tensorflow/compression

Data compression in TensorFlow

TensorFlow Compression

TensorFlow Compression (TFC) contains data compression tools for TensorFlow.

You can use this library to build your own ML models with end-to-end optimized data compression built in. It's useful to find storage-efficient representations of your data (images, features, examples, etc.) while only sacrificing a small fraction of model performance. Take a look at the lossy data compression tutorial or the model compression tutorial to get started.

For a more in-depth introduction from a classical data compression perspective, consider our paper on nonlinear transform coding, or watch @jonycgn's talk on learned image compression. For an introduction to lossy data compression from a machine learning perspective, take a look at @yiboyang's review paper.

The library contains (see the API docs for details):

• Range coding (a.k.a. arithmetic coding) implementations in the form of flexible TF ops written in C++. These include an optional "overflow" functionality that embeds an Elias gamma code into the range encoded bit sequence, making it possible to encode alphabets containing the entire set of signed integers rather than just a finite range.

• Entropy model classes which simplify the process of designing rate–distortion optimized codes. During training, they act like likelihood models. Once training is completed, they encode floating point tensors into optimized bit sequences by automating the design of range coding tables and calling the range coder implementation behind the scenes.

• Additional TensorFlow functions and Keras layers that are useful in the context of learned data compression, such as methods to numerically find quantiles of density functions, take expectations with respect to dithering noise, convolution layers with more flexible padding options and support for reparameterizing kernels and biases in the Fourier domain, and an implementation of generalized divisive normalization (GDN).

Important update: As of February 1, 2024, TensorFlow Compression is in maintenance mode. This means concretely:

• The full feature set of TFC is frozen. No new features will be developed, but the repository will receive maintenance fixes.

• Going forward, new TFC packages will only work with TensorFlow 2.14. This is due to an incompatibility introduced in the Keras version shipped with TF 2.15, which would require a rewrite of our layer and entropy model classes.

• To ensure existing models can still be run with TF 2.15 and later, we are releasing a new package tensorflow-compression-ops, which only contains the C++ ops. These will be updated as long as possible for newer TF versions.

• Binary packages are provided for both options on pypi.org: TFC and TFC ops.

Documentation & getting help

Refer to the API documentation for a complete description of the classes and functions this package implements.

Please post all questions or comments on Discussions. Only file Issues for actual bugs or feature requests. On Discussions, you may get a faster answer, and you help other people find the question or answer more easily later.

Installation

Note: Precompiled packages are currently only provided for Linux and Darwin/Mac OS. To use these packages on Windows, consider installing TensorFlow using the instructions for WSL2 or using a TensorFlow Docker image, and then installing the Linux package.

Set up an environment in which you can install precompiled binary Python packages using the pip command. Refer to the TensorFlow installation instructions for more information on how to set up such a Python environment.

The current version of TensorFlow Compression requires TensorFlow 2. For versions compatible with TensorFlow 1, see our previous releases.

pip

To install TFC via pip, run the following command:

python -m pip install tensorflow-compression

To test that the installation works correctly, you can run the unit tests with:

python -m tensorflow_compression.all_tests

Once the command finishes, you should see a message OK (skipped=29) or similar in the last line.

Colab

You can try out TFC live in a Colab. The following command installs the latest version of TFC that is compatible with the installed TensorFlow version. Run it in a cell before executing your Python code:

%pip install tensorflow-compression~=$(pip show tensorflow | perl -p -0777 -e 's/.*Version: (\d+\.\d+).*/\1.0/sg')

Note: The binary packages of TFC are tied to TF with the same minor version (e.g., TFC 2.9.1 requires TF 2.9.x), and Colab sometimes lags behind a few days in deploying the latest version of TensorFlow. As a result, using pip install tensorflow-compression naively might attempt to upgrade TF, which can create problems.

Docker

To use a Docker container (e.g. on Windows), be sure to install Docker (e.g., Docker Desktop), use a TensorFlow Docker image, and then run the pip install command inside the Docker container, not on the host. For instance, you can use a command line like this:

docker run tensorflow/tensorflow:latest bash -c \
    "python -m pip install tensorflow-compression &&
     python -m tensorflow_compression.all_tests"

This will fetch the TensorFlow Docker image if it's not already cached, install the pip package and then run the unit tests to confirm that it works.

Anaconda

It seems that Anaconda ships its own binary version of TensorFlow which is incompatible with our pip package. To solve this, always install TensorFlow via pip rather than conda. For example, this creates an Anaconda environment with CUDA libraries, and then installs TensorFlow and TensorFlow Compression:

conda create --name ENV_NAME python cudatoolkit cudnn
conda activate ENV_NAME
python -m pip install tensorflow-compression

Depending on the requirements of the tensorflow pip package, you may need to pin the CUDA libraries to specific versions. If you aren't using a GPU, CUDA is of course not necessary.

Usage

We recommend importing the library from your Python code as follows:

import tensorflow as tf
import tensorflow_compression as tfc

Using a pre-trained model to compress an image

In the models directory, you'll find a python script tfci.py. Download the file and run:

python tfci.py -h

This will give you a list of options. Briefly, the command

python tfci.py compress <model> <PNG file>

will compress an image using a pre-trained model and write a file ending in .tfci. Execute python tfci.py models to give you a list of supported pre-trained models. The command

python tfci.py decompress <TFCI file>

will decompress a TFCI file and write a PNG file. By default, an output file will be named like the input file, only with the appropriate file extension appended (any existing extensions will not be removed).

Training your own model

The models directory contains several implementations of published image compression models to enable easy experimentation. Note that in order to reproduce published results, more tuning of the code and training dataset may be necessary. Use the tfci.py script above to access published models.

The following instructions talk about a re-implementation of the model published in:

"End-to-end optimized image compression"
J. Ballé, V. Laparra, E. P. Simoncelli
https://arxiv.org/abs/1611.01704

Note that the models directory is not contained in the pip package. The models are meant to be downloaded individually. Download the file bls2017.py and run:

python bls2017.py -h

This will list the available command line options for the implementation. Training can be as simple as the following command:

python bls2017.py -V train

This will use the default settings. Note that unless a custom training dataset is provided via --train_glob, the CLIC dataset will be downloaded using TensorFlow Datasets.

The most important training parameter is --lambda, which controls the trade-off between bitrate and distortion that the model will be optimized for. The number of channels per layer is important, too: models tuned for higher bitrates (or, equivalently, lower distortion) tend to require transforms with a greater approximation capacity (i.e. more channels), so to optimize performance, you want to make sure that the number of channels is large enough (or larger). This is described in more detail in:

"Efficient nonlinear transforms for lossy image compression"
J. Ballé
https://arxiv.org/abs/1802.00847

If you wish, you can monitor progress with Tensorboard. To do this, create a Tensorboard instance in the background before starting the training, then point your web browser to port 6006 on your machine:

tensorboard --logdir=/tmp/train_bls2017 &

When training has finished, the Python script saves the trained model to the directory specified with --model_path (by default, bls2017 in the current directory) in TensorFlow's SavedModel format. The script can then be used to compress and decompress images as follows. The same saved model must be accessible to both commands.

python bls2017.py [options] compress original.png compressed.tfci
python bls2017.py [options] decompress compressed.tfci reconstruction.png

Building pip packages

This section describes the necessary steps to build your own pip packages of TensorFlow Compression. This may be necessary to install it on platforms for which we don't provide precompiled binaries (currently only Linux and Darwin).

To be compatible with the official TensorFlow pip package, the TFC pip package must be linked against a matching version of the C libraries. For this reason, to build the official Linux pip packages, we use these Docker images and use the same toolchain that TensorFlow uses.

Inside the Docker container, the following steps need to be taken:

1. Clone the tensorflow/compression repo from GitHub.
2. Run tools/build_pip_pkg.sh inside the cloned repo.

For example:

git clone https://github.com/tensorflow/compression.git /tensorflow_compression
docker run -i --rm \
    -v /tmp/tensorflow_compression:/tmp/tensorflow_compression\
    -v /tensorflow_compression:/tensorflow_compression \
    -w /tensorflow_compression \
    -e "BAZEL_OPT=--config=manylinux_2_17_x86_64" \
    tensorflow/build:latest-python3.10 \
    bash tools/build_pip_pkg.sh /tmp/tensorflow_compression <custom-version>

For Darwin, the Docker image and specifying the toolchain is not necessary. We just build the package like this (note that you may want to create a clean Python virtual environment to do this):

git clone https://github.com/tensorflow/compression.git /tensorflow_compression
cd /tensorflow_compression
BAZEL_OPT="--macos_minimum_os=10.14" bash \
  tools/build_pip_pkg.sh \
  /tmp/tensorflow_compression <custom-version>

In both cases, the wheel file is created inside /tmp/tensorflow_compression.

To test the created package, first install the resulting wheel file:

python -m pip install /tmp/tensorflow_compression/tensorflow_compression-*.whl

Then run the unit tests (Do not run the tests in the workspace directory where the WORKSPACE file lives. In that case, the Python interpreter would attempt to import tensorflow_compression packages from the source tree, rather than from the installed package system directory):

pushd /tmp
python -m tensorflow_compression.all_tests
popd

When done, you can uninstall the pip package again:

python -m pip uninstall tensorflow-compression

Evaluation

We provide evaluation results for several image compression methods in terms of different metrics in different colorspaces. Please see the results subdirectory for more information.

