ora-io/keras2circom

python tool to transpile a tf.keras model into a circom circuit

keras2circom

keras2circom is a python tool that transpiles a tf.keras model into a circom circuit.

Installation

First, clone the repository:

git clone https://github.com/socathie/keras2circom.git

Then, install the dependencies. You can use pip:

pip install -r requirements.txt

If you use conda, you can also create a new environment with the following command:

conda env create -f environment.yml

You will also need to install circom and snarkjs. You can run the following commands to install them:

bash setup-circom.sh

Last but not least, run

npm install

Usage

To use the package, you can run the following command:

python main.py <model_path> [-o <output_dir>] [--raw]

For example, to transpile the model in models/model.h5 into a circom circuit, you can run:

python main.py models/model.h5

The output will be in the output directory.

If you want to transpile the model into a circom circuit with "raw" output, i.e. no ArgMax at the end, you can run:

python main.py models/model.h5 --raw

Testing

To test the package, you can run the following command:

npm test

keras2circom's Issues

witeness of the output circuit.circom file generation failed

Hi, i am a new learner in zkml and i am using this keras2circom with dokcer python3.9.16, and i got this problem,
when i tried to generate witness with the output circuit.circom and circuit.json, i had this error:

~/Desktop/PrivacyComputing/proof-of-sql/practices/keras2circom/output]$ node circuit_js/generate_witness.js circuit_js/circuit.wasm circuit.json circuit.wtns /Desktop/PrivacyComputing/proof-of-sql/practices/keras2circom/output/circuit_js/witness_calculator.js:167 throw new Error(Not all inputs have been set. Only ${input_counter} out of ${this.instance.exports.getInputSize()}); ^ Error: Not all inputs have been set. Only 674 out of 1458 at WitnessCalculator._doCalculateWitness (/Desktop/PrivacyComputing/proof-of-sql/practices/keras2circom/output/circuit_js/witness_calculator.js:167:12) at WitnessCalculator.calculateWTNSBin (/Desktop/PrivacyComputing/proof-of-sql/practices/keras2circom/output/circuit_js/witness_calculator.js:212:20) at /Desktop/PrivacyComputing/proof-of-sql/practices/keras2circom/output/circuit_js/generate_witness.js:15:38

and here are the logs when i generated the circom file:

root@d4a5c9f0cdc9:/usr/wzj# python main.py models/alt_model.h5 2023-04-11 02:38:43.782470: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-04-11 02:38:44.014700: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0. 2023-04-11 02:38:44.023624: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory 2023-04-11 02:38:44.023696: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. 2023-04-11 02:38:44.064497: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2023-04-11 02:38:45.111990: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory 2023-04-11 02:38:45.112222: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory 2023-04-11 02:38:45.112262: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly. 2023-04-11 02:38:46.222013: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory 2023-04-11 02:38:46.222106: W tensorflow/stream_executor/cuda/cuda_driver.cc:263] failed call to cuInit: UNKNOWN ERROR (303) 2023-04-11 02:38:46.222317: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (d4a5c9f0cdc9): /proc/driver/nvidia/version does not exist 2023-04-11 02:38:46.222715: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.

am i using keras2circom wrong? looking forward to your anser, thank you.

python main.py failed

Hello sir, sorry to bother, but i got an error when i run python main.py keras2circom/test.h5 with test.h5 generated by my mnist.py which is copied from your best_practice.ipynb.

