spaCy/examples/training/pretrain_textcat.py

"""This script is experimental.

Try pre-training the CNN component of the text categorizer using a cheap
language modelling-like objective. Specifically, we load pretrained vectors
(from something like word2vec, GloVe, FastText etc), and use the CNN to
predict the tokens' pretrained vectors. This isn't as easy as it sounds:
we're not merely doing compression here, because heavy dropout is applied,
including over the input words. This means the model must often (50% of the time)
use the context in order to predict the word.

To evaluate the technique, we're pre-training with the 50k texts from the IMDB
corpus, and then training with only 100 labels. Note that it's a bit dirty to
pre-train with the development data, but also not *so* terrible: we're not using
the development labels, after all --- only the unlabelled text.
"""
import plac
import random
import spacy
import thinc.extra.datasets
from spacy.util import minibatch, use_gpu, compounding
from spacy._ml import Tok2Vec
from spacy.pipeline import TextCategorizer
import numpy


def load_texts(limit=0):
    train, dev = thinc.extra.datasets.imdb()
    train_texts, train_labels = zip(*train)
    dev_texts, dev_labels = zip(*train)
    train_texts = list(train_texts)
    dev_texts = list(dev_texts)
    random.shuffle(train_texts)
    random.shuffle(dev_texts)
    if limit >= 1:
        return train_texts[:limit]
    else:
        return list(train_texts) + list(dev_texts)


def load_textcat_data(limit=0):
    """Load data from the IMDB dataset."""
    # Partition off part of the train data for evaluation
    train_data, eval_data = thinc.extra.datasets.imdb()
    random.shuffle(train_data)
    train_data = train_data[-limit:]
    texts, labels = zip(*train_data)
    eval_texts, eval_labels = zip(*eval_data)
    cats = [{"POSITIVE": bool(y), "NEGATIVE": not bool(y)} for y in labels]
    eval_cats = [{"POSITIVE": bool(y), "NEGATIVE": not bool(y)} for y in eval_labels]
    return (texts, cats), (eval_texts, eval_cats)


def prefer_gpu():
    used = spacy.util.use_gpu(0)
    if used is None:
        return False
    else:
        import cupy.random

        cupy.random.seed(0)
        return True


def build_textcat_model(tok2vec, nr_class, width):
    from thinc.v2v import Model, Softmax, Maxout
    from thinc.api import flatten_add_lengths, chain
    from thinc.t2v import Pooling, sum_pool, mean_pool, max_pool
    from thinc.misc import Residual, LayerNorm
    from spacy._ml import logistic, zero_init

    with Model.define_operators({">>": chain}):
        model = (
            tok2vec
            >> flatten_add_lengths
            >> Pooling(mean_pool)
            >> Softmax(nr_class, width)
        )
    model.tok2vec = tok2vec
    return model


def block_gradients(model):
    from thinc.api import wrap

    def forward(X, drop=0.0):
        Y, _ = model.begin_update(X, drop=drop)
        return Y, None

    return wrap(forward, model)


def create_pipeline(width, embed_size, vectors_model):
    print("Load vectors")
    nlp = spacy.load(vectors_model)
    print("Start training")
    textcat = TextCategorizer(
        nlp.vocab,
        labels=["POSITIVE", "NEGATIVE"],
        model=build_textcat_model(
            Tok2Vec(width=width, embed_size=embed_size), 2, width
        ),
    )

    nlp.add_pipe(textcat)
    return nlp


def train_tensorizer(nlp, texts, dropout, n_iter):
    # temp fix to avoid import issues cf https://github.com/explosion/spaCy/issues/4200
    import tqdm

    tensorizer = nlp.create_pipe("tensorizer")
    nlp.add_pipe(tensorizer)
    optimizer = nlp.begin_training()
    for i in range(n_iter):
        losses = {}
        for i, batch in enumerate(minibatch(tqdm.tqdm(texts))):
            docs = [nlp.make_doc(text) for text in batch]
            tensorizer.update((docs, None), losses=losses, sgd=optimizer, drop=dropout)
        print(losses)
    return optimizer