Citation

If you use this library for research purposes, please cite:

@software{tfc_github,
  author = "Ballé, Johannes and Hwang, Sung Jin and Agustsson, Eirikur",
  title = "{T}ensor{F}low {C}ompression: Learned Data Compression",
  url = "http://github.com/tensorflow/compression",
  version = "2.14.1",
  year = "2024",
}

In the above BibTeX entry, names are top contributors sorted by number of commits. Please adjust version number and year according to the version that was actually used.

Note that this is not an officially supported Google product.

compression's Issues

FAILED (errors=79, skipped=11) python -m tensorflow_compression.python.all_test

Hi, I tried to test the environment using python -m tensorflow_compression.python.all_test, but

Running tests under Python 3.6.9: /home/xuxinzi/anaconda3/envs/compress/bin/python
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_build_decompress
W0814 13:21:54.124725 140542371317504 deprecation_wrapper.py:119] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py:225: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

W0814 13:21:54.299077 140542371317504 deprecation_wrapper.py:119] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models.py:630: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.

W0814 13:21:54.308104 140542371317504 deprecation_wrapper.py:119] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models.py:648: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.

[ OK ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_build_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_channels_first
W0814 13:21:55.052779 140542371317504 deprecation_wrapper.py:119] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/autograph/converters/directives.py:117: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

W0814 13:21:55.285466 140542371317504 deprecation.py:323] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py:60: TensorFlowTestCase.test_session (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use self.session() or self.cached_session() instead.
2019-08-14 13:21:55.285949: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 AVX512F FMA
2019-08-14 13:21:55.295784: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcuda.so.1
2019-08-14 13:21:55.734689: W tensorflow/compiler/xla/service/platform_util.cc:256] unable to create StreamExecutor for CUDA:0: failed initializing StreamExecutor for CUDA device ordinal 0: Internal: failed call to cuDevicePrimaryCtxRetain: CUDA_ERROR_OUT_OF_MEMORY: out of memory; total memory reported: 8338604032
2019-08-14 13:21:55.735069: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x55ed92a30ed0 executing computations on platform CUDA. Devices:
2019-08-14 13:21:55.735115: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (0): GeForce RTX 2080, Compute Capability 7.5
2019-08-14 13:21:55.735124: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (1): GeForce GTX 1070 Ti, Compute Capability 6.1
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_channels_first
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_channels_last
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_channels_last
[ RUN] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_codec
W0814 13:21:56.746465 140542371317504 deprecation_wrapper.py:119] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py:98: The name tf.train.GradientDescentOptimizer is deprecated. Please use tf.compat.v1.train.GradientDescentOptimizer instead.

[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_codec
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_codec_init
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_codec_init
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_compress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_compress
[ RUN] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_decompress
[ FAILED] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_decompress
[ RUN] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_entropy_estimates
W0814 13:21:58.176367 140542371317504 deprecation_wrapper.py:119] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py:250: The name tf.log is deprecated. Please use tf.math.log instead.

[ FAILED] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_entropy_estimates
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_noise
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_noise
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_normalization
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_normalization
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_quantization
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_quantization
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_quantization_init
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_quantization_init

[ RUN ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest.test_session
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_channels_first
[ FAILED ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_channels_first
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_channels_last
[ FAILED ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_channels_last
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_igdn
[ FAILED ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_igdn
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_invalid_data_format
[ OK ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_invalid_data_format
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_rgdn
[ FAILED ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_rgdn
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_session
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_unknown_dim
[ OK ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_unknown_dim
[ RUN ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_wrong_dims
[ OK ] tensorflow_compression.python.layers.gdn_test.GDNTest.test_wrong_dims
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_build_decompress
[ OK ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_build_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_codec
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_codec
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_codec_mean
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_codec_mean
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_compress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_compress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_decompress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_distribution
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_distribution
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_entropy_estimates
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_entropy_estimates
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_multiple_dimensions
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_multiple_dimensions
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_noise
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_noise
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_quantile_function
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_quantile_function
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_quantization
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_quantization
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_quantization_mean
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_quantization_mean
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest.test_session
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_build_decompress
W0814 13:22:02.914414 140542371317504 deprecation.py:323] From /home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models.py:1082: add_dispatch_support..wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.where in 2.0, which has the same broadcast rule as np.where
[ OK ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_build_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_codec
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_codec
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_codec_mean
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_codec_mean
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_compress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_compress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_decompress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_distribution
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_distribution
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_entropy_estimates
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_entropy_estimates
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_multiple_dimensions
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_multiple_dimensions
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_noise
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_noise
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_quantile_function
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_quantile_function
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_quantization
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_quantization
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_quantization_mean
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_quantization_mean
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest.test_session
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_build_decompress
[ OK ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_build_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_codec
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_codec
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_codec_mean
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_codec_mean
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_compress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_compress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_decompress
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_decompress
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_distribution
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_distribution
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_entropy_estimates
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_entropy_estimates
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_multiple_dimensions
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_multiple_dimensions
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_noise
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_noise
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_quantile_function
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_quantile_function
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_quantization
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_quantization
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_quantization_mean
[ FAILED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_quantization_mean
[ RUN ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest.test_session
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_disconnected
[ FAILED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_disconnected
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_identity
[ FAILED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_identity
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_identity_if_towards
[ FAILED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_identity_if_towards
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_invalid
[ OK ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_lower_bound_invalid
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_session
[ SKIPPED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_session
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_disconnected
[ FAILED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_disconnected
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_identity
[ FAILED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_identity
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_identity_if_towards
[ FAILED ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_identity_if_towards
[ RUN ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_invalid
[ OK ] tensorflow_compression.python.ops.math_ops_test.MathTest.test_upper_bound_invalid
[ RUN ] tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest.test_same_padding_conv
[ FAILED ] tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest.test_same_padding_conv
[ RUN ] tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest.test_same_padding_corr
[ FAILED ] tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest.test_same_padding_corr
[ RUN ] tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest.test_session
[ SKIPPED ] tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest.test_session
[ RUN ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_nonnegative_parameterizer
[ FAILED ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_nonnegative_parameterizer
[ RUN ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_positive_parameterizer
[ FAILED ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_positive_parameterizer
[ RUN ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_rdft_parameterizer
[ FAILED ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_rdft_parameterizer
[ RUN ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_session
[ RUN ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_static_parameterizer
[ FAILED ] tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest.test_static_parameterizer
[ RUN ] tensorflow_compression.python.ops.range_coding_ops_test.RangeCodingOpsTest.test_readme_example
[ FAILED ] tensorflow_compression.python.ops.range_coding_ops_test.RangeCodingOpsTest.test_readme_example
[ RUN ] tensorflow_compression.python.ops.range_coding_ops_test.RangeCodingOpsTest.test_session
[ SKIPPED ] tensorflow_compression.python.ops.range_coding_ops_test.RangeCodingOpsTest.test_session
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_bias_activation
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_bias_activation
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_same_padding
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_same_padding
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_same_zeros_spatial
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_same_zeros_spatial
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_valid_channels
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_valid_channels
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_valid_spatial
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_1d_valid_spatial
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_bias_activation
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_bias_activation
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_same_padding
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_same_padding
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_same_zeros_spatial
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_same_zeros_spatial
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_valid_channels
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_valid_channels
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_valid_spatial
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_2d_valid_spatial
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_bias_activation
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_bias_activation
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_same_padding
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_same_padding
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_same_zeros_spatial
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_same_zeros_spatial
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_valid_channels
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_valid_channels
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_valid_spatial
[ FAILED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_3d_valid_spatial
[ RUN ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_session
[ SKIPPED ] tensorflow_compression.python.layers.signal_conv_test.SignalTest.test_session
[ RUN ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_irdft1_matrix
[ FAILED ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_irdft1_matrix
[ RUN ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_irdft2_matrix
[ FAILED ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_irdft2_matrix
[ RUN ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_irdft3_matrix
[ FAILED ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_irdft3_matrix
[ RUN ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_session
[ SKIPPED ] tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest.test_session

