SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Fix CI with change of name of nlp (#7054)

Browse Source 
* nlp -> datasets

* More nlp -> datasets

* Woopsie

* More nlp -> datasets

* One last
This commit is contained in:
Sylvain Gugger
2020-09-10 14:51:08 -04:00
committed by GitHub
parent e9a2f772bc
commit 514486739c
13 changed files with 42 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/transformers/__init__.py
View File

@@ -83,7 +83,7 @@ from .file_utils import (
add_start_docstrings,
cached_path,
is_apex_available,
is_nlp_available,
is_datasets_available,
is_psutil_available,
is_py3nvml_available,
is_tf_available,

10
src/transformers/file_utils.py
View File

@@ -66,12 +66,12 @@ except (ImportError, AssertionError):
try:
import nlp # noqa: F401
import datasets # noqa: F401
_nlp_available = True
_datasets_available = True
except ImportError:
_nlp_available = False
_datasets_available = False
try:
from torch.hub import _get_torch_home
@@ -155,8 +155,8 @@ def is_torch_tpu_available():
return _torch_tpu_available
def is_nlp_available():
return _nlp_available
def is_datasets_available():
return _datasets_available
def is_psutil_available():

30
src/transformers/trainer.py
View File

@@ -20,7 +20,7 @@ from torch.utils.data.sampler import RandomSampler, Sampler, SequentialSampler
from tqdm.auto import tqdm, trange
from .data.data_collator import DataCollator, DataCollatorWithPadding, default_data_collator
from .file_utils import is_nlp_available, is_torch_tpu_available
from .file_utils import is_datasets_available, is_torch_tpu_available
from .integrations import (
default_hp_search_backend,
is_comet_available,
@@ -65,8 +65,8 @@ else:
_use_native_amp = True
from torch.cuda.amp import autocast
if is_nlp_available():
import nlp
if is_datasets_available():
import datasets
if is_torch_tpu_available():
import torch_xla.core.xla_model as xm
@@ -179,10 +179,10 @@ class Trainer:
:obj:`eval_dataset`. Will default to :func:`~transformers.default_data_collator` if no ``tokenizer`` is
provided, an instance of :func:`~transformers.DataCollatorWithPadding` otherwise.
train_dataset (:obj:`torch.utils.data.dataset.Dataset`, `optional`):
The dataset to use for training. If it is an :obj:`nlp.Dataset`, columns not accepted by the
The dataset to use for training. If it is an :obj:`datasets.Dataset`, columns not accepted by the
``model.forward()`` method are automatically removed.
eval_dataset (:obj:`torch.utils.data.dataset.Dataset`, `optional`):
The dataset to use for evaluation. If it is an :obj:`nlp.Dataset`, columns not accepted by the
The dataset to use for evaluation. If it is an :obj:`datasets.Dataset`, columns not accepted by the
``model.forward()`` method are automatically removed.
tokenizer (:class:`PreTrainedTokenizerBase`, `optional`):
The tokenizer used to preprocess the data. If provided, will be used to automatically pad the inputs the
@@ -280,10 +280,10 @@ class Trainer:
FutureWarning,
)
if is_nlp_available():
if isinstance(train_dataset, nlp.Dataset):
if is_datasets_available():
if isinstance(train_dataset, datasets.Dataset):
self._remove_unused_columns(self.train_dataset, description="training")
if isinstance(eval_dataset, nlp.Dataset):
if isinstance(eval_dataset, datasets.Dataset):
self._remove_unused_columns(self.eval_dataset, description="evaluation")
self.global_step = None
@@ -294,7 +294,7 @@ class Trainer:
self.hp_search_backend = None
self.use_tune_checkpoints = False
def _remove_unused_columns(self, dataset: "nlp.Dataset", description: Optional[str] = None):
def _remove_unused_columns(self, dataset: "datasets.Dataset", description: Optional[str] = None):
if not self.args.remove_unused_columns:
return
# Inspect model forward signature to keep only the arguments it accepts.
@@ -364,12 +364,12 @@ class Trainer:
Args:
eval_dataset (:obj:`torch.utils.data.dataset.Dataset`, `optional`):
If provided, will override :obj:`self.eval_dataset`. If it is an :obj:`nlp.Dataset`, columns not
If provided, will override :obj:`self.eval_dataset`. If it is an :obj:`datasets.Dataset`, columns not
accepted by the ``model.forward()`` method are automatically removed.
"""
if eval_dataset is None and self.eval_dataset is None:
raise ValueError("Trainer: evaluation requires an eval_dataset.")
elif eval_dataset is not None and is_nlp_available() and isinstance(eval_dataset, nlp.Dataset):
elif eval_dataset is not None and is_datasets_available() and isinstance(eval_dataset, datasets.Dataset):
self._remove_unused_columns(eval_dataset, description="evaluation")
eval_dataset = eval_dataset if eval_dataset is not None else self.eval_dataset
eval_sampler = self._get_eval_sampler(eval_dataset)
@@ -393,10 +393,10 @@ class Trainer:
Args:
eval_dataset (:obj:`torch.utils.data.dataset.Dataset`, `optional`):
The test dataset to use. If it is an :obj:`nlp.Dataset`, columns not accepted by the
The test dataset to use. If it is an :obj:`datasets.Dataset`, columns not accepted by the
``model.forward()`` method are automatically removed.
"""
if is_nlp_available() and isinstance(test_dataset, nlp.Dataset):
if is_datasets_available() and isinstance(test_dataset, datasets.Dataset):
self._remove_unused_columns(test_dataset, description="test")
test_sampler = self._get_eval_sampler(test_dataset)
@@ -1200,7 +1200,7 @@ class Trainer:
Args:
eval_dataset (:obj:`Dataset`, `optional`):
Pass a dataset if you wish to override :obj:`self.eval_dataset`. If it is an :obj:`nlp.Dataset`,
Pass a dataset if you wish to override :obj:`self.eval_dataset`. If it is an :obj:`datasets.Dataset`,
columns not accepted by the ``model.forward()`` method are automatically removed.
Returns:
@@ -1227,7 +1227,7 @@ class Trainer:
Args:
test_dataset (:obj:`Dataset`):
Dataset to run the predictions on. If it is an :obj:`nlp.Dataset`, columns not accepted by the
Dataset to run the predictions on. If it is an :obj:`datasets.Dataset`, columns not accepted by the
``model.forward()`` method are automatically removed.
Returns: