Character Level CNN

Live Demo

Links & Resources

Character Level Convolutional Neural Network

I've been for a quite long time been interested in character level NLP models for text classification and this model really caught my attention.

character cnn

So I decided to implement it for these various reasons:

  • Based on the paper, it's powerful in text classification (see benchmark) even though it doesn't have any notion of semantics
  • You don't need to apply any text preprocessing (tokenization, lemmatization, stemming ...) while using it
  • Its handles misspelled words and out-of-vocabulary tokes, by desing
  • It's fast to train
  • It doesn't require storing a large word embedding matrix. Hence, it's lightweight and you can deploy it in production easily

When I shared my implementation on Twitter, it quickly went viral:

Dependencies

  • numpy
  • pandas
  • sklearn
  • PyTorch 0.4.1
  • tensorboardX
  • Tensorflow (to be able to run TensorboardX)

Structure of the code

At the root of the project, you will have:

  • train.py: used for training a model
  • predict.py: used for the testing and inference
  • config.json: a configuration file for storing model parameters (number of filters, neurons)

  • src: a folder that contains:

    • cnn_model.py: the actual CNN model (model initialization and forward method)
    • data_loader.py: the script responsible of passing the data to the training after processing it
    • utils.py: a set of utility functions for text preprocessing (url/hashtag/user_mention removal)

How to use the code

Training

Launch train.py with the following arguments:

  • data_path: path of the data. Data should be in csv format with at least a column for text and a column for the label
  • validation_split: the ratio of validation data. default to 0.2
  • label_column: column name of the labels
  • text_column: column name of the texts
  • max_rows: the maximum number of rows to load from the dataset. (I mainly use this for testing to go faster)
  • chunksize: size of the chunks when loading the data using pandas. default to 500000
  • encoding: default to utf-8
  • steps: text preprocessing steps to include on the text like hashtag or url removal
  • group_labels: whether or not to group labels. Default to None.
  • use_sampler: whether or not to use a weighted sampler to overcome class imbalance
  • alphabet: default to "abcdefghijklmnopqrstuvwxyz 0123456789,;.!?:'"/\|_@#$%^&*~`+-=<>()[]{}" (normally you should not modify it)
  • number_of_characters: default 70
  • extra_characters: additional characters that you'd add to the alphabet. For example uppercase letters or accented characters
  • max_length: the maximum length to fix for all the documents. default to 150 but should be adapted to your data
  • epochs: number of epochs
  • batch_size: batch size, default to 128.
  • optimizer: adam or sgd, default to sgd
  • learning_rate: default to 0.01
  • class_weights: whether or not to use class weights in the cross entropy loss
  • focal_loss: whether or not to use the focal loss
  • gamma: gamma parameter of the focal loss. default to 2
  • alpha: alpha parameter of the focal loss. default to 0.25
  • schedule: number of epochs by which the learning rate decreases by half (learning rate scheduling works only for sgd), default to 3. set it to 0 to disable it
  • patience: maximum number of epochs to wait without improvement of the validation loss, default to 3
  • early_stopping: to choose whether or not to early stop the training. default to 0. set to 1 to enable it.
  • checkpoint: to choose to save the model on disk or not. default to 1, set to 0 to disable model checkpoint
  • workers: number of workers in PyTorch DataLoader, default to 1
  • log_path: path of tensorboard log file
  • output: path of the folder where models are saved
  • model_name: prefix name of saved models

Here's an example:

python train.py --data_path=/data/tweets.csv --max_rows=200000

Prediction

Launch predict.py with the following arguments:

  • model: path of the pre-trained model
  • text: input text
  • steps: list of preprocessing steps, default to lower
  • alphabet: default to "abcdefghijklmnopqrstuvwxyz 0123456789-,;.!?:'"\/|_@#$%^&*~`+-=<>()[]{}\n"
  • number_of_characters: default to 70
  • extra_characters: additional characters that you'd add to the alphabet. For example uppercase letters or accented characters
  • max_length: the maximum length to fix for all the documents. default to 150 but should be adapted to your data

Example usage:

python predict.py ./models/pretrained_model.pth --text="I love pizza !" --max_length=150

Results

I have tested this model on a set of french labeled customer reviews (of over 3 millions rows). I reported the metrics in TensorboardX.

I got the following results

F1 score Accuracy
train 0.965 0.9366
test 0.945 0.915
training metrics

Download pretrained models

Sentiment analysis model on French customer reviews (3M documents): download link

When using it set max_length to 300 and extra_characters to "éàèùâêîôûçëïü" (accented letters)

Contributions - PR are welcome:

Here's a non-exhaustive list of potential future features to add:

  • Adapt the loss for multi-class classification
  • Log training and validation metrics for each epoch to a text file
  • Provide notebook tutorials

If you feel like adding a feature or impproving something do not hesitate to submit a pull request