======================================================================

ERROR: test_channels_first (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_channels_first (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 144, in test_channels_first
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_channels_last (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_channels_last (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 123, in test_channels_last
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_codec (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 101, in test_codec
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec_init (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_codec_init (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 82, in test_codec_init
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_compress (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_compress (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 164, in test_compress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_decompress (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_decompress (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 216, in test_decompress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 254, in test_entropy_estimates
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_noise (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_noise (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 35, in test_noise
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_normalization (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_normalization (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 236, in test_normalization
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_quantization (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 64, in test_quantization
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization_init (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)
test_quantization_init (tensorflow_compression.python.layers.entropy_models_test.EntropyBottleneckTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 48, in test_quantization_init
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_channels_first (tensorflow_compression.python.layers.gdn_test.GDNTest)
test_channels_first (tensorflow_compression.python.layers.gdn_test.GDNTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 59, in test_channels_first
y = self._run_gdn(x, x.shape, False, False, "channels_first")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 34, in _run_gdn
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_channels_last (tensorflow_compression.python.layers.gdn_test.GDNTest)
test_channels_last (tensorflow_compression.python.layers.gdn_test.GDNTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 52, in test_channels_last
y = self._run_gdn(x, x.shape, False, False, "channels_last")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 34, in _run_gdn
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_igdn (tensorflow_compression.python.layers.gdn_test.GDNTest)
test_igdn (tensorflow_compression.python.layers.gdn_test.GDNTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 73, in test_igdn
y = self._run_gdn(x, x.shape, True, False, "channels_last")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 34, in _run_gdn
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_rgdn (tensorflow_compression.python.layers.gdn_test.GDNTest)
test_rgdn (tensorflow_compression.python.layers.gdn_test.GDNTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 79, in test_rgdn
y = self._run_gdn(x, x.shape, False, True, "channels_last")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/gdn_test.py", line 34, in _run_gdn
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_codec (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 330, in test_codec
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec_mean (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_codec_mean (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 349, in test_codec_mean
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_compress (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_compress (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 394, in test_compress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_decompress (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_decompress (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 418, in test_decompress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_distribution (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_distribution (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 455, in test_distribution
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 479, in test_entropy_estimates
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_multiple_dimensions (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_multiple_dimensions (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 372, in test_multiple_dimensions
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_noise (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_noise (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 275, in test_noise
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantile_function (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_quantile_function (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 440, in test_quantile_function
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_quantization (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 291, in test_quantization
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization_mean (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)
test_quantization_mean (tensorflow_compression.python.layers.entropy_models_test.GaussianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 308, in test_quantization_mean
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_codec (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 330, in test_codec
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec_mean (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_codec_mean (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 349, in test_codec_mean
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_compress (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_compress (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 394, in test_compress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_decompress (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_decompress (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 418, in test_decompress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_distribution (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_distribution (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 455, in test_distribution
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 479, in test_entropy_estimates
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_multiple_dimensions (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_multiple_dimensions (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 372, in test_multiple_dimensions
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_noise (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_noise (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 275, in test_noise
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantile_function (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_quantile_function (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 440, in test_quantile_function
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_quantization (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 291, in test_quantization
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization_mean (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)
test_quantization_mean (tensorflow_compression.python.layers.entropy_models_test.LaplacianConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 308, in test_quantization_mean
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_codec (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 330, in test_codec
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_codec_mean (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_codec_mean (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 349, in test_codec_mean
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_compress (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_compress (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 394, in test_compress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_decompress (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_decompress (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 418, in test_decompress
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_distribution (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_distribution (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 455, in test_distribution
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_entropy_estimates (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 479, in test_entropy_estimates
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_multiple_dimensions (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_multiple_dimensions (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 372, in test_multiple_dimensions
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_noise (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_noise (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 275, in test_noise
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantile_function (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_quantile_function (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 440, in test_quantile_function
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_quantization (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 291, in test_quantization
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_quantization_mean (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)
test_quantization_mean (tensorflow_compression.python.layers.entropy_models_test.LogisticConditionalTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/entropy_models_test.py", line 308, in test_quantization_mean
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_lower_bound_disconnected (tensorflow_compression.python.ops.math_ops_test.MathTest)
test_lower_bound_disconnected (tensorflow_compression.python.ops.math_ops_test.MathTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 92, in test_lower_bound_disconnected
self._test_lower_bound("disconnected")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 84, in _test_lower_bound
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_lower_bound_identity (tensorflow_compression.python.ops.math_ops_test.MathTest)
test_lower_bound_identity (tensorflow_compression.python.ops.math_ops_test.MathTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 95, in test_lower_bound_identity
self._test_lower_bound("identity")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 84, in _test_lower_bound
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_lower_bound_identity_if_towards (tensorflow_compression.python.ops.math_ops_test.MathTest)
test_lower_bound_identity_if_towards (tensorflow_compression.python.ops.math_ops_test.MathTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 98, in test_lower_bound_identity_if_towards
self._test_lower_bound("identity_if_towards")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 84, in _test_lower_bound
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_upper_bound_disconnected (tensorflow_compression.python.ops.math_ops_test.MathTest)
test_upper_bound_disconnected (tensorflow_compression.python.ops.math_ops_test.MathTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 54, in test_upper_bound_disconnected
self._test_upper_bound("disconnected")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 46, in _test_upper_bound
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_upper_bound_identity (tensorflow_compression.python.ops.math_ops_test.MathTest)
test_upper_bound_identity (tensorflow_compression.python.ops.math_ops_test.MathTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 57, in test_upper_bound_identity
self._test_upper_bound("identity")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 46, in _test_upper_bound
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_upper_bound_identity_if_towards (tensorflow_compression.python.ops.math_ops_test.MathTest)
test_upper_bound_identity_if_towards (tensorflow_compression.python.ops.math_ops_test.MathTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 60, in test_upper_bound_identity_if_towards
self._test_upper_bound("identity_if_towards")
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/math_ops_test.py", line 46, in _test_upper_bound
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_same_padding_conv (tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest)
test_same_padding_conv (tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/padding_ops_test.py", line 60, in test_same_padding_conv
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_same_padding_corr (tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest)
test_same_padding_corr (tensorflow_compression.python.ops.padding_ops_test.PaddingOpsTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/padding_ops_test.py", line 40, in test_same_padding_corr
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_nonnegative_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)
test_nonnegative_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 58, in test_nonnegative_parameterizer
tf.initializers.random_uniform(), shape)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 33, in _test_parameterizer
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_positive_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)
test_positive_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 66, in test_positive_parameterizer
tf.initializers.random_uniform(), shape)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 33, in _test_parameterizer
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_rdft_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)
test_rdft_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 50, in test_rdft_parameterizer
tf.initializers.ones(), shape)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 33, in _test_parameterizer
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_static_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)
test_static_parameterizer (tensorflow_compression.python.layers.parameterizers_test.ParameterizersTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 42, in test_static_parameterizer
tf.initializers.random_uniform(), shape)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/parameterizers_test.py", line 33, in _test_parameterizer
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_readme_example (tensorflow_compression.python.ops.range_coding_ops_test.RangeCodingOpsTest)
test_readme_example (tensorflow_compression.python.ops.range_coding_ops_test.RangeCodingOpsTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/range_coding_ops_test.py", line 45, in test_readme_example
with self.cached_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1965, in cached_session
graph, config, force_gpu, crash_if_inconsistent_args=True)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_1d_bias_activation (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_1d_bias_activation (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 1D convolutions with bias and activation.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 368, in test_1d_bias_activation
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_1d_same_padding (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_1d_same_padding (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 1D same convolutions with different padding modes.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 345, in test_1d_same_padding
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_1d_same_zeros_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_1d_same_zeros_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 1D same_zeros convolutions with different supports/strides.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 322, in test_1d_same_zeros_spatial
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_1d_valid_channels (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_1d_valid_channels (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 1D valid convolutions with multiple channels/filters.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 298, in test_1d_valid_channels
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_1d_valid_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_1d_valid_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 1D valid convolutions with different supports/strides.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 276, in test_1d_valid_spatial
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_2d_bias_activation (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_2d_bias_activation (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 2D convolutions with bias and activation.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 484, in test_2d_bias_activation
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_2d_same_padding (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_2d_same_padding (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 2D same convolutions with different padding modes.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 461, in test_2d_same_padding
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_2d_same_zeros_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_2d_same_zeros_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 2D same_zeros convolutions with different supports/strides.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 438, in test_2d_same_zeros_spatial
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_2d_valid_channels (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_2d_valid_channels (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 2D valid convolutions with multiple channels/filters.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 414, in test_2d_valid_channels
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_2d_valid_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_2d_valid_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 2D valid convolutions with different supports/strides.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 392, in test_2d_valid_spatial
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_3d_bias_activation (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_3d_bias_activation (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 3D convolutions with bias and activation.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 600, in test_3d_bias_activation
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_3d_same_padding (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_3d_same_padding (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 3D same convolutions with different padding modes.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 577, in test_3d_same_padding
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_3d_same_zeros_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_3d_same_zeros_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 3D same_zeros convolutions with different supports/strides.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 554, in test_3d_same_zeros_spatial
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_3d_valid_channels (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_3d_valid_channels (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 3D valid convolutions with multiple channels/filters.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 530, in test_3d_valid_channels
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_3d_valid_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
test_3d_valid_spatial (tensorflow_compression.python.layers.signal_conv_test.SignalTest)
Test 3D valid convolutions with different supports/strides.

