davidbelanger/spen

Structured Prediction Energy Networks in Torch

spen's Issues

Prediction part code

Hi, could you please tell me where is the code for solving equation (2) of your paper? I spent a long time studying your code but failed to find it.

Questions regarding the model

Since I haven't gone deeply into this package yet, just wonder if this package/tool can allow us to define arbitrary graphs (that have exact inference) for structured prediction task.

I couldn't find which part of the code is used to "build" the network.

Multi-Label Classification's data link down

Hi,

I'm studying your paper and apparently the link for getting the data of the Multi-Label Classification task is down !
I'm wondering where could I find these data ?

Best,

Nicolas

Installing problem

Hi, thank you for the great project. I am trying to run it, and having a wired problem in all different computers.

0. SSH to my ubuntu machine

1. Install Torch (from http://torch.ch/docs/getting-started.html)

git clone https://github.com/torch/distro.git ~/torch --recursive
cd ~/torch; bash install-deps;
./install.sh
source ~/.bashrc

2. Install CUDA development kit - This is needed for cutorch (From nvidia)

wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_9.0.176-1_amd64.deb

sudo dpkg -i cuda-repo-ubuntu1604_9.0.176-1_amd64.deb``sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/7fa2af80.pub

sudo apt-get update

sudo apt-get install cuda

3. Dependency for SPEN (From SPEN github)
   cd ~
   git clone https://github.com/davidBelanger/SPEN.git
   luarocks install torch
   luarocks install nn
   luarocks install cutorch
   luarocks install cunn
   luarocks install optim
   luarocks install nngraph

git clone https://github.com/davidBelanger/torch-util.git

4. Run
   chmod +x quick_start_tagging.sh # example
   ./quick_start_tagging.sh

But I still get

heejin@hubuntu:~/SPEN$ ./quick_start_tagging.sh
Creating Synthetic Train/Test Data by Drawing from a CRF
/home/heejin/torch/install/bin/luajit: /home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: /home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: /home/heejin/torch/install/share/lua/5.1/torch/init.lua:102: class nn.OneHot has been already assigned a parent class

stack traceback:
[C]: in function 'error'
/home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: in function 'require'
test/test_data_serialization_and_loading.lua:1: in main chunk
[C]: in function 'dofile'
...ejin/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x00405d50

Training a CRF on the Data to Test that Obtainable Accuracy Approaches the Accuracy of the True Model
/home/heejin/torch/install/bin/luajit: /home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: /home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: /home/heejin/torch/install/share/lua/5.1/torch/init.lua:102: class nn.OneHot has been already assigned a parent class

stack traceback:
[C]: in function 'error'
/home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: in function 'require'
test/test_chain_crf_mle_from_file.lua:2: in main chunk
[C]: in function 'dofile'
...ejin/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x00405d50

Training a SPEN on the Data using tag_cmd.sh
tag_cmd.sh: 5: [: 0: unexpected operator
-gradient_clip 1.0 -optim_method adam -evaluation_frequency 25 -save_frequency 25 -adam_epsilon 1e-8 -batches_per_epoch 100 -learning_rate_decay 0.0 -learning_rate 0.001 -num_epochs 30 -training_mode pretrain_unaries
running in tag-runs/Sat_Oct__7_16:48:17_CDT_2017
/home/heejin/torch/install/bin/luajit: /home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: /home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: /home/heejin/torch/install/share/lua/5.1/torch/init.lua:102: class nn.OneHot has been already assigned a parent class

stack traceback:
[C]: in function 'error'
/home/heejin/torch/install/share/lua/5.1/trepl/init.lua:389: in function 'require'
main.lua:4: in main chunk
[C]: in function 'dofile'
...ejin/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x00405d50

Could you help please?

Thank you.

Denoising example

Hello David, I cannot get working SPEN denoising example. First I got an exception about inference_net being nil in RNNInference.lua, which I fixed somehow by setting inference_net = nn.Identity. But now I get an error because inference_net is nil:

2: attempt to index local 'module' (a nil value)
stack traceback:
        ...e/dulyanov/torch/install/share/lua/5.1/nn/Sequential.lua:12: in function 'add'
        ./RNNInference.lua:68: in function '__init'

Can you please try to reproduce it?

I used this fork to run the experiments, since it provided a helpful script. Wondering if denoising worked for xuexue.