spaCy/spacy/cli/train.py
adrianeboyd 90c52128dc Improve train CLI with base model (#4911)
Improve train CLI with a provided base model so that you can:

* add a new component
* extend an existing component
* replace an existing component

When the final model and best model are saved, reenable any disabled
components and merge the meta information to include the full pipeline
and accuracy information for all components in the base model plus the
newly added components if needed.
2020-01-16 01:58:51 +01:00

660 lines
28 KiB
Python

# coding: utf8
from __future__ import unicode_literals, division, print_function
import plac
import os
import tqdm
from pathlib import Path
from thinc.neural._classes.model import Model
from timeit import default_timer as timer
import shutil
import srsly
from wasabi import msg
import contextlib
import random
from .._ml import create_default_optimizer
from ..attrs import PROB, IS_OOV, CLUSTER, LANG
from ..gold import GoldCorpus
from ..compat import path2str
from .. import util
from .. import about
@plac.annotations(
# fmt: off
lang=("Model language", "positional", None, str),
output_path=("Output directory to store model in", "positional", None, Path),
train_path=("Location of JSON-formatted training data", "positional", None, Path),
dev_path=("Location of JSON-formatted development data", "positional", None, Path),
raw_text=("Path to jsonl file with unlabelled text documents.", "option", "rt", Path),
base_model=("Name of model to update (optional)", "option", "b", str),
pipeline=("Comma-separated names of pipeline components", "option", "p", str),
replace_components=("Replace components from base model", "flag", "R", bool),
vectors=("Model to load vectors from", "option", "v", str),
n_iter=("Number of iterations", "option", "n", int),
n_early_stopping=("Maximum number of training epochs without dev accuracy improvement", "option", "ne", int),
n_examples=("Number of examples", "option", "ns", int),
use_gpu=("Use GPU", "option", "g", int),
version=("Model version", "option", "V", str),
meta_path=("Optional path to meta.json to use as base.", "option", "m", Path),
init_tok2vec=("Path to pretrained weights for the token-to-vector parts of the models. See 'spacy pretrain'. Experimental.", "option", "t2v", Path),
parser_multitasks=("Side objectives for parser CNN, e.g. 'dep' or 'dep,tag'", "option", "pt", str),
entity_multitasks=("Side objectives for NER CNN, e.g. 'dep' or 'dep,tag'", "option", "et", str),
noise_level=("Amount of corruption for data augmentation", "option", "nl", float),
orth_variant_level=("Amount of orthography variation for data augmentation", "option", "ovl", float),
eval_beam_widths=("Beam widths to evaluate, e.g. 4,8", "option", "bw", str),
gold_preproc=("Use gold preprocessing", "flag", "G", bool),
learn_tokens=("Make parser learn gold-standard tokenization", "flag", "T", bool),
textcat_multilabel=("Textcat classes aren't mutually exclusive (multilabel)", "flag", "TML", bool),
textcat_arch=("Textcat model architecture", "option", "ta", str),
textcat_positive_label=("Textcat positive label for binary classes with two labels", "option", "tpl", str),
verbose=("Display more information for debug", "flag", "VV", bool),
debug=("Run data diagnostics before training", "flag", "D", bool),
# fmt: on
)
def train(
lang,
output_path,
train_path,
dev_path,
raw_text=None,
base_model=None,
pipeline="tagger,parser,ner",
replace_components=False,
vectors=None,
n_iter=30,
n_early_stopping=None,
n_examples=0,
use_gpu=-1,
version="0.0.0",
meta_path=None,
init_tok2vec=None,
parser_multitasks="",
entity_multitasks="",
noise_level=0.0,
orth_variant_level=0.0,
eval_beam_widths="",
gold_preproc=False,
learn_tokens=False,
textcat_multilabel=False,
textcat_arch="bow",
textcat_positive_label=None,
verbose=False,
debug=False,
):
"""
Train or update a spaCy model. Requires data to be formatted in spaCy's
JSON format. To convert data from other formats, use the `spacy convert`
command.
