From b8c83784dfd541dbae8851a45c66a204d5cd82a7 Mon Sep 17 00:00:00 2001
From: Mateusz Klimaszewski <mk.klimaszewski@gmail.com>
Date: Tue, 5 Jan 2021 16:37:59 +0100
Subject: [PATCH] Extend installation and pre-trained models documentation, fix
installation, update trainer to be in sync with AllenNLP 1.2.1.
---
combo/data/api.py | 3 ++
combo/training/trainer.py | 38 ++++++++------
docs/installation.md | 18 ++++++-
docs/models.md | 2 +
scripts/train.py | 108 ++++++++++++++++++++++++++++++++++++--
setup.py | 9 ++--
6 files changed, 151 insertions(+), 27 deletions(-)
diff --git a/combo/data/api.py b/combo/data/api.py
index 7d44917..4ab7f1a 100644
--- a/combo/data/api.py
+++ b/combo/data/api.py
@@ -36,6 +36,9 @@ class Sentence:
"metadata": self.metadata,
})
+ def __len__(self):
+ return len(self.tokens)
+
class _TokenList(conllu.TokenList):
diff --git a/combo/training/trainer.py b/combo/training/trainer.py
index 26bd75f..f74873e 100644
--- a/combo/training/trainer.py
+++ b/combo/training/trainer.py
@@ -68,10 +68,7 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
self.validate_every_n = 5
@overrides
- def train(self) -> Dict[str, Any]:
- """
- Trains the supplied model with the supplied parameters.
- """
+ def _try_train(self) -> Dict[str, Any]:
try:
epoch_counter = self._restore_checkpoint()
except RuntimeError:
@@ -87,7 +84,7 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
logger.info("Beginning training.")
val_metrics: Dict[str, float] = {}
- this_epoch_val_metric: float = None
+ this_epoch_val_metric: float
metrics: Dict[str, Any] = {}
epochs_trained = 0
training_start_time = time.time()
@@ -97,19 +94,24 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
metrics["best_validation_" + key] = value
for callback in self._epoch_callbacks:
- callback(self, metrics={}, epoch=-1, is_master=True)
+ callback(self, metrics={}, epoch=-1, is_master=self._master)
for epoch in range(epoch_counter, self._num_epochs):
epoch_start_time = time.time()
train_metrics = self._train_epoch(epoch)
+ if self._master and self._checkpointer is not None:
+ self._checkpointer.save_checkpoint(epoch, self, save_model_only=True)
+
+ # Wait for the master to finish saving the model checkpoint
+ if self._distributed:
+ dist.barrier()
+
# get peak of memory usage
- if "cpu_memory_MB" in train_metrics:
- metrics["peak_cpu_memory_MB"] = max(
- metrics.get("peak_cpu_memory_MB", 0), train_metrics["cpu_memory_MB"]
- )
for key, value in train_metrics.items():
- if key.startswith("gpu_"):
+ if key.startswith("gpu_") and key.endswith("_memory_MB"):
+ metrics["peak_" + key] = max(metrics.get("peak_" + key, 0), value)
+ elif key.startswith("worker_") and key.endswith("_memory_MB"):
metrics["peak_" + key] = max(metrics.get("peak_" + key, 0), value)
if self._validation_data_loader is not None:
@@ -129,9 +131,9 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
self.model,
val_loss,
val_reg_loss,
- num_batches=num_batches,
batch_loss=None,
batch_reg_loss=None,
+ num_batches=num_batches,
reset=True,
world_size=self._world_size,
cuda_device=self.cuda_device,
@@ -139,7 +141,7 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
# Check validation metric for early stopping
this_epoch_val_metric = val_metrics[self._validation_metric]
- # self._metric_tracker.add_metric(this_epoch_val_metric)
+ self._metric_tracker.add_metric(this_epoch_val_metric)
train_metrics["patience"] = self._metric_tracker._patience
if self._metric_tracker.should_stop_early():
@@ -184,7 +186,7 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
if self._momentum_scheduler:
self._momentum_scheduler.step(this_epoch_val_metric)
- if self._master:
+ if self._master and self._checkpointer is not None:
self._checkpointer.save_checkpoint(
epoch, self, is_best_so_far=self._metric_tracker.is_best_so_far()
)
@@ -209,11 +211,13 @@ class GradientDescentTrainer(training.GradientDescentTrainer):
epochs_trained += 1
- # make sure pending events are flushed to disk and files are closed properly
- self._tensorboard.close()
+ for callback in self._end_callbacks:
+ callback(self, metrics=metrics, epoch=epoch, is_master=self._master)
# Load the best model state before returning
- best_model_state = self._checkpointer.best_model_state()
+ best_model_state = (
+ None if self._checkpointer is None else self._checkpointer.best_model_state()
+ )
if best_model_state:
self.model.load_state_dict(best_model_state)
diff --git a/docs/installation.md b/docs/installation.md
index 6371094..bf741f9 100644
--- a/docs/installation.md
+++ b/docs/installation.md
@@ -7,7 +7,23 @@ python setup.py develop
combo --helpfull
```
+### Virtualenv example:
+```bash
+python -m venv venv
+source venv/bin/activate
+pip install --upgrade pip
+python setup.py develop
+```
+
## Problems & solutions
* **jsonnet** installation error
-use `conda install -c conda-forge jsonnet=0.15.0`
+Run `conda install -c conda-forge jsonnet=0.15.0` and re-run installation.