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 508, in test_3d_valid_spatial
for data_format in self.data_formats:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/layers/signal_conv_test.py", line 225, in data_formats
if tf.test.is_gpu_available(cuda_only=True):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1368, in is_gpu_available
for local_device in device_lib.list_local_devices():
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/device_lib.py", line 41, in list_local_devices
for s in pywrap_tensorflow.list_devices(session_config=session_config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 2202, in list_devices
return ListDevices()
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_irdft1_matrix (tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest)
test_irdft1_matrix (tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/spectral_ops_test.py", line 35, in test_irdft1_matrix
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_irdft2_matrix (tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest)
test_irdft2_matrix (tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/spectral_ops_test.py", line 45, in test_irdft2_matrix
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

======================================================================
ERROR: test_irdft3_matrix (tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest)
test_irdft3_matrix (tensorflow_compression.python.ops.spectral_ops_test.SpectralOpsTest)

Traceback (most recent call last):
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow_compression/python/ops/spectral_ops_test.py", line 55, in test_irdft3_matrix
with self.test_session() as sess:
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/contextlib.py", line 81, in enter
return next(self.gen)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 1986, in test_session
graph, config, force_gpu, crash_if_inconsistent_args=False)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2752, in _get_cached_session
graph=graph, config=config, force_gpu=force_gpu)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/framework/test_util.py", line 2742, in _create_session
return ErrorLoggingSession(graph=graph, config=prepare_config(config))
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1570, in init
super(Session, self).init(target, graph, config=config)
File "/home/xuxinzi/anaconda3/envs/compress/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 693, in init
self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InvalidArgumentError: device CUDA:0 not supported by XLA service
while setting up XLA_GPU_JIT device number 0

---

Ran 99 tests in 13.451s

FAILED (errors=79, skipped=11)

I have no idea why these errors occured although I can successfully import tensorflow and tensorflow_compression. My settings are listed below:

absl-py 0.7.1
astor 0.8.0
certifi 2019.6.16
gast 0.2.2
google-pasta 0.1.7
grpcio 1.22.0
h5py 2.9.0
Keras-Applications 1.0.8
Keras-Preprocessing 1.1.0
Markdown 3.1.1
numpy 1.17.0
pip 19.1.1
protobuf 3.9.1
scipy 1.3.1
setuptools 41.0.1
six 1.12.0
tensorboard 1.14.0
tensorflow-compression 1.2b2
tensorflow-estimator 1.14.0
tensorflow-gpu 1.14.0
termcolor 1.1.0
Werkzeug 0.15.5
wheel 0.33.4
wrapt 1.11.2

Reimplementation on learned image compression with Pytorch toolbox

Thanks for your wonderful works on learned image compression.
Our team has reimplemented the ' End-to-end Optimized Image Compression(ICLR17)' and 'Variational image compression with a scale hyperprior(ICLR18)' with Pytorch toolbox.
Besides, our team has released our code on Github.
ICLR17 link: https://github.com/liujiaheng/iclr_17_compression
ICLR18 link: https://github.com/liujiaheng/compression

Thanks.

When the model running in aGPU，we will get different result each time

When the trained model compress or decompress a image in a GPU, we got different result each time. But when compress or decompress image in cpu, we will get correct result (only one)each time

EntropyBottleneck: make training flag compatible with TensorFlow Booleans.

compression/tensorflow_compression/python/layers/entropy_models.py

Lines 416 to 423 in d156bbe

	# Add noise or quantize.
	if training:
	noise = random_ops.random_uniform(array_ops.shape(values), -half, half)
	values = math_ops.add_n([values, noise])
	elif self.optimize_integer_offset:
	values = math_ops.round(values - self._medians) + self._medians
	else:
	values = math_ops.round(values)

The EntropyBottleneck layer currently only takes Python Booleans for its training flag. A more comfortable option would be to also accept TensorFlow Booleans so that the same layer can be used for evaluation of test-performance during training (similar to the training flag in Batch Normalization layers).

As far as I can see, adding in a tf.cond here is unproblematic and shouldn't hurt performance, but I may be missing something. I can submit a PR if that helps.

no file of compare_gan

Dear all:
Thanks for releasing such great work.
I cannot find the file named compare_gan. Did you miss this part?
Looking forward to your reply.
Best regards.

Jillian

pip install fails for Python 3.7

I'm using conda and had Python 3.7 + TensorFlow 1.13, and doing pip install tensorflow-compression just gave me

Could not find a version that satisfies the requirement tensorflow-compression

Downgrading to Python 3.6 worked.

Might be useful to mention in the readme, but might also be a quite rare to use conda for TF.

Problem when training bls2017.py

Describe the bug
When I execute this command python3 bls2017.py --verbose train --train_glob="images/*.png", the training is first appears to work well but after that, this error occures:

After that, the file "train" appears but when decompressing compressed images with bls2017.py, the result image is all black.

Expected behavior
I think the reconstruction problem is the result of the training error so i expect to have a solution for the training problem.

System please complete the following information):

• ubuntu 18.04
• python 3.6.9
• TensorFlow 1.15.0
• tensorflow-compression 1.3

Errors when running test scripts

Hello,

The script python compression/python/ops/coder_ops_test.py works perfectly. However I get errors with the second script python compression/python/layers/entropybottleneck_test.py.

Following is the trace.

Any idea what's going on?

I'm using Ubuntu 16.04 with Tensorflow 1.8 GPU.

`EEEEEEEEEEEEEE.E

ERROR: test_build_decompress (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 257, in test_build_decompress
layer.decompress(bitstrings, input_shape[1:], channels=5)
File " /compression/python/layers/entropybottleneck.py", line 533, in decompress
with ops.name_scope(self._name_scope()):
AttributeError: 'EntropyBottleneck' object has no attribute '_name_scope'

======================================================================
ERROR: test_channels_first (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 175, in test_channels_first
noisy, _ = layer(inputs, training=True)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_channels_last (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 154, in test_channels_last
noisy, _ = layer(inputs, training=True)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_codec (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 91, in test_codec
bitstrings = layer.compress(inputs)
File " /compression/python/layers/entropybottleneck.py", line 464, in compress
with ops.name_scope(self._name_scope()):
AttributeError: 'EntropyBottleneck' object has no attribute '_name_scope'

======================================================================
ERROR: test_codec_clipping (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 137, in test_codec_clipping
bitstrings = layer.compress(inputs)
File " /compression/python/layers/entropybottleneck.py", line 464, in compress
with ops.name_scope(self._name_scope()):
AttributeError: 'EntropyBottleneck' object has no attribute '_name_scope'

======================================================================
ERROR: test_codec_optimized_offset (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 114, in test_codec_optimized_offset
bitstrings = layer.compress(inputs)
File " /compression/python/layers/entropybottleneck.py", line 464, in compress
with ops.name_scope(self._name_scope()):
AttributeError: 'EntropyBottleneck' object has no attribute '_name_scope'

======================================================================
ERROR: test_compress (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 197, in test_compress
bitstrings = layer.compress(inputs)
File " /compression/python/layers/entropybottleneck.py", line 464, in compress
with ops.name_scope(self._name_scope()):
AttributeError: 'EntropyBottleneck' object has no attribute '_name_scope'

======================================================================
ERROR: test_decompress (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 242, in test_decompress
layer.build(self.expected.shape)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_entropy_estimates (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 296, in test_entropy_estimates
_, likelihood = layer(inputs, training=True)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_noise (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 40, in test_noise
noisy, _ = layer(inputs, training=True)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_normalization (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 282, in test_normalization
_, likelihood = layer(inputs, training=True)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_pmf_normalization (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 263, in test_pmf_normalization
layer.build((None, 10))
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_quantization (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 53, in test_quantization
quantized, _ = layer(inputs, training=False)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_quantization_optimized_offset (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 70, in test_quantization_optimized_offset
quantized, _ = layer(inputs, training=False)
File "/site-packages/tensorflow/python/keras/_impl/keras/engine/base_layer.py", line 314, in call
output = super(Layer, self).call(inputs, *args, **kwargs)
File "/site-packages/tensorflow/python/layers/base.py", line 699, in call
self.build(input_shapes)
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

======================================================================
ERROR: test_visualize (main.EntropyBottleneckTest)

Traceback (most recent call last):
File "compression/python/layers/entropybottleneck_test.py", line 272, in test_visualize
layer.build((None, 10))
File " /compression/python/layers/entropybottleneck.py", line 368, in build
getter=cdf_getter, trainable=False)
TypeError: add_variable() got an unexpected keyword argument 'getter'

---

Ran 16 tests in 0.900s

FAILED (errors=15)
`

the problem of print

hello,

I ran the example bls2017 to compress images. The file "compressed.bin" could be generated, but there was nothing to print (MSE, PSNR etc.). Could you help me solve this problem?

Thank you very much!

Difference with hyperprior paper: hyperprior (decoder) loses a ReLU

The model defined in compression/models/bmshj2018.py differs from the one shown in the paper (whose figure is shown below):

The paper has 3 ReLUs in the hyper-decoder, but the HyperSynthesisTransform class omits the last ReLU.

Are the results shown in the paper based on the paper's definition, or the published code?

I am trying to replicate Balle2018 to compare the results of my work, for that I used the pytorch implementation defined in https://github.com/liujiaheng/compression and I was getting good results, but I noticed several differences from the paper's definition. One is the lack of a ReLU which is present in this code as well. liujiaheng also replaces the gaussian distribution by a laplace distribution and the hyper-encoder absolute value with an exponential function (that also doesn't seem to be mentioned in a paper). Once I revert those modifications I no longer seem to get as good results (although training is still far from complete).

edit: liujiaheng's exponential function is in the decoder, I put it back because it seems to be mentioned in the paper's equations and possibly in your code as well. I edited the post accordingly with strikethrough, though that means I don't know yet the performance of each configuration.