"""
util.fix_random_seed()
util.set_env_log(verbose)
# Make sure all files and paths exists if they are needed
train_path = util.ensure_path(train_path)
dev_path = util.ensure_path(dev_path)
meta_path = util.ensure_path(meta_path)
output_path = util.ensure_path(output_path)
if raw_text is not None:
raw_text = list(srsly.read_jsonl(raw_text))
if not train_path or not train_path.exists():
msg.fail("Training data not found", train_path, exits=1)
if not dev_path or not dev_path.exists():
msg.fail("Development data not found", dev_path, exits=1)
if meta_path is not None and not meta_path.exists():
msg.fail("Can't find model meta.json", meta_path, exits=1)
meta = srsly.read_json(meta_path) if meta_path else {}
if output_path.exists() and [p for p in output_path.iterdir() if p.is_dir()]:
msg.warn(
"Output directory is not empty",
"This can lead to unintended side effects when saving the model. "
"Please use an empty directory or a different path instead. If "
"the specified output path doesn't exist, the directory will be "
"created for you.",
)
if not output_path.exists():
output_path.mkdir()
# Take dropout and batch size as generators of values -- dropout
# starts high and decays sharply, to force the optimizer to explore.
# Batch size starts at 1 and grows, so that we make updates quickly
# at the beginning of training.
dropout_rates = util.decaying(
util.env_opt("dropout_from", 0.2),
util.env_opt("dropout_to", 0.2),
util.env_opt("dropout_decay", 0.0),
)
batch_sizes = util.compounding(
util.env_opt("batch_from", 100.0),
util.env_opt("batch_to", 1000.0),
util.env_opt("batch_compound", 1.001),
)
if not eval_beam_widths:
eval_beam_widths = [1]
else:
eval_beam_widths = [int(bw) for bw in eval_beam_widths.split(",")]
if 1 not in eval_beam_widths:
eval_beam_widths.append(1)
eval_beam_widths.sort()
has_beam_widths = eval_beam_widths != [1]
# Set up the base model and pipeline. If a base model is specified, load
# the model and make sure the pipeline matches the pipeline setting. If
# training starts from a blank model, intitalize the language class.
pipeline = [p.strip() for p in pipeline.split(",")]
disabled_pipes = None
pipes_added = False
msg.text("Training pipeline: {}".format(pipeline))
if base_model:
msg.text("Starting with base model '{}'".format(base_model))
nlp = util.load_model(base_model)
if nlp.lang != lang:
msg.fail(
"Model language ('{}') doesn't match language specified as "
"`lang` argument ('{}') ".format(nlp.lang, lang),
exits=1,
)
for pipe in pipeline:
pipe_cfg = {}
if pipe == "parser":
pipe_cfg = {"learn_tokens": learn_tokens}
elif pipe == "textcat":
pipe_cfg = {
"exclusive_classes": not textcat_multilabel,
"architecture": textcat_arch,
"positive_label": textcat_positive_label,
}
if pipe not in nlp.pipe_names:
msg.text("Adding component to base model '{}'".format(pipe))
nlp.add_pipe(nlp.create_pipe(pipe, config=pipe_cfg))
pipes_added = True
elif replace_components:
msg.text("Replacing component from base model '{}'".format(pipe))
nlp.replace_pipe(pipe, nlp.create_pipe(pipe, config=pipe_cfg))
pipes_added = True
else:
if pipe == "textcat":
textcat_cfg = nlp.get_pipe("textcat").cfg
base_cfg = {
"exclusive_classes": textcat_cfg["exclusive_classes"],
"architecture": textcat_cfg["architecture"],
"positive_label": textcat_cfg["positive_label"],
}
if base_cfg != pipe_cfg:
msg.fail(
"The base textcat model configuration does"
"not match the provided training options. "
"Existing cfg: {}, provided cfg: {}".format(
base_cfg, pipe_cfg
),
exits=1,
)
msg.text("Extending component from base model '{}'".format(pipe))
disabled_pipes = nlp.disable_pipes([p for p in nlp.pipe_names if p not in pipeline])
else:
msg.text("Starting with blank model '{}'".format(lang))
lang_cls = util.get_lang_class(lang)
nlp = lang_cls()
for pipe in pipeline:
if pipe == "parser":