+
+* No package 'sentencepiece' found
+
+Run `pip install sentencepiece` and re-run installation.
+
+* Missing Cython error
+
+Run `pip install spacy==2.3.2` manually and re-run installation.
diff --git a/docs/models.md b/docs/models.md
index 25a7f70..94eed03 100644
--- a/docs/models.md
+++ b/docs/models.md
@@ -4,6 +4,8 @@ COMBO provides pre-trained models for:
- morphosyntactic prediction (i.e. part-of-speech tagging, morphosyntactic analysis, lemmatisation and dependency parsing) trained on the treebanks from [Universal Dependencies repository](https://universaldependencies.org),
- enhanced dependency parsing trained on IWPT 2020 shared task [data](https://universaldependencies.org/iwpt20/data.html).
+Pre-trained models list with the **evaluation results** is available in the [spreadsheet](https://docs.google.com/spreadsheets/d/1WFYc2aLRa1jw7le030HOacv9fc4zmtqiZtRQY6gl5mc/edit?usp=sharing)
+Please notice that the name in the brackets matches the name used in [Automatic Download](models.md#Automatic download).
## Manual download
The pre-trained models can be downloaded from [here](http://mozart.ipipan.waw.pl/~mklimaszewski/models/).
diff --git a/scripts/train.py b/scripts/train.py
index 9390888..4bd342a 100644
--- a/scripts/train.py
+++ b/scripts/train.py
@@ -6,38 +6,71 @@ from absl import flags
from scripts import utils
+# # ls -1 | xargs -i echo "\"{}\","
+# UD 2.7
TREEBANKS = [
"UD_Afrikaans-AfriBooms",
+ "UD_Akkadian-PISANDUB",
+ "UD_Akkadian-RIAO",
+ "UD_Akuntsu-TuDeT",
+ "UD_Albanian-TSA",
+ "UD_Amharic-ATT",
+ "UD_Ancient_Greek-Perseus",
+ "UD_Ancient_Greek-PROIEL",
+ "UD_Apurina-UFPA",
"UD_Arabic-NYUAD",
"UD_Arabic-PADT",
+ "UD_Arabic-PUD",
"UD_Armenian-ArmTDP",
+ "UD_Assyrian-AS",
+ "UD_Bambara-CRB",
"UD_Basque-BDT",
"UD_Belarusian-HSE",
+ "UD_Bhojpuri-BHTB",
"UD_Breton-KEB",
"UD_Bulgarian-BTB",
+ "UD_Buryat-BDT",
+ "UD_Cantonese-HK",
"UD_Catalan-AnCora",
+ "UD_Chinese-CFL",
+ "UD_Chinese-GSD",
+ "UD_Chinese-GSDSimp",
+ "UD_Chinese-HK",
+ "UD_Chinese-PUD",
+ "UD_Chukchi-HSE",
+ "UD_Classical_Chinese-Kyoto",
+ "UD_Coptic-Scriptorium",
"UD_Croatian-SET",
"UD_Czech-CAC",
"UD_Czech-CLTT",
"UD_Czech-FicTree",
"UD_Czech-PDT",
+ "UD_Czech-PUD",
"UD_Danish-DDT",
"UD_Dutch-Alpino",
"UD_Dutch-LassySmall",
"UD_English-ESL",
"UD_English-EWT",
"UD_English-GUM",
+ "UD_English-GUMReddit",
"UD_English-LinES",