Here is the mnist.py file:

import json
import os
from tensorflow.keras import Model
from tensorflow.keras.layers import (
    Input,
    AveragePooling2D,
    BatchNormalization,
    Conv2D,
    Dense,
    GlobalAveragePooling2D,
    Lambda,  # only for polynomial activation in the form of `Lambda(lambda x: x**2+x)`
    Softmax,
)
from tensorflow.keras.utils import to_categorical
current_dir = os.getcwd()
from tensorflow.keras.datasets import mnist
(X_train, y_train), (X_test, y_test) = mnist.load_data(path=current_dir + '/mnist.npz')
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)
X_train = X_train.reshape(X_train.shape[0], 28, 28, 1)
X_test = X_test.reshape(X_test.shape[0], 28, 28, 1)
inputs = Input(shape=(28,28,1))
out = Conv2D(4, 3, use_bias=False)(inputs)
out = BatchNormalization()(out)
out = Lambda(lambda x: x**2+x)(out) # best practice: use polynomial activation instead of ReLU
out = AveragePooling2D()(out) # best practice: use AveragePooling2D instead of MaxPooling2D
out = Conv2D(16, 3, use_bias=False)(out)
out = BatchNormalization()(out)
out = Lambda(lambda x: x**2+x)(out)
out = AveragePooling2D()(out)
out = GlobalAveragePooling2D()(out) # best practice: use GlobalAveragePooling2D instead of Flatten
out = Dense(10, activation=None)(out)
out = Softmax()(out)
model = Model(inputs, out)
model.summary()
model.compile(
    loss='categorical_crossentropy',
    optimizer='adam',
    metrics=['acc']
)
model.fit(X_train, y_train, epochs=100, batch_size=128, validation_data=(X_test, y_test))
model.save('test.h5')
model2 = Model(model.input, model.layers[-2].output)
model2.layers[-1]
X = X_test[[0]]
y = model2.predict(X)
y
for layer in model.layers:
    print(layer.__class__.__name__, layer.get_config())
    try:
        print(layer.get_config()['function'])
    except:
        pass
    print(layer.get_input_shape_at(0), layer.get_output_shape_at(0))
    try:
        print(layer.get_weights()[0].shape)
        print(layer.get_weights()[1].shape)
    except:
        pass
with open("test.json", "w") as f:
    json.dump({'X': X.flatten().tolist(), 'y': y.flatten().tolist()}, f)

and here is the error:

root@984cb5a21ce9:/usr/wzj# python main.py keras2circom/test.h5                                                                                             
2023-04-12 02:16:30.752955: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA                                            
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.                                                                 
2023-04-12 02:16:31.020164: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.                    
2023-04-12 02:16:31.045363: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory                                                                                    
2023-04-12 02:16:31.045446: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2023-04-12 02:16:31.094098: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2023-04-12 02:16:32.528631: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
2023-04-12 02:16:32.528909: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2023-04-12 02:16:32.528941: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
2023-04-12 02:16:33.942876: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory
2023-04-12 02:16:33.942948: W tensorflow/stream_executor/cuda/cuda_driver.cc:263] failed call to cuInit: UNKNOWN ERROR (303)
2023-04-12 02:16:33.943119: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (984cb5a21ce9): /proc/driver/nvidia/version does not exist
2023-04-12 02:16:33.943566: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Traceback (most recent call last):
  File "/usr/wzj/main.py", line 25, in <module>
    main()
  File "/usr/wzj/main.py", line 21, in main
    transpiler.transpile(args['<model.h5>'], args['--output'], args['--raw'])
  File "/usr/wzj/keras2circom/transpiler.py", line 16, in transpile
    circuit.add_components(transpile_layer(layer))
  File "/usr/wzj/keras2circom/transpiler.py", line 80, in transpile_layer
    raise ValueError('Only polynomial activation functions are supported')
ValueError: Only polynomial activation functions are supported

and i printed the value of s.ratio() in transpiler.py:

0.6615384615384615

is this some kind of accuracy problem?

Not all inputs have been set

Hello. I attempted to run the example you provided and got a circuit.circom and a circuit.json file. But I found the number of input signals in circuit.json is different from the number needed by circuits.circom. So i got the following error when generating witness.

Not all inputs have been set. Only 1074 out of 17155
circuit: main.cpp:278: int main(int, char**): Assertion `false' failed.
Aborted (core dumped)