pipe_cfg = {"learn_tokens": learn_tokens}
elif pipe == "textcat":
pipe_cfg = {
"exclusive_classes": not textcat_multilabel,
"architecture": textcat_arch,
"positive_label": textcat_positive_label,
}
else:
pipe_cfg = {}
nlp.add_pipe(nlp.create_pipe(pipe, config=pipe_cfg))
if vectors:
msg.text("Loading vector from model '{}'".format(vectors))
_load_vectors(nlp, vectors)
# Multitask objectives
multitask_options = [("parser", parser_multitasks), ("ner", entity_multitasks)]
for pipe_name, multitasks in multitask_options:
if multitasks:
if pipe_name not in pipeline:
msg.fail(
"Can't use multitask objective without '{}' in the "
"pipeline".format(pipe_name)
)
pipe = nlp.get_pipe(pipe_name)
for objective in multitasks.split(","):
pipe.add_multitask_objective(objective)
# Prepare training corpus
msg.text("Counting training words (limit={})".format(n_examples))
corpus = GoldCorpus(train_path, dev_path, limit=n_examples)
n_train_words = corpus.count_train()
if base_model and not pipes_added:
# Start with an existing model, use default optimizer
optimizer = create_default_optimizer(Model.ops)
else:
# Start with a blank model, call begin_training
optimizer = nlp.begin_training(lambda: corpus.train_tuples, device=use_gpu)
nlp._optimizer = None
# Load in pretrained weights
if init_tok2vec is not None:
components = _load_pretrained_tok2vec(nlp, init_tok2vec)
msg.text("Loaded pretrained tok2vec for: {}".format(components))
# Verify textcat config
if "textcat" in pipeline:
textcat_labels = nlp.get_pipe("textcat").cfg.get("labels", [])
if textcat_positive_label and textcat_positive_label not in textcat_labels:
msg.fail(
"The textcat_positive_label (tpl) '{}' does not match any "
"label in the training data.".format(textcat_positive_label),
exits=1,
)
if textcat_positive_label and len(textcat_labels) != 2:
msg.fail(
"A textcat_positive_label (tpl) '{}' was provided for training "
"data that does not appear to be a binary classification "
"problem with two labels.".format(textcat_positive_label),
exits=1,
)
train_docs = corpus.train_docs(
nlp,
noise_level=noise_level,
gold_preproc=gold_preproc,
max_length=0,
ignore_misaligned=True,
)
train_labels = set()
if textcat_multilabel:
multilabel_found = False
for text, gold in train_docs:
train_labels.update(gold.cats.keys())
if list(gold.cats.values()).count(1.0) != 1:
multilabel_found = True
if not multilabel_found and not base_model:
msg.warn(
"The textcat training instances look like they have "
"mutually-exclusive classes. Remove the flag "
"'--textcat-multilabel' to train a classifier with "
"mutually-exclusive classes."
)
if not textcat_multilabel:
for text, gold in train_docs:
train_labels.update(gold.cats.keys())
if list(gold.cats.values()).count(1.0) != 1 and not base_model:
msg.warn(
"Some textcat training instances do not have exactly "
"one positive label. Modifying training options to "
"include the flag '--textcat-multilabel' for classes "
"that are not mutually exclusive."
)
nlp.get_pipe("textcat").cfg["exclusive_classes"] = False
textcat_multilabel = True
break
if base_model and set(textcat_labels) != train_labels:
msg.fail(
"Cannot extend textcat model using data with different "
"labels. Base model labels: {}, training data labels: "
"{}.".format(textcat_labels, list(train_labels)),
exits=1,
)
if textcat_multilabel:
msg.text(
"Textcat evaluation score: ROC AUC score macro-averaged across "
"the labels '{}'".format(", ".join(textcat_labels))
)
elif textcat_positive_label and len(textcat_labels) == 2:
msg.text(
"Textcat evaluation score: F1-score for the "
"label '{}'".format(textcat_positive_label)
)
elif len(textcat_labels) > 1:
if len(textcat_labels) == 2:
msg.warn(
"If the textcat component is a binary classifier with "
"exclusive classes, provide '--textcat_positive_label' for "
"an evaluation on the positive class."