"UD_English-ParTUT",
"UD_English-Pronouns",
+ "UD_English-PUD",
+ "UD_Erzya-JR",
"UD_Estonian-EDT",
"UD_Estonian-EWT",
+ "UD_Faroese-FarPaHC",
+ "UD_Faroese-OFT",
"UD_Finnish-FTB",
+ "UD_Finnish-OOD",
+ "UD_Finnish-PUD",
"UD_Finnish-TDT",
"UD_French-FQB",
"UD_French-FTB",
"UD_French-GSD",
"UD_French-ParTUT",
+ "UD_French-PUD",
"UD_French-Sequoia",
"UD_French-Spoken",
"UD_Galician-CTG",
@@ -45,60 +78,120 @@ TREEBANKS = [
"UD_German-GSD",
"UD_German-HDT",
"UD_German-LIT",
+ "UD_German-PUD",
+ "UD_Gothic-PROIEL",
"UD_Greek-GDT",
"UD_Hebrew-HTB",
"UD_Hindi_English-HIENCS",
"UD_Hindi-HDTB",
+ "UD_Hindi-PUD",
"UD_Hungarian-Szeged",
+ "UD_Icelandic-IcePaHC",
+ "UD_Icelandic-PUD",
+ "UD_Indonesian-CSUI",
"UD_Indonesian-GSD",
+ "UD_Indonesian-PUD",
"UD_Irish-IDT",
"UD_Italian-ISDT",
"UD_Italian-ParTUT",
"UD_Italian-PoSTWITA",
+ "UD_Italian-PUD",
"UD_Italian-TWITTIRO",
"UD_Italian-VIT",
- "UD_Japanese-BCCWJ",
+ # "UD_Japanese-BCCWJ", no data
"UD_Japanese-GSD",
"UD_Japanese-Modern",
+ "UD_Japanese-PUD",
+ "UD_Karelian-KKPP",
"UD_Kazakh-KTB",
+ "UD_Khunsari-AHA",
+ "UD_Komi_Permyak-UH",
+ "UD_Komi_Zyrian-IKDP",
+ "UD_Komi_Zyrian-Lattice",
"UD_Korean-GSD",
"UD_Korean-Kaist",
+ "UD_Korean-PUD",
+ "UD_Kurmanji-MG",
"UD_Latin-ITTB",
+ "UD_Latin-LLCT",
"UD_Latin-Perseus",
"UD_Latin-PROIEL",
"UD_Latvian-LVTB",
"UD_Lithuanian-ALKSNIS",
"UD_Lithuanian-HSE",
+ "UD_Livvi-KKPP",
"UD_Maltese-MUDT",
+ "UD_Manx-Cadhan",
"UD_Marathi-UFAL",
+ "UD_Mbya_Guarani-Dooley",
+ "UD_Mbya_Guarani-Thomas",
+ "UD_Moksha-JR",
+ "UD_Munduruku-TuDeT",
+ "UD_Naija-NSC",
+ "UD_Nayini-AHA",
+ "UD_North_Sami-Giella",
+ "UD_Norwegian-Bokmaal",
+ "UD_Norwegian-Nynorsk",
+ "UD_Norwegian-NynorskLIA",
+ "UD_Old_Church_Slavonic-PROIEL",
+ "UD_Old_French-SRCMF",
+ "UD_Old_Russian-RNC",
+ "UD_Old_Russian-TOROT",
+ "UD_Old_Turkish-Tonqq",
+ "UD_Persian-PerDT",
"UD_Persian-Seraji",
"UD_Polish-LFG",
"UD_Polish-PDB",
+ "UD_Polish-PUD",
"UD_Portuguese-Bosque",
"UD_Portuguese-GSD",
+ "UD_Portuguese-PUD",
"UD_Romanian-Nonstandard",
"UD_Romanian-RRT",
"UD_Romanian-SiMoNERo",
"UD_Russian-GSD",
+ "UD_Russian-PUD",
"UD_Russian-SynTagRus",
"UD_Russian-Taiga",
+ "UD_Sanskrit-UFAL",
+ "UD_Sanskrit-Vedic",
+ "UD_Scottish_Gaelic-ARCOSG",
"UD_Serbian-SET",
+ "UD_Skolt_Sami-Giellagas",
"UD_Slovak-SNK",
"UD_Slovenian-SSJ",
"UD_Slovenian-SST",
+ "UD_Soi-AHA",