KeyError: u'UnboundedIndexRangeEncode'

Hi,
I try to compress an image by running the command at 'example' folder: "python tfci.py compress mbt2018-mean-mse-1 test.png". But it failed. I did not modify anything to the original codes. Do you have any idea about it?

Environment:
python 2.7
tensorflow1.13-gpu

Error Message:
Traceback (most recent call last):
File "tfci.py", line 326, in
app.run(main, flags_parser=parse_args)
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/absl/app.py", line 300, in run
_run_main(main, args)
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/absl/app.py", line 251, in _run_main
sys.exit(main(argv))
File "tfci.py", line 318, in main
args.target_bpp, args.bpp_strict)
File "tfci.py", line 154, in compress
bitstring = compress_image(model, input_image)
File "tfci.py", line 110, in compress_image
signature_defs = import_metagraph(model)
File "tfci.py", line 88, in import_metagraph
tf.train.import_meta_graph(metagraph)
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1435, in import_meta_graph
meta_graph_or_file, clear_devices, import_scope, **kwargs)[0]
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/tensorflow/python/training/saver.py", line 1457, in _import_meta_graph_with_return_elements
**kwargs))
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/tensorflow/python/framework/meta_graph.py", line 806, in import_scoped_meta_graph_with_return_elements
return_elements=return_elements)
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
return func(*args, **kwargs)
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/tensorflow/python/framework/importer.py", line 399, in import_graph_def
_RemoveDefaultAttrs(op_dict, producer_op_list, graph_def)
File "/home/gcs/App/anaconda3/envs/tensorflow1.13/lib/python2.7/site-packages/tensorflow/python/framework/importer.py", line 159, in _RemoveDefaultAttrs
op_def = op_dict[node.op]
KeyError: u'UnboundedIndexRangeEncode'

Problem when executing eagerly

Hello!

I'm trying to use some of your layers, but I see they're not fully compatible with eager execution (even though I see some if's to treat cases of eager execution in your code).

One example is the self.add_loss in the entropy bottleneck layer.

I've started reading the keras code, and I've seen this commentary:

Note that add_loss is not supported when executing eagerly. Instead,
variable regularizers may be added through add_variable. Activity
regularization is not supported directly (but such losses may be returned
from Layer.call()).

I don't know if simply using the add_variable instead would fix this problem.

I have not tried any solution yet. I've just wanted to share the problem with the code maintainer.

Maybe anybody seeing this can point out a solution before I do.

; )

range_coding_ops.cc: UnboundedIndexRangeEncode Documentation on overflow flag

Hey there!

In line 171 of range_coding_ops.cc, it says

When a value from data is in the interval [offset[i], offset[i] + m - 2), then the value is range encoded using the CDF values. The last entry in each CDF (the one at m - 1) is an overflow code.

However, you usually refer to m, the size of the CDF, as the number of interval boundaries CDF has, not the number of intervals (unless I'm very much mistaken here). So the CDF has m-1, intervals, the last one of which has the index m-2, which then closes the interval [offset[i], offset[i] + m - 2) so that the CDF is complete.

From the implementation in https://github.com/tensorflow/compression/blob/master/tensorflow_compression/cc/kernels/unbounded_index_range_coding_kernels.cc#L209:

const int32 max_value = cdf_size(cdf_index) - 2;

So, as far as I can tell, the implementation is m-2 for the overflow flag and I think it would be clearer to state it in the documentation as well.

This is a small issue, but it can be confusing if someone is trying to understand or implement your coding protocol ;)

Problem when I tried to create a Gaussian mixture distribution

When I tried to add a Gaussian mixture distribution in entropy_models.py, I had some problems.

1. The self._indexes in the build function of the SymmetricConditional class:
   def loop_body(indexes, scale):
   return indexes - tf.cast(self.scale <= scale, tf.int32)
   self._indexes = tf.foldr(
   loop_body, scale_table[:-1],
   initializer=initializer, back_prop=False, name="compute_indexes")
   why the self._indexes need to subtract the scale?
2. The pmf in the build function of the SymmetricConditional class:
   upper = self._standardized_cumulative((.5 - samples) / samples_scale)
   lower = self._standardized_cumulative((-.5 - samples) / samples_scale)
   pmf = upper - lower
   but there is a difference in calculating the pmf in the https://github.com/ZhengxueCheng/Learned-Image-Compression-with-GMM-and-Attention
   pmf = (0.5 * (1 + scipy.special.erf((samples + 0.5 - mu[0]) / ((sigma[0] + TINY) * 2 ** 0.5))) -
   0.5 * (1 + scipy.special.erf((samples - 0.5 - mu[0]) / ((sigma[0] + TINY) * 2 ** 0.5)))) * weight[0] +
   (0.5 * (1 + scipy.special.erf((samples + 0.5 - mu[1]) / ((sigma[1] + TINY) * 2 ** 0.5))) -
   0.5 * (1 + scipy.special.erf((samples - 0.5 - mu[1]) / ((sigma[1] + TINY) * 2 ** 0.5)))) * weight[1] +
   (0.5 * (1 + scipy.special.erf((samples + 0.5 - mu[2]) / ((sigma[2] + TINY) * 2 ** 0.5))) -
   0.5 * (1 + scipy.special.erf((samples - 0.5 - mu[2]) / ((sigma[2] + TINY) * 2 ** 0.5)))) * weight[2]
   the divided is scale_table in yours, and the the divided is scale in the referenced project. I'm a little confused about this.
   Could you give me some suggestions about those problems? Thank you.

Confused about evaluation and decompress

Hi,

I'm trying to build up an image compression model using the entropybottleneck here.

What I'm curious that is there any difference on reconstruction image between via evaluation way and via compression/decompression way.

Evaluation. What the difference on x_ ??

# Build autoencoder.
x = tf.placeholder(tf.float32, shape=[None, 16, 16, 1])
y = forward_transform(x)
y_, likelihoods = EntropyBottleneck()(y, training=False)
x_ = backward_transform(y_)

compression/decompression

x = tf.placeholder(tf.float32, shape=[None, 16, 16, 1])
y = forward_transform(x)
strings = EntropyBottleneck().compress(y)
shape = tf.shape(y)[1:]

entropy_bottleneck = EntropyBottleneck(dtype=tf.float32)
y_ = entropy_bottleneck.decompress(strings, shape, channels=5)
x_ = backward_transform(y_)

Another problem is, could you hint me a little because I'm following your guide in readme and trying to have a experiment on cifar10 dataset using simple forward/backward architecture, convolution layer with strides to downscale image and convolution transpose layer to upscale the image. Also add tf.contrib.gdn and inverse gdn into the forward/backward architecture between every conv/conv_trans layer except before the bottleneck layer.
The results are very frustrating after several day work.

I'll be very appreciated if there is any advice. Thanks.

wrong reconstructed image test on gpu

Describe the bug
Based on 'bmshj2018.py', don't change anything, I just trained a new model with 30000 steps, compressed and decompressed my test imgae (a 'png' type image with size '2560x3840') on Gpu, the recon image is wrong. There is no problem tested on cpu. The same step ckpt encode and decode 'kodim01.png' whcih size is '512x768' on gpu, the recon image is right, even endode on gpu and decode on cpu

Wrong decompressed img

System

• ubuntu 16.04
• Python 3.6.7
• conda install tensorFlow-gpu== 1.15.0
• tensorflow-compression==1.3
• gpu version [GeForce GTX 1080 Ti]

Additional context
Actually, my test image test on your releaze mode named 'mbt2018-mean-mse-1' are always correct no matter gpu or cpu been used when coding or decoding. It means GPU and CPU can be mixed used when coding and decoding eventhough recon sigma haved tiny difference.

KeyError: 'UnboundedIndexRangeEncode' while loading metagraph file.

I want to use the pretrained model and do a fine-tuning on my dataset. But this error happens while I'm trying to load the .metagraph file. Am I doing it in a wrong way since I didn't find sufficient references on Google on loading a .metagraph file. Most are about .meta file.

