Skip to content
Snippets Groups Projects
Select Git revision
  • master default protected
  • update_links
  • load_quant_aware_train_checkpoint
  • kd_fixes
  • api_changes
  • lattice_regularization
  • inception_support
  • ax_integration
  • ptq_propagate_settings
  • gste
  • 4-bit-qat
  • scale_approximation
  • v0.3.2
  • amc_before_pytorch1.1_merge
  • v0.3.1
  • v0.3.0
  • v0.2.1
  • v0.2.0
  • v0.1.0
19 results
Compare
user avatar
LSTM: Modular implementation + Post-Train Quantization Sample (#196)
Lev Zlotnik authored 
* Introduce a modular, Python-level implementation of LSTM/LSTMCell
  using existing PyTorch nn.Modules as building blocks
* This allows quantization of weights and internal activations of
  LSTM layers using the existing Quantizer. 
  (In the PyTorch implementation of RNN/LSTM only the weights are 
  exposed at the Python level, whereas the internal activations are 
  "hidden" in C++ code.)
* Supports stacked (multi-layer) and bi-directional LSTM
* Implemented conversion functions from PyTorch LSTM module to
  our LSTM module and vice-versa
* Tests for modular implementation correctness and for conversions
* Jupyter notebook showing post-training quantization of a language
  model
a3c8d86f
History
user avatar a3c8d86f
History
Name Last commit Last update
..
data/wikitext-2
README.md
data.py
generate.py
logging.conf
main.py
manual_lstm_pretrained_stats.yaml
manual_lstm_pretrained_stats_new.yaml
model.py
quantize_lstm.ipynb
README.md

Word-level language modeling RNN

This example trains a multi-layer RNN (Elman, GRU, or LSTM) on a language modeling task. By default, the training script uses the Wikitext-2 dataset, provided. The trained model can then be used by the generate script to generate new text.

python main.py --cuda --epochs 6        # Train a LSTM on Wikitext-2 with CUDA, reaching perplexity of 117.61
python main.py --cuda --epochs 6 --tied # Train a tied LSTM on Wikitext-2 with CUDA, reaching perplexity of 110.44
python main.py --cuda --tied            # Train a tied LSTM on Wikitext-2 with CUDA for 40 epochs, reaching perplexity of 87.17
python generate.py                      # Generate samples from the trained LSTM model.

The model uses the nn.RNN module (and its sister modules nn.GRU and nn.LSTM) which will automatically use the cuDNN backend if run on CUDA with cuDNN installed.

During training, if a keyboard interrupt (Ctrl-C) is received, training is stopped and the current model is evaluated against the test dataset.

The main.py script accepts the following arguments:

optional arguments:
  -h, --help         show this help message and exit
  --data DATA        location of the data corpus
  --model MODEL      type of recurrent net (RNN_TANH, RNN_RELU, LSTM, GRU)
  --emsize EMSIZE    size of word embeddings
  --nhid NHID        number of hidden units per layer
  --nlayers NLAYERS  number of layers
  --lr LR            initial learning rate
  --clip CLIP        gradient clipping
  --epochs EPOCHS    upper epoch limit
  --batch-size N     batch size
  --bptt BPTT        sequence length
  --dropout DROPOUT  dropout applied to layers (0 = no dropout)
  --decay DECAY      learning rate decay per epoch
  --tied             tie the word embedding and softmax weights
  --seed SEED        random seed
  --cuda             use CUDA
  --log-interval N   report interval
  --save SAVE        path to save the final model

With these arguments, a variety of models can be tested. As an example, the following arguments produce slower but better models:

python main.py --cuda --emsize 650 --nhid 650 --dropout 0.5 --epochs 40           # Test perplexity of 80.97
python main.py --cuda --emsize 650 --nhid 650 --dropout 0.5 --epochs 40 --tied    # Test perplexity of 75.96
python main.py --cuda --emsize 1500 --nhid 1500 --dropout 0.65 --epochs 40        # Test perplexity of 77.42
python main.py --cuda --emsize 1500 --nhid 1500 --dropout 0.65 --epochs 40 --tied # Test perplexity of 72.30

Perplexities on PTB are equal or better than Recurrent Neural Network Regularization (Zaremba et al. 2014) and are similar to Using the Output Embedding to Improve Language Models (Press & Wolf 2016 and Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling (Inan et al. 2016), though both of these papers have improved perplexities by using a form of recurrent dropout (variational dropout).