+ "UD_South_Levantine_Arabic-MADAR",
"UD_Spanish-AnCora",
"UD_Spanish-GSD",
+ "UD_Spanish-PUD",
"UD_Swedish-LinES",
+ "UD_Swedish-PUD",
"UD_Swedish_Sign_Language-SSLC",
"UD_Swedish-Talbanken",
+ "UD_Swiss_German-UZH",
+ "UD_Tagalog-TRG",
+ "UD_Tagalog-Ugnayan",
+ "UD_Tamil-MWTT",
"UD_Tamil-TTB",
"UD_Telugu-MTG",
+ "UD_Thai-PUD",
+ "UD_Tupinamba-TuDeT",
+ "UD_Turkish-BOUN",
"UD_Turkish-GB",
+ "UD_Turkish_German-SAGT",
"UD_Turkish-IMST",
+ "UD_Turkish-PUD",
"UD_Ukrainian-IU",
+ "UD_Upper_Sorbian-UFAL",
"UD_Urdu-UDTB",
"UD_Uyghur-UDT",
"UD_Vietnamese-VTB",
+ "UD_Warlpiri-UFAL",
+ "UD_Welsh-CCG",
+ "UD_Wolof-WTB",
+ "UD_Yoruba-YTB",
]
FLAGS = flags.FLAGS
@@ -136,7 +229,7 @@ def run(_):
embeddings_file = None
if embeddings_dir:
embeddings_dir = pathlib.Path(embeddings_dir) / language
- embeddings_file = [f for f in embeddings_dir.iterdir() if "vectors" in f.name and ".vec.gz" in f.name]
+ embeddings_file = [f for f in embeddings_dir.iterdir() if "vectors" in f.name and ".vec" in f.name]
assert len(embeddings_file) == 1, f"Couldn't find embeddings file."
embeddings_file = embeddings_file[0]
@@ -153,14 +246,19 @@ def run(_):
else f"--pretrained_transformer_name {utils.LANG2TRANSFORMER[language]}"}
--serialization_dir {serialization_dir}
--config_path {pathlib.Path.cwd() / 'config.template.jsonnet'}
- --word_batch_size 2500
--notensorboard
"""
- # no XPOS datasets
- if treebank in ["UD_Hungarian-Szeged", "UD_Armenian-ArmTDP"]:
+ # Datasets without XPOS
+ if treebank in {"UD_Armenian-ArmTDP", "UD_Basque-BDT", "UD_Hungarian-Szeged"}:
command = command + " --targets deprel,head,upostag,lemma,feats"
+ # Reduce word_batch_size
+ word_batch_size = 2500
+ if treebank in {"UD_German-HDT"}:
+ word_batch_size = 1000
+ command = command + f" --word_batch_size {word_batch_size}"
+
utils.execute_command(command)
diff --git a/setup.py b/setup.py
index fdaa2be..8d4db9d 100644
--- a/setup.py
+++ b/setup.py
@@ -6,15 +6,16 @@ REQUIREMENTS = [
'allennlp==1.2.1',
'conllu==2.3.2',
'dataclasses;python_version<"3.7"',
- 'joblib==0.14.1',
'jsonnet==0.15.0',
- 'requests==2.23.0',
+ 'numpy==1.19.4',
'overrides==3.1.0',
- 'tensorboard==2.1.0',
+ 'requests==2.23.0',
+ 'spacy==2.3.2',
+ 'scikit-learn<=0.23.2',
'torch==1.6.0',
'tqdm==4.43.0',
'transformers>=3.4.0,<3.5',
- 'urllib3>=1.25.11',
+ 'urllib3==1.25.11',
]
setup(
--
GitLab