This is the error information:

/home/zzktb/.virtualenvs/tf1.14_py3/bin/python /home/zzktb/PycharmProjects/End2EndVideoCompression/tfoptflow/intra/intra.py
2019-08-14 15:10:50.004926: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcuda.so.1
2019-08-14 15:10:50.008137: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.008704: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1640] Found device 0 with properties: 
name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.582
pciBusID: 0000:01:00.0
2019-08-14 15:10:50.008798: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0
2019-08-14 15:10:50.009523: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcublas.so.10.0
2019-08-14 15:10:50.010185: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcufft.so.10.0
2019-08-14 15:10:50.010338: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcurand.so.10.0
2019-08-14 15:10:50.011303: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcusolver.so.10.0
2019-08-14 15:10:50.012014: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcusparse.so.10.0
2019-08-14 15:10:50.014199: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudnn.so.7
2019-08-14 15:10:50.014270: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.014877: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.015432: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1763] Adding visible gpu devices: 0
2019-08-14 15:10:50.015655: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-08-14 15:10:50.089384: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.090085: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x53433c0 executing computations on platform CUDA. Devices:
2019-08-14 15:10:50.090099: I tensorflow/compiler/xla/service/service.cc:175]   StreamExecutor device (0): GeForce GTX 1080 Ti, Compute Capability 6.1
2019-08-14 15:10:50.115681: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3192000000 Hz
2019-08-14 15:10:50.116478: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x540f790 executing computations on platform Host. Devices:
2019-08-14 15:10:50.116496: I tensorflow/compiler/xla/service/service.cc:175]   StreamExecutor device (0): <undefined>, <undefined>
2019-08-14 15:10:50.116636: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.117236: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1640] Found device 0 with properties: 
name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.582
pciBusID: 0000:01:00.0
2019-08-14 15:10:50.117261: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0
2019-08-14 15:10:50.117270: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcublas.so.10.0
2019-08-14 15:10:50.117277: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcufft.so.10.0
2019-08-14 15:10:50.117284: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcurand.so.10.0
2019-08-14 15:10:50.117292: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcusolver.so.10.0
2019-08-14 15:10:50.117299: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcusparse.so.10.0
2019-08-14 15:10:50.117306: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudnn.so.7
2019-08-14 15:10:50.117338: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.117918: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.118472: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1763] Adding visible gpu devices: 0
2019-08-14 15:10:50.118490: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0
2019-08-14 15:10:50.119416: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1181] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-08-14 15:10:50.119425: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1187]      0 
2019-08-14 15:10:50.119428: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1200] 0:   N 
2019-08-14 15:10:50.119602: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.120196: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-08-14 15:10:50.120782: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1326] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 9245 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1)
Traceback (most recent call last):
  File "/home/zzktb/PycharmProjects/End2EndVideoCompression/tfoptflow/intra/intra.py", line 32, in <module>
    saver = tf.compat.v1.train.import_meta_graph(meta_path)
  File "/home/zzktb/.virtualenvs/tf1.14_py3/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1449, in import_meta_graph
    **kwargs)[0]
  File "/home/zzktb/.virtualenvs/tf1.14_py3/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1473, in _import_meta_graph_with_return_elements
    **kwargs))
  File "/home/zzktb/.virtualenvs/tf1.14_py3/lib/python3.6/site-packages/tensorflow/python/framework/meta_graph.py", line 857, in import_scoped_meta_graph_with_return_elements
    return_elements=return_elements)
  File "/home/zzktb/.virtualenvs/tf1.14_py3/lib/python3.6/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
    return func(*args, **kwargs)
  File "/home/zzktb/.virtualenvs/tf1.14_py3/lib/python3.6/site-packages/tensorflow/python/framework/importer.py", line 400, in import_graph_def
    _RemoveDefaultAttrs(op_dict, producer_op_list, graph_def)
  File "/home/zzktb/.virtualenvs/tf1.14_py3/lib/python3.6/site-packages/tensorflow/python/framework/importer.py", line 160, in _RemoveDefaultAttrs
    op_def = op_dict[node.op]
KeyError: 'UnboundedIndexRangeEncode'

Process finished with exit code 1

Here's the snippet that I'm using which is based on tensorflow docs
https://www.tensorflow.org/api_docs/python/tf/train/import_meta_graph

import tensorflow as tf
if __name__ == '__main__':
    meta_path = '/home/zzktb/Downloads/bmshj2018-hyperprior-msssim-8.metagraph'

    with tf.compat.v1.Session() as sess:
        saver = tf.compat.v1.train.import_meta_graph(meta_path)
        saver.restore(sess, meta_path.split('.')[0])

        print('Loaded.')

Since this is not working, I tried another way as in tfci.py file as follows:

import tensorflow as tf


def load_metagraph(file):
    """
    Load metagraph file.
    :param file:
    :return: MetaGraphDef protocol buffer
    """
    with tf.io.gfile.GFile(file, 'rb') as f:
        string = f.read()

    meta_graph = tf.MetaGraphDef()
    meta_graph.ParseFromString(string)
    tf.train.import_meta_graph(meta_graph)
    return meta_graph


if __name__ == '__main__':
    meta_path = '/home/zzktb/Downloads/bmshj2018-hyperprior-msssim-8.metagraph'

    with tf.Graph().as_default():
        meta_graph = load_metagraph(meta_path)

        print('Loaded.')

And the same error information happens when executing tf.train.import_meta_graph(meta_graph). So anyone can enlighten me on how to correctly restore this MetaGraph and do a fine-tuning. I need the pre-trained variables.

undefined symbol: _ZN10tensorflow8internal21CheckOpMessageBuilder9NewStringEv

when I run:
import tensorflow_compression
that's an error:
tensorflow.python.framework.errors_impl.NotFoundError: /data/huangyujun/anaconda2/envs/tensorflow1/lib/python3.6/site-packages/tensorflow_compression/python/ops/../../cc/libtemsorflow_compression.so: undefined symbol: _ZN10tensorflow8internal21CheckOpMessageBuilder9NewStringEv

I have tried version 1.3, 1.2, 1.2b2, the same error raised.

Thanks for the help!

undefined symbol: _ZN10tensorflow8internal21CheckOpMessageBuilder9NewStringEv on Linux

Using Arch Linux and gcc 9.3.0, python 3.8, and tensorflow 2.1, I get the following error:

>>> import tensorflow_compression
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/trougnouf/.local/lib/python3.8/site-packages/tensorflow_compression/__init__.py", line 30, in <module>
    from tensorflow_compression.python.entropy_models.continuous_batched import *
  File "/home/trougnouf/.local/lib/python3.8/site-packages/tensorflow_compression/python/entropy_models/continuous_batched.py", line 20, in <module>
    from tensorflow_compression.python.entropy_models import continuous_base
  File "/home/trougnouf/.local/lib/python3.8/site-packages/tensorflow_compression/python/entropy_models/continuous_base.py", line 24, in <module>
    from tensorflow_compression.python.ops import range_coding_ops
  File "/home/trougnouf/.local/lib/python3.8/site-packages/tensorflow_compression/python/ops/range_coding_ops.py", line 26, in <module>
    ops = namespace_helper.get_ops(load_library.load_op_library(
  File "/usr/lib/python3.8/site-packages/tensorflow_core/python/framework/load_library.py", line 57, in load_op_library
    lib_handle = py_tf.TF_LoadLibrary(library_filename)
tensorflow.python.framework.errors_impl.NotFoundError: /home/trougnouf/.local/lib/python3.8/site-packages/tensorflow_compression/python/ops/../../cc/libtensorflow_compression.so: undefined symbol: _ZN10tensorflow8internal21CheckOpMessageBuilder9NewStringEv

Note that I had to add 'Programming Language :: Python :: 3.8' in build_pip_pkg.py and make a change ( if "linux" in ctx.os.name: library_filename = "libtensorflow_framework.so.2") in workspace.bzl to compile in the first place.

I tried compiling with bazel run -c opt --copt "-D_GLIBCXX_USE_CXX11_ABI=1" :build_pip_pkg as that solution came up a few times, but that didn't get rid of the error.

I know that newer versions of Linux / libs are not officially supported but I would rather not learn TensorFlow 1 now if I can help it, so I'm posting this as a placeholder.

Could you upload the trained model

The project works, but lack of the model, could you upload?

Is the entropy bottleneck layer the network of side information?

Hi,

I found that in your example bls2017.py, a function tfc.EntropyBottleneck is called. Does this function work for the side information network (mentioned in the paper 'Variational Image Compression With a Scale Hyperprior')?

Thank you so much!

Multi GPU training

Thank you for the repository! If I am not wrong the current implementation does not support multiple GPU training. Any plans to do the same in future?
Also what is the average training time on V100 GPU for the entire training i.e 2 million steps

The problem about quantization

Hello,

I want to achieve the quantization value of the image and map the quantization value to other sequence in my research. But I did not understand EntropyBottleneck at all, so I am confused with the following code and can not find the quantization value. Can you help me to get the quantization value?
y = analysis_transform(x, args.num_filters)
entropy_bottleneck = tfc.EntropyBottleneck()
string = entropy_bottleneck.compress(y)
string = tf.squeeze(string, axis=0)

Transform the quantized image back (if requested).

y_hat, likelihoods = entropy_bottleneck(y, training=False)
x_hat = synthesis_transform(y_hat, args.num_filters)

Thank you very much.

libtensorflow_compression.so not found

Hi, when I try to import tensorflow_compression, I get the following error:

tensorflow.python.framework.errors_impl.NotFoundError: /home/user/Github/compression/tensorflow_compression/python/ops/../../cc/libtensorflow_compression.so: cannot open shared object file: No such file or directory

I searched system-wise but still couldn't find a file with this name. Could anyone please help? Below are my settings:

python=3.4
tensorflow-gpu=1.13.1
tensorflow_compression=1.2

Many thanks.

Problems understanding the auxiliary loss in the entropy model

Hi Ballé,

I hope it’s ok to open an issue with regards to me not understanding the logic behind a concept in the code:

In “Variational Image Compression with A Scale Hyperprior”, Appendix 6.1 you describe how to build the entropy model for the hyperprior bottleneck-layer via a learned density model. This is mostly implemented through the EntropyModel class in ‘entropy_models.py’. Inside the example script, in the training function you use the imported entropy model to get your entropy loss ‘train_bpp’ → this part is clear.