def train_textcat(nlp, n_texts, n_iter=10):
    # temp fix to avoid import issues cf https://github.com/explosion/spaCy/issues/4200
    import tqdm

    textcat = nlp.get_pipe("textcat")
    tok2vec_weights = textcat.model.tok2vec.to_bytes()
    (train_texts, train_cats), (dev_texts, dev_cats) = load_textcat_data(limit=n_texts)
    print(
        "Using {} examples ({} training, {} evaluation)".format(
            n_texts, len(train_texts), len(dev_texts)
        )
    )
    train_data = list(zip(train_texts, [{"cats": cats} for cats in train_cats]))

    # get names of other pipes to disable them during training
    other_pipes = [pipe for pipe in nlp.pipe_names if pipe != "textcat"]
    with nlp.disable_pipes(*other_pipes):  # only train textcat
        optimizer = nlp.begin_training()
        textcat.model.tok2vec.from_bytes(tok2vec_weights)
        print("Training the model...")
        print("{:^5}\t{:^5}\t{:^5}\t{:^5}".format("LOSS", "P", "R", "F"))
        for i in range(n_iter):
            losses = {"textcat": 0.0}
            # batch up the examples using spaCy's minibatch
            batches = minibatch(tqdm.tqdm(train_data), size=2)
            for batch in batches:
                nlp.update(batch, sgd=optimizer, drop=0.2, losses=losses)
            with textcat.model.use_params(optimizer.averages):
                # evaluate on the dev data split off in load_data()
                scores = evaluate_textcat(nlp.tokenizer, textcat, dev_texts, dev_cats)
            print(
                "{0:.3f}\t{1:.3f}\t{2:.3f}\t{3:.3f}".format(  # print a simple table
                    losses["textcat"],
                    scores["textcat_p"],
                    scores["textcat_r"],
                    scores["textcat_f"],
                )
            )


def evaluate_textcat(tokenizer, textcat, texts, cats):
    docs = (tokenizer(text) for text in texts)
    tp = 1e-8
    fp = 1e-8
    tn = 1e-8
    fn = 1e-8
    for i, doc in enumerate(textcat.pipe(docs)):
        gold = cats[i]
        for label, score in doc.cats.items():
            if label not in gold:
                continue
            if score >= 0.5 and gold[label] >= 0.5:
                tp += 1.0
            elif score >= 0.5 and gold[label] < 0.5:
                fp += 1.0
            elif score < 0.5 and gold[label] < 0.5:
                tn += 1
            elif score < 0.5 and gold[label] >= 0.5:
                fn += 1
    precision = tp / (tp + fp)
    recall = tp / (tp + fn)
    f_score = 2 * (precision * recall) / (precision + recall)
    return {"textcat_p": precision, "textcat_r": recall, "textcat_f": f_score}


@plac.annotations(
    width=("Width of CNN layers", "positional", None, int),
    embed_size=("Embedding rows", "positional", None, int),
    pretrain_iters=("Number of iterations to pretrain", "option", "pn", int),
    train_iters=("Number of iterations to pretrain", "option", "tn", int),
    train_examples=("Number of labelled examples", "option", "eg", int),
    vectors_model=("Name or path to vectors model to learn from"),
)
def main(
    width,
    embed_size,
    vectors_model,
    pretrain_iters=30,
    train_iters=30,
    train_examples=1000,
):
    random.seed(0)
    numpy.random.seed(0)
    use_gpu = prefer_gpu()
    print("Using GPU?", use_gpu)

    nlp = create_pipeline(width, embed_size, vectors_model)
    print("Load data")
    texts = load_texts(limit=0)
    print("Train tensorizer")
    optimizer = train_tensorizer(nlp, texts, dropout=0.2, n_iter=pretrain_iters)
    print("Train textcat")
    train_textcat(nlp, train_examples, n_iter=train_iters)


if __name__ == "__main__":
    plac.call(main)