)
msg.text(
"Textcat evaluation score: F1-score macro-averaged across "
"the labels '{}'".format(", ".join(textcat_labels))
)
else:
msg.fail(
"Unsupported textcat configuration. Use `spacy debug-data` "
"for more information."
)
# fmt: off
row_head, output_stats = _configure_training_output(pipeline, use_gpu, has_beam_widths)
row_widths = [len(w) for w in row_head]
row_settings = {"widths": row_widths, "aligns": tuple(["r" for i in row_head]), "spacing": 2}
# fmt: on
print("")
msg.row(row_head, **row_settings)
msg.row(["-" * width for width in row_settings["widths"]], **row_settings)
try:
iter_since_best = 0
best_score = 0.0
for i in range(n_iter):
train_docs = corpus.train_docs(
nlp,
noise_level=noise_level,
orth_variant_level=orth_variant_level,
gold_preproc=gold_preproc,
max_length=0,
ignore_misaligned=True,
)
if raw_text:
random.shuffle(raw_text)
raw_batches = util.minibatch(
(nlp.make_doc(rt["text"]) for rt in raw_text), size=8
)
words_seen = 0
with tqdm.tqdm(total=n_train_words, leave=False) as pbar:
losses = {}
for batch in util.minibatch_by_words(train_docs, size=batch_sizes):
if not batch:
continue
docs, golds = zip(*batch)
nlp.update(
docs,
golds,
sgd=optimizer,
drop=next(dropout_rates),
losses=losses,
)
if raw_text:
# If raw text is available, perform 'rehearsal' updates,
# which use unlabelled data to reduce overfitting.
raw_batch = list(next(raw_batches))
nlp.rehearse(raw_batch, sgd=optimizer, losses=losses)
if not int(os.environ.get("LOG_FRIENDLY", 0)):
pbar.update(sum(len(doc) for doc in docs))
words_seen += sum(len(doc) for doc in docs)
with nlp.use_params(optimizer.averages):
util.set_env_log(False)
epoch_model_path = output_path / ("model%d" % i)
nlp.to_disk(epoch_model_path)
nlp_loaded = util.load_model_from_path(epoch_model_path)
for beam_width in eval_beam_widths:
for name, component in nlp_loaded.pipeline:
if hasattr(component, "cfg"):
component.cfg["beam_width"] = beam_width
dev_docs = list(
corpus.dev_docs(
nlp_loaded,
gold_preproc=gold_preproc,
ignore_misaligned=True,
)
)
nwords = sum(len(doc_gold[0]) for doc_gold in dev_docs)
start_time = timer()
scorer = nlp_loaded.evaluate(dev_docs, verbose=verbose)
end_time = timer()
if use_gpu < 0:
gpu_wps = None
cpu_wps = nwords / (end_time - start_time)
else:
gpu_wps = nwords / (end_time - start_time)
with Model.use_device("cpu"):
nlp_loaded = util.load_model_from_path(epoch_model_path)
for name, component in nlp_loaded.pipeline:
if hasattr(component, "cfg"):
component.cfg["beam_width"] = beam_width
dev_docs = list(
corpus.dev_docs(
nlp_loaded,
gold_preproc=gold_preproc,
ignore_misaligned=True,
)
)
start_time = timer()
scorer = nlp_loaded.evaluate(dev_docs, verbose=verbose)
end_time = timer()
cpu_wps = nwords / (end_time - start_time)
acc_loc = output_path / ("model%d" % i) / "accuracy.json"
srsly.write_json(acc_loc, scorer.scores)
# Update model meta.json
meta["lang"] = nlp.lang
meta["pipeline"] = nlp.pipe_names
meta["spacy_version"] = ">=%s" % about.__version__
if beam_width == 1:
meta["speed"] = {
"nwords": nwords,
"cpu": cpu_wps,
"gpu": gpu_wps,
}
meta.setdefault("accuracy", {})
for component in nlp.pipe_names:
for metric in _get_metrics(component):
meta["accuracy"][metric] = scorer.scores[metric]
else:
meta.setdefault("beam_accuracy", {})
