Move test model folders (#17034)
* move test model folders (TODO: fix imports and others) * fix (potentially partially) imports (in model test modules) * fix (potentially partially) imports (in tokenization test modules) * fix (potentially partially) imports (in feature extraction test modules) * fix import utils.test_modeling_tf_core * fix path ../fixtures/ * fix imports about generation.test_generation_flax_utils * fix more imports * fix fixture path * fix get_test_dir * update module_to_test_file * fix get_tests_dir from wrong transformers.utils * update config.yml (CircleCI) * fix style * remove missing imports * update new model script * update check_repo * update SPECIAL_MODULE_TO_TEST_MAP * fix style * add __init__ * update self-scheduled * fix add_new_model scripts * check one way to get location back * python setup.py build install * fix import in test auto * update self-scheduled.yml * update slack notification script * Add comments about artifact names * fix for yolos Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
This commit is contained in:
Yih-Dar
0
tests/models/mt5/__init__.py
Normal file
0
tests/models/mt5/__init__.py
Normal file
61
tests/models/mt5/test_modeling_flax_mt5.py
Normal file
61
tests/models/mt5/test_modeling_flax_mt5.py
Normal file
@@ -0,0 +1,61 @@
|
||||
# Copyright 2021 The HuggingFace Team. All rights reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import unittest
|
||||
|
||||
from transformers import is_flax_available
|
||||
from transformers.testing_utils import require_flax, require_sentencepiece, require_tokenizers, require_torch, slow
|
||||
|
||||
|
||||
if is_flax_available():
|
||||
import optax
|
||||
from flax.training.common_utils import onehot
|
||||
from transformers import AutoTokenizer, FlaxMT5ForConditionalGeneration
|