Now to the part that I don't understand yet: During training, you also get a second loss, the auxiliary loss. Could you describe why you need this second loss? I’ll put my thoughts below, and would we be quite happy if anyone could correct them and/or comment on them:

---

From the code it looks like the second loss exists because we do not encode the entire bottleneck layer via range encoding, but only encode the symbols in given probability-interval [tail_mass/2, 1-tail_mass/2] via range encoding, and everything outside the interval via Golomb coding. Why do we do that? Is it because outliers are inefficient to encode via range encoding?

---

Given we split our symbols into a part we encode via range encoding and a part we encode via Golomb coding, then we obviously have the problem that the learned continuous PMF/CDF does not match the discrete, outlier-removed data anymore. (Reason: The CDF was trained over the entire input data). Thus, I think the second part of the build() function of the class EntropyBottleneck, line 551 to 655 in the ‘entropy_models.py’ script, is about getting the quantized PMF over the values [cdf_inverse(tail_mass/2), cdf_inverse(1-tail_mass/2)] given the continuous learned PMF over the interval [-inf, +inf].

1. Is this correct?

2. Why do we need to do that at all? Can we not just cut-off the learned PMF at the desired quantiles determined by ‘tail_mass’ and then just re-normalize the remaining sum?

Thanks for your help!

How it's related to tensorflow quantization methods?

How it's related to tensorflow built-in quantization methods? i.e. post-training quantization and quantization-aware training.

ContinuousBatchedEntropyModel.bits/8 differ significantly from compress strings length

I get a different size for the side information strings returned by continuous_batched.compress (z) than that reported by continuous_batched.bit/8 (bits).
The difference is especially significant at very low bpp, where an estimate of 0.07705 bpp becomes 0.3538 bpp.

The bottlenecks are instantiated as follow:

entropy_bottleneck = tfc.ContinuousBatchedEntropyModel(
        prior=tfc.DeepFactorized(), coding_rank=3, compression=not training)
conditional_bottleneck = tfc.ContinuousIndexedEntropyModel(
        prior_fn=tfc.NoisyNormal, index_ranges=64,parameter_fns=dict(
                loc=lambda _: 0., scale=lambda i: tf.exp(i / 8 - 5),),
                coding_rank=3,compression=True)
[...]
y = analysis_transform(x)
z = hyper_analysis_transform(abs(y))
z_tilde = entropy_bottleneck.quantize(z)
sigma = hyper_synthesis_transform(z_tilde)
y_compressed = conditional_bottleneck.compress(y, sigma)
z_compressed = entropy_bottleneck.compress(z)

print('eval: {}'.format(self.entropy_bottleneck.bits(z, training=False)/8))
print('train: {}'.format(self.entropy_bottleneck.bits(z, training=True)/8))
print('compress: {}'.format(len(self.entropy_bottleneck.compress(z).numpy()[0])))

results:

eval: [535.6596]
train: [18.767015]
compress: 12374

This is inconsistent with the ContinuousBatchedEntropyModel.bits function which states that if training is set to false then a lower-bound is returned, and vice versa. There is actually a test for this condition in continuous_batched_test.test_information_bounds but it's only ever instantiated with a NoisyNormal distribution (not learned).

System:

• Arch Linux
• Python version 3.8
• TensorFlow 2.2.0rc3
• tensorflow-compression version 6ece08b

edit:

• The conditional entropy reported bits are accurate so the side information is taking up most of the bits.
• Encoding z_tilde=entropy_bottleneck.quantize(z) with lzma yields a 792-bytes file, which is closer to the value reported by entropy_bottleneck.bits

Problem when decompress compressed.bin

enviroment:
tensorflow1.13 cuda10 compression-1.1

lee@home:~/HungSheng/end2EndCompression/compression-1.1/examples$ python bls2017.py decompress compressed.bin reconstruction.png
WARNING:tensorflow:From /home/lee/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
Traceback (most recent call last):
File "bls2017.py", line 338, in
decompress()
File "bls2017.py", line 267, in decompress
x_hat = x_hat[0, :x_shape[0], :x_shape[1], :]
File "/home/lee/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/array_ops.py", line 597, in _slice_helper
_check_index(s.stop)
File "/home/lee/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/array_ops.py", line 516, in _check_index
raise TypeError(_SLICE_TYPE_ERROR + ", got {!r}".format(idx))
TypeError: Only integers, slices (:), ellipsis (...), tf.newaxis (None) and scalar tf.int32/tf.int64 tensors are valid indices, got 756

The detailed settings for the results optimized for MS-SSIM

Hi, thanks for your excellent work and for sharing the code!

I could not find the detailed settings for the results optimized for MS-SSIM here.

Since the magnitude of the MSE-based distortion loss is quite different from the MS-SSIM-based distortion loss, I wonder what setting of lambda (the trade-off parameter) is used in the experiment measured by MS-SSIM. Hope to hear your response, thanks!

Error occurred when the command line is "import tensorflow_compression"

My environment is python3.6, tensorflow-gpu-14.0,tensorflow-compression-1.2b2 on linux.
the command line:
pip3 install --upgrade tensorflow-gpu
pip3 install tensorflow-compression
All of the above is were installed successfully.
However, when I input import tensorflow_compression, error occurred:

Traceback (most recent call last):
File "", line 1, in
File "/home/xuxinzi/research-model-compression/proj/compression/tensorflow_compression/init.py", line 31, in
from tensorflow_compression.python.layers.entropy_models import *
File "/home/xuxinzi/research-model-compression/proj/compression/tensorflow_compression/python/layers/entropy_models.py", line 28, in
from tensorflow_compression.python.ops import range_coding_ops
File "/home/xuxinzi/research-model-compression/proj/compression/tensorflow_compression/python/ops/range_coding_ops.py", line 28, in
"../../cc/libtensorflow_compression.so")))
File "/home/xuxinzi/.local/lib/python3.6/site-packages/tensorflow/python/framework/load_library.py", line 61, in load_op_library
lib_handle = py_tf.TF_LoadLibrary(library_filename)
tensorflow.python.framework.errors_impl.NotFoundError: /home/xuxinzi/research-model-compression/proj/compression/tensorflow_compression/python/ops/../../cc/libtensorflow_compression.so: cannot open shared object file: No such file or directory

Cudnn version

Hi, I used commad lines to config the environment:

conda create --name ENV_NAME python=3.6 cudatoolkit=10.0 cudnn
conda activate ENV_NAME
pip install tensorflow-gpu tensorflow-compression

My conda list is:

_libgcc_mutex 0.1 main
absl-py 0.7.1
astor 0.8.0
ca-certificates 2019.5.15 1
certifi 2019.6.16 py36_1
cudatoolkit 10.0.130 0
cudnn 7.6.0 cuda10.0_0
gast 0.2.2
google-pasta 0.1.7
grpcio 1.23.0
h5py 2.9.0
Keras-Applications 1.0.8
Keras-Preprocessing 1.1.0
libedit 3.1.20181209 hc058e9b_0
libffi 3.2.1 hd88cf55_4
libgcc-ng 9.1.0 hdf63c60_0
libstdcxx-ng 9.1.0 hdf63c60_0
Markdown 3.1.1
ncurses 6.1 he6710b0_1
numpy 1.17.0
openssl 1.1.1c h7b6447c_1
pip 19.2.2 py36_0
protobuf 3.9.1
python 3.6.9 h265db76_0
readline 7.0 h7b6447c_5
scipy 1.3.1
setuptools 41.0.1 py36_0
six 1.12.0
sqlite 3.29.0 h7b6447c_0
tensorboard 1.14.0
tensorflow-compression 1.2
tensorflow-estimator 1.14.0
tensorflow-gpu 1.14.0
termcolor 1.1.0
tk 8.6.8 hbc83047_0
Werkzeug 0.15.5
wheel 0.33.4 py36_0
wrapt 1.11.2
xz 5.2.4 h14c3975_4
zlib 1.2.11 h7b6447c_3

And I ran test code successfully. However, I met an error when running tfci.py. The error says

Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.

I searched on the Internet and it says that the tensorflow-gpu version doesn't match with cudnn+cuda version. I use tensorflow-gpu1.14.0. What's your cudnn version?

Op type not registered 'PmfToQuantizedCdf' in binary running

Describe the bug
When I use freeze_graph to generate pb model, I get an error
tensorflow.python.framework.errors_impl.NotFoundError: Op type not registered 'PmfToQuantizedCdf' in binary running on ...

System:

• ubuntu 18
• python3.6 anconda
• tf1.15
• tensorflow-compression1.3

I want to know what the reason is or how to solve it?
Looking forward to receiving a reply, thank you!

Issue while training with Resource Exchaust Error. I have configuration as per attached image and gpu usage while training on png dataset.

Problem When pip install tensorflow-compression

when I install this package by using "pip install tensorflow-compression"
It comes out " Could not find a version that satisfies the requirement tensorflow-compression (from versions: ) No matching distribution found for tensorflow-compression"
How can I fix it. Thanks for reply!

libtensorflow_compression.so: cannot open shared object file

Describe the bug
When importing the library, it looks for libtensorflow_compression.so in cc folder but this does not exist in the master version and hence throws the error.