meta.setdefault("beam_speed", {})
for component in nlp.pipe_names:
for metric in _get_metrics(component):
meta["beam_accuracy"][metric] = scorer.scores[metric]
meta["beam_speed"][beam_width] = {
"nwords": nwords,
"cpu": cpu_wps,
"gpu": gpu_wps,
}
meta["vectors"] = {
"width": nlp.vocab.vectors_length,
"vectors": len(nlp.vocab.vectors),
"keys": nlp.vocab.vectors.n_keys,
"name": nlp.vocab.vectors.name,
}
meta.setdefault("name", "model%d" % i)
meta.setdefault("version", version)
meta["labels"] = nlp.meta["labels"]
meta_loc = output_path / ("model%d" % i) / "meta.json"
srsly.write_json(meta_loc, meta)
util.set_env_log(verbose)
progress = _get_progress(
i,
losses,
scorer.scores,
output_stats,
beam_width=beam_width if has_beam_widths else None,
cpu_wps=cpu_wps,
gpu_wps=gpu_wps,
)
if i == 0 and "textcat" in pipeline:
textcats_per_cat = scorer.scores.get("textcats_per_cat", {})
for cat, cat_score in textcats_per_cat.items():
if cat_score.get("roc_auc_score", 0) < 0:
msg.warn(
"Textcat ROC AUC score is undefined due to "
"only one value in label '{}'.".format(cat)
)
msg.row(progress, **row_settings)
# Early stopping
if n_early_stopping is not None:
current_score = _score_for_model(meta)
if current_score < best_score:
iter_since_best += 1
else:
iter_since_best = 0
best_score = current_score
if iter_since_best >= n_early_stopping:
msg.text(
"Early stopping, best iteration "
"is: {}".format(i - iter_since_best)
)
msg.text(
"Best score = {}; Final iteration "
"score = {}".format(best_score, current_score)
)
break
finally:
best_pipes = nlp.pipe_names
if disabled_pipes:
disabled_pipes.restore()
with nlp.use_params(optimizer.averages):
final_model_path = output_path / "model-final"
nlp.to_disk(final_model_path)
final_meta = srsly.read_json(output_path / "model-final" / "meta.json")
msg.good("Saved model to output directory", final_model_path)
with msg.loading("Creating best model..."):
best_model_path = _collate_best_model(final_meta, output_path, best_pipes)
msg.good("Created best model", best_model_path)
def _score_for_model(meta):
""" Returns mean score between tasks in pipeline that can be used for early stopping. """
mean_acc = list()
pipes = meta["pipeline"]
acc = meta["accuracy"]
if "tagger" in pipes:
mean_acc.append(acc["tags_acc"])
if "parser" in pipes:
mean_acc.append((acc["uas"] + acc["las"]) / 2)
if "ner" in pipes:
mean_acc.append((acc["ents_p"] + acc["ents_r"] + acc["ents_f"]) / 3)
if "textcat" in pipes:
mean_acc.append(acc["textcat_score"])
return sum(mean_acc) / len(mean_acc)
@contextlib.contextmanager
def _create_progress_bar(total):
if int(os.environ.get("LOG_FRIENDLY", 0)):
yield
else:
pbar = tqdm.tqdm(total=total, leave=False)
yield pbar
def _load_vectors(nlp, vectors):
util.load_model(vectors, vocab=nlp.vocab)
for lex in nlp.vocab:
values = {}
for attr, func in nlp.vocab.lex_attr_getters.items():
# These attrs are expected to be set by data. Others should
# be set by calling the language functions.
if attr not in (CLUSTER, PROB, IS_OOV, LANG):
values[lex.vocab.strings[attr]] = func(lex.orth_)
lex.set_attrs(**values)
lex.is_oov = False
def _load_pretrained_tok2vec(nlp, loc):
"""Load pretrained weights for the 'token-to-vector' part of the component
models, which is typically a CNN. See 'spacy pretrain'. Experimental.
