https://github.com/carpedm20/lstm-char-cnn-tensorflow

in progress
https://github.com/carpedm20/lstm-char-cnn-tensorflow

cnn lstm nlp tensorflow

Last synced: about 1 month ago
JSON representation

in progress

• Host: GitHub
• URL: https://github.com/carpedm20/lstm-char-cnn-tensorflow
• Owner: carpedm20
• License: mit
• Created: 2015-12-10T16:31:03.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2018-07-27T11:04:42.000Z (about 6 years ago)
• Last Synced: 2024-07-31T02:16:22.191Z (about 1 month ago)
• Topics: cnn, lstm, nlp, tensorflow
• Language: Python
• Homepage:
• Size: 8.76 MB
• Stars: 761
• Watchers: 60
• Forks: 244
• Open Issues: 17
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-tensorflow - Character-Aware Neural Language Models - TensorFlow implementation of [Character-Aware Neural Language Models](http://arxiv.org/abs/1508.06615) (Models/Projects)
• Awesome-TensorFlow-Chinese - Character-Aware Neural Language Models - TensorFlow implementation of [Character-Aware Neural Language Models](http://arxiv.org/abs/1508.06615) (模型项目 / 微信群)
• awesome-tensorflow - Character-Aware Neural Language Models - TensorFlow implementation of [Character-Aware Neural Language Models](http://arxiv.org/abs/1508.06615) (Models/Projects)

README

        
Character-Aware Neural Language Models

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

Tensorflow implementation of [Character-Aware Neural Language Models](http://arxiv.org/abs/1508.06615). The original code of author can be found [here](https://github.com/yoonkim/lstm-char-cnn).

![model.png](./assets/model.png)

This implementation contains:

1. Word-level and Character-level Convolutional Neural Network

2. Highway Network

3. Recurrent Neural Network Language Model

*The current implementation has a performance issue. See [#3](https://github.com/carpedm20/lstm-char-cnn-tensorflow/issues/3).*

Prerequisites

-------------

- Python 2.7 or Python 3.3+

- [Tensorflow](https://www.tensorflow.org/)

Usage

-----

To train a model with `ptb` dataset:

    $ python main.py --dataset ptb

To test an existing model:

    $ python main.py --dataset ptb --forward_only True

To see all training options, run:

    $ python main.py --help

which will print

    usage: main.py [-h] [--epoch EPOCH] [--word_embed_dim WORD_EMBED_DIM]

                  [--char_embed_dim CHAR_EMBED_DIM]

                  [--max_word_length MAX_WORD_LENGTH] [--batch_size BATCH_SIZE]

                  [--seq_length SEQ_LENGTH] [--learning_rate LEARNING_RATE]

                  [--decay DECAY] [--dropout_prob DROPOUT_PROB]

                  [--feature_maps FEATURE_MAPS] [--kernels KERNELS]

                  [--model MODEL] [--data_dir DATA_DIR] [--dataset DATASET]

                  [--checkpoint_dir CHECKPOINT_DIR]

                  [--forward_only [FORWARD_ONLY]] [--noforward_only]

                  [--use_char [USE_CHAR]] [--nouse_char] [--use_word [USE_WORD]]

                  [--nouse_word]

    optional arguments:

      -h, --help            show this help message and exit

      --epoch EPOCH         Epoch to train [25]

      --word_embed_dim WORD_EMBED_DIM

                            The dimension of word embedding matrix [650]

      --char_embed_dim CHAR_EMBED_DIM

                            The dimension of char embedding matrix [15]

      --max_word_length MAX_WORD_LENGTH

                            The maximum length of word [65]

      --batch_size BATCH_SIZE

                            The size of batch images [100]

      --seq_length SEQ_LENGTH

                            The # of timesteps to unroll for [35]

      --learning_rate LEARNING_RATE

                            Learning rate [1.0]

      --decay DECAY         Decay of SGD [0.5]

      --dropout_prob DROPOUT_PROB

                            Probability of dropout layer [0.5]

      --feature_maps FEATURE_MAPS

                            The # of feature maps in CNN

                            [50,100,150,200,200,200,200]

      --kernels KERNELS     The width of CNN kernels [1,2,3,4,5,6,7]

      --model MODEL         The type of model to train and test [LSTM, LSTMTDNN]

      --data_dir DATA_DIR   The name of data directory [data]

      --dataset DATASET     The name of dataset [ptb]

      --checkpoint_dir CHECKPOINT_DIR

                            Directory name to save the checkpoints [checkpoint]

      --forward_only [FORWARD_ONLY]

                            True for forward only, False for training [False]

      --noforward_only

      --use_char [USE_CHAR]

                            Use character-level language model [True]

      --nouse_char

      --use_word [USE_WORD]

                            Use word-level language [False]

      --nouse_word

but more options can be found in [models/LSTMTDNN](./models/LSTMTDNN.py) and [models/TDNN](./models/TDNN.py).

Performance

-----------

**Failed to reproduce the results of paper (2016.02.12)**. If you are looking for a code that reproduced the paper's result, see https://github.com/mkroutikov/tf-lstm-char-cnn.

![loss](./assets/performance.png)

The perplexity on the test sets of Penn Treebank (PTB) corpora.

|       Name      | Character embed | LSTM hidden units | Paper (Y Kim 2016) |  This repo. |

|:---------------:|:---------------:|:-----------------:|:------------------:|:-----------:|

| LSTM-Char-Small |        15       |        100        |        92.3        | in progress |

| LSTM-Char-Large |        15       |        150        |        78.9        | in progress |

Author

------

Taehoon Kim / [@carpedm20](http://carpedm20.github.io/)