Adversarial Text Classification

Code for Adversarial Training Methods for Semi-Supervised Text Classification and Semi-Supervised Sequence Learning.

Requirements

TensorFlow >= v1.3

End-to-end IMDB Sentiment Classification

Fetch data

$ wget http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz \
    -O /tmp/imdb.tar.gz
$ tar -xf /tmp/imdb.tar.gz -C /tmp

The directory /tmp/aclImdb contains the raw IMDB data.

Generate vocabulary

$ IMDB_DATA_DIR=/tmp/imdb
$ python gen_vocab.py \
    --output_dir=$IMDB_DATA_DIR \
    --dataset=imdb \
    --imdb_input_dir=/tmp/aclImdb \
    --lowercase=False

Vocabulary and frequency files will be generated in $IMDB_DATA_DIR.

Generate training, validation, and test data

$ python gen_data.py \
    --output_dir=$IMDB_DATA_DIR \
    --dataset=imdb \
    --imdb_input_dir=/tmp/aclImdb \
    --lowercase=False \
    --label_gain=False

$IMDB_DATA_DIR contains TFRecords files.

Pretrain IMDB Language Model

$ PRETRAIN_DIR=/tmp/models/imdb_pretrain
$ python pretrain.py \
    --train_dir=$PRETRAIN_DIR \
    --data_dir=$IMDB_DATA_DIR \
    --vocab_size=86934 \
    --embedding_dims=256 \
    --rnn_cell_size=1024 \
    --num_candidate_samples=1024 \
    --batch_size=256 \
    --learning_rate=0.001 \
    --learning_rate_decay_factor=0.9999 \
    --max_steps=100000 \
    --max_grad_norm=1.0 \
    --num_timesteps=400 \
    --keep_prob_emb=0.5 \
    --normalize_embeddings

$PRETRAIN_DIR contains checkpoints of the pretrained language model.

Train classifier

Most flags stay the same, save for the removal of candidate sampling and the addition of pretrained_model_dir, from which the classifier will load the pretrained embedding and LSTM variables, and flags related to adversarial training and classification.

$ TRAIN_DIR=/tmp/models/imdb_classify
$ python train_classifier.py \
    --train_dir=$TRAIN_DIR \
    --pretrained_model_dir=$PRETRAIN_DIR \
    --data_dir=$IMDB_DATA_DIR \
    --vocab_size=86934 \
    --embedding_dims=256 \
    --rnn_cell_size=1024 \
    --cl_num_layers=1 \
    --cl_hidden_size=30 \
    --batch_size=64 \
    --learning_rate=0.0005 \
    --learning_rate_decay_factor=0.9998 \
    --max_steps=15000 \
    --max_grad_norm=1.0 \
    --num_timesteps=400 \
    --keep_prob_emb=0.5 \
    --normalize_embeddings \
    --adv_training_method=vat \
    --perturb_norm_length=5.0

Evaluate on test data

$ EVAL_DIR=/tmp/models/imdb_eval
$ python evaluate.py \
    --eval_dir=$EVAL_DIR \
    --checkpoint_dir=$TRAIN_DIR \
    --eval_data=test \
    --run_once \
    --num_examples=25000 \
    --data_dir=$IMDB_DATA_DIR \
    --vocab_size=86934 \
    --embedding_dims=256 \
    --rnn_cell_size=1024 \
    --batch_size=256 \
    --num_timesteps=400 \
    --normalize_embeddings

Code Overview

The main entry points are the binaries listed below. Each training binary builds a VatxtModel, defined in graphs.py, which in turn uses graph building blocks defined in inputs.py (defines input data reading and parsing), layers.py (defines core model components), and adversarial_losses.py (defines adversarial training losses). The training loop itself is defined in train_utils.py.

Binaries

Pretraining: pretrain.py
Classifier Training: train_classifier.py
Evaluation: evaluate.py

Command-Line Flags

Flags related to distributed training and the training loop itself are defined in train_utils.py.

Flags related to model hyperparameters are defined in graphs.py.

Flags related to adversarial training are defined in adversarial_losses.py.

Flags particular to each job are defined in the main binary files.

Data Generation

Vocabulary generation: gen_vocab.py
Data generation: gen_data.py

Command-line flags defined in document_generators.py control which dataset is processed and how.

Contact for Issues

Ryan Sepassi, @rsepassi
Andrew M. Dai, @a-dai adai@google.com
Takeru Miyato, @takerum (Original implementation)

drinky78/adversarial_text