"""
with loc.open("rb") as file_:
weights_data = file_.read()
loaded = []
for name, component in nlp.pipeline:
if hasattr(component, "model") and hasattr(component.model, "tok2vec"):
component.tok2vec.from_bytes(weights_data)
loaded.append(name)
return loaded
def _collate_best_model(meta, output_path, components):
bests = {}
meta.setdefault("accuracy", {})
for component in components:
bests[component] = _find_best(output_path, component)
best_dest = output_path / "model-best"
shutil.copytree(path2str(output_path / "model-final"), path2str(best_dest))
for component, best_component_src in bests.items():
shutil.rmtree(path2str(best_dest / component))
shutil.copytree(
path2str(best_component_src / component), path2str(best_dest / component)
)
accs = srsly.read_json(best_component_src / "accuracy.json")
for metric in _get_metrics(component):
meta["accuracy"][metric] = accs[metric]
srsly.write_json(best_dest / "meta.json", meta)
return best_dest
def _find_best(experiment_dir, component):
accuracies = []
for epoch_model in experiment_dir.iterdir():
if epoch_model.is_dir() and epoch_model.parts[-1] != "model-final":
accs = srsly.read_json(epoch_model / "accuracy.json")
scores = [accs.get(metric, 0.0) for metric in _get_metrics(component)]
accuracies.append((scores, epoch_model))
if accuracies:
return max(accuracies)[1]
else:
return None
def _get_metrics(component):
if component == "parser":
return ("las", "uas", "las_per_type", "token_acc")
elif component == "tagger":
return ("tags_acc",)
elif component == "ner":
return ("ents_f", "ents_p", "ents_r", "ents_per_type")
elif component == "textcat":
return ("textcat_score",)
return ("token_acc",)
def _configure_training_output(pipeline, use_gpu, has_beam_widths):
row_head = ["Itn"]
output_stats = []
for pipe in pipeline:
if pipe == "tagger":
row_head.extend(["Tag Loss ", " Tag % "])
output_stats.extend(["tag_loss", "tags_acc"])
elif pipe == "parser":
row_head.extend(["Dep Loss ", " UAS ", " LAS "])
output_stats.extend(["dep_loss", "uas", "las"])
elif pipe == "ner":
row_head.extend(["NER Loss ", "NER P ", "NER R ", "NER F "])
output_stats.extend(["ner_loss", "ents_p", "ents_r", "ents_f"])
elif pipe == "textcat":
row_head.extend(["Textcat Loss", "Textcat"])
output_stats.extend(["textcat_loss", "textcat_score"])
row_head.extend(["Token %", "CPU WPS"])
output_stats.extend(["token_acc", "cpu_wps"])
if use_gpu >= 0:
row_head.extend(["GPU WPS"])
output_stats.extend(["gpu_wps"])
if has_beam_widths:
row_head.insert(1, "Beam W.")
return row_head, output_stats
def _get_progress(
itn, losses, dev_scores, output_stats, beam_width=None, cpu_wps=0.0, gpu_wps=0.0
):
scores = {}
for stat in output_stats:
scores[stat] = 0.0
scores["dep_loss"] = losses.get("parser", 0.0)
scores["ner_loss"] = losses.get("ner", 0.0)
scores["tag_loss"] = losses.get("tagger", 0.0)
scores["textcat_loss"] = losses.get("textcat", 0.0)
scores["cpu_wps"] = cpu_wps
scores["gpu_wps"] = gpu_wps or 0.0
scores.update(dev_scores)
formatted_scores = []
for stat in output_stats:
format_spec = "{:.3f}"
if stat.endswith("_wps"):
format_spec = "{:.0f}"
formatted_scores.append(format_spec.format(scores[stat]))
result = [itn + 1]
result.extend(formatted_scores)
if beam_width is not None:
result.insert(1, beam_width)
return result