To Reproduce
Steps to reproduce the behavior:

1. Install tf2.1,tfp0.9.0 and tfc master version
2. import tensorflow_compression as tfc
3. Error:

Traceback (most recent call last):
File "", line 1, in
File "/home/work/compression-master/tensorflow_compression/init.py", line 30, in
from tensorflow_compression.python.entropy_models.continuous_batched import *
File "/home/work/compression-master/tensorflow_compression/python/entropy_models/continuous_batched.py", line 22, in
from tensorflow_compression.python.entropy_models import continuous_base
File "/home/work/compression-master/tensorflow_compression/python/entropy_models/continuous_base.py", line 24, in
from tensorflow_compression.python.ops import range_coding_ops
File "/home/work/compression-master/tensorflow_compression/python/ops/range_coding_ops.py", line 28, in
"../../cc/libtensorflow_compression.so")))
File "/home/virtualenvs/tfcomp/lib/python3.6/site-packages/tensorflow_core/python/framework/load_library.py", line 57, in load_op_library
lib_handle = py_tf.TF_LoadLibrary(library_filename)
tensorflow.python.framework.errors_impl.NotFoundError: /home/work/compression-master/tensorflow_compression/python/ops/../../cc/libtensorflow_compression.so: cannot open shared object file: No such file or directory

Expected behavior
Should be able to import without issues

System (please complete the following information):

• OS Ubuntu20.04
• Python version 3.6.9, built from source
• TensorFlow version 2.1.2, pip
• tensorflow-compression version master

I need a tf2 compatible tfc version hence trying to use the master version

docker image

Has anybody built the docker image for this project successfully?

Unable to retain the original resolution on reconstruction.

Hello,

I am compressing an image of resolution 426 x 240.

The architecture for analysis transformation is:

self._layers = [
            tfc.SignalConv2D(
                self.num_filters, (3, 3), name="layer_0", corr=True, strides_down=2,
                padding="same_zeros", use_bias=True,
                activation=tfc.GDN(name="gdn_0")),

            tfc.SignalConv2D(
                self.num_filters, (3, 3), name="layer_1", corr=True, strides_down=2,
                padding="same_zeros", use_bias=True,
                activation=tfc.GDN(name="gdn_1")),

            tfc.SignalConv2D(
                self.num_filters, (3, 3), name="layer_2", corr=True, strides_down=2,
                padding="same_zeros", use_bias=True,
                activation=tfc.GDN(name="gdn_2")),

            tfc.SignalConv2D(
                self.num_filters, (3, 3), name="layer_3", corr=True, strides_down=2,
                padding="same_zeros", use_bias=True,
                activation=None),
        ]

So the shape of latent representation(excluding batch and channel) is 27 x 15.

The architecture of synthesis transform is exactly inverse of the analysis transform. But the shape of the reconstructed image is 432 x 240. I tried different padding options, but I couldn’t figure out what combination would get me the original shape. Some combinations would reconstruct the output of the same height but different width, and some would reconstruct the output of the same width but different height. This problem seems not only for this specific resolution but also for other resolutions as well (when the height and width aren’t the multiple of 16).

Here is a complete network architecture, https://github.com/tukilabs/Video-Compression-Net/blob/master/utils/network.py
(class AnalysisTransform and class SynthesisTransform)

PS, I’m using this wonderful work for video compression.

Tfci.py looking for CUDA libs even if tensorflow-gpu is not installed

I'm trying to encode few images with the pretrained models.
The system is a VM running on a cluster, OS is Ubuntu 18.04. Python is 3.6.9, tensorflow is 1.15 and tensorflow-compression is 1.3. The system has no GPU, so I can't use tensorflow-gpu.
Following the readme file instructions, I run:
python3 tfci.py compress bmshj2018-factorized-mse-3 test_img_001.png

then I get the following output:
2020-02-22 22:41:08.854368: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/home/user/research/codecs/kakadu:/home/user/research/codecs/kakadu 2020-02-22 22:41:08.854410: E tensorflow/stream_executor/cuda/cuda_driver.cc:318] failed call to cuInit: UNKNOWN ERROR (303) 2020-02-22 22:41:08.854451: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (matlabWS): /proc/driver/nvidia/version does not exist 2020-02-22 22:41:08.854810: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: FMA 2020-02-22 22:41:08.889214: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3491910000 Hz 2020-02-22 22:41:08.892134: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x4c44dc0 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2020-02-22 22:41:08.892369: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version INFO:tensorflow:Saver not created because there are no variables in the graph to restore I0222 22:41:09.164153 140027371087680 saver.py:1503] Saver not created because there are no variables in the graph to restore

It's looking for CUDA libs, even if there's no GPU. It's like if it's stuck with using tensorflow-gpu.
I run the previous version of tensorflow-compression in the past on the exact same system without any problems, so I think it's a bug with the 1.3.

Thanks for the help.

Python 3.7 / Linux Wheel

Unfortunately, PyPi only has wheels up to python 3.6 for linux available (https://pypi.org/project/tensorflow-compression/#files). If you could add 3.7, that would make things even a little smoother.

Anyway, thanks a lot for you work on the library!

Problem with EntropyBottleneck

Hello,
I'm trying to use some of your layers ,but I find some problems.When I read the bls.py ,I find you used an auxiliary loss，and the code is 'aux_step = aux_optimizer.minimize(entropy_bottleneck.losses[0])' , but I couldn't find losses in the layer of EntropyBottleneck. Can you help with this?

Problems about images used to train

Hello,
In your paper"END-TO-END OPTIMIZED IMAGE COMPRESSION" , i find that you used 6507 ImageNet images as training set .But i didn't find them. Could you tell me how to download them?

Thank you very much.

How to pick the 6507 images as training set

Hi,

Thanks a lot for your brilliant work. :)
I just wonder how you extract the 6507-image training dataset from imagenet as you mentioned in the paper End To End Optimized Image Compression. Did you pick the images just randomly or by following any rules?

BTW I read the earlier issue #29 and found that I should send a email to ask the question, but I have no idea how to send an email to a github repository's mailing list...

Would you please help me? Thanks so much

The calculation of MSE is not correct

Hi, Johannes:
Thanks a lot for your released code. There is a small error in your calculation of MSE, since the num_pixels = batchsize* patchsizepatchsize, which is reasonable for bpp, but for MSE, it should be num_pixels = batchsize patchsizepatchsize3, or you can use tf.reduce_mean instead.
What's more, I find that the R-D loss curve oscillates a lot in training phase, and it does not converge with small lambda(such as 0.001). Can you give me some advice to train the model well.
Best regards.

Question about the encoding range for \hat{y}?

Hi!
I saw the SCALES_MIN=0.11, SCALES_MAX=256 in training file, and I find the maximal length in entropy_model.py at line830 is about 1024. But I wonder what happends when \hat{y} is out of range?

Thanks!

Is it possible to get the encoded sub-patches from a encoded image directly?

Hi folks,
I'm wondering if it is possible to get the encoded sub-patches from a encoded image directly, rather than decoding the encoded image and extract the sub-patches then encode the sub-patch again.

Any one with clues will be appreciated.

Make a release

Is it possible to make a release, so that my docker file could pull a frozen version instead of the master?

forward_transform

Hi.
This is probably a very stupid question but where is the function y = forward_transform(x) defined? I'm getting an error that it is not defined. Is that supposed to be replaced by some custom-defined forward transform function?
Thanks!

Decode some incorrect pictures

Hello, when I was trying this code，In the same intermediate file, sometimes it is normal to decode the picture, sometimes it is abnormal to decode the picture

The question about quantization

Hello,

Thanks for your reply to my previous question. I have some new questions to consult you.

1. The q in your paper do not appear in your code, so the quantization representation is just y_hat and do not consider q in code? I ran the code and the value of y_hat is not an integer, just as followings:
   [[[[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
   [ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
   [ 0.59345233 1.0139648 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
   ...
   [ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
   [ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
   [ 0.59345233 2.0139647 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]]

[[ 0.59345233 1.0139648 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 -0.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 1.0139648 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
...
[ 0.59345233 -1.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 1.0139648 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]]

[[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 -1.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
...
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]]

...

[[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
...
[ 0.59345233 -0.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 1.0139648 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 1.0139648 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]]

[[ 0.59345233 -0.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 -0.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
...
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 -0.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]]

[[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
...
[ 0.59345233 -0.9860352 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]
[ 0.59345233 0.01396477 0.13712975 ... 0.1063276 -0.2717389 -0.42567667]]]]
By the way, how can I get the level of the y_hat (i.e. the quantitative level), as you can see there are many ellipses in the above value?

2. Is the compressed string made up of binary streams? How can I open the file of compressed.bin as I want to use the content of the compressed string if it is a binary stream.

Thank you very much.

Getting stuck in one point during training

Hello,
I use python 3.6 installed on miniconda in a Linux OS. I have installed tensorflow 1.15 and tensorflow compression 1.3 on the machine. I ran the command "python -m hific.train --config mselpips --ckpt_dir ckpts/mse_lpips --num_steps 1M --tfds_dataset_name coco2014" and the code starts training but after the step "Downloading LPIPS weights: http://rail.eecs.berkeley.edu/models/lpips/net-lin_alex_v0.1.pb -> lpips_weight__net-lin_alex_v0.1.pb", it gets idle and does not do anything.

In advance, I appreciate your time and answers.