Home

Awesome

docker-torch-rnn

Docker images for using torch-rnn (https://github.com/jcjohnson/torch-rnn)

Available tags

How to

More details here: https://github.com/jcjohnson/torch-rnn#usage

CPU Only

  1. Start bash in the container

    • docker run --rm -ti crisbal/torch-rnn:base bash
  2. Preprocess the sample data

    python scripts/preprocess.py \
    --input_txt data/tiny-shakespeare.txt \
    --output_h5 data/tiny-shakespeare.h5 \
    --output_json data/tiny-shakespeare.json
    
  3. Train

    th train.lua \
    -input_h5 data/tiny-shakespeare.h5 \
    -input_json data/tiny-shakespeare.json \
    -gpu -1
    
  4. Sample

    • th sample.lua -checkpoint cv/checkpoint_10000.t7 -length 2000 -gpu -1

CUDA

  1. Install nvidia-docker

  2. Start bash in the container

    • nvidia-docker run --rm -ti lordalfred/docker-torch-rnn:cuda10.0 bash
  3. Preprocess the sample data

    python scripts/preprocess.py \
    --input_txt data/tiny-shakespeare.txt \
    --output_h5 data/tiny-shakespeare.h5 \
    --output_json data/tiny-shakespeare.json
    
  4. Train

    th train.lua \
    -input_h5 data/tiny-shakespeare.h5 \
    -input_json data/tiny-shakespeare.json
    
  5. Sample

    • th sample.lua -checkpoint cv/checkpoint_10000.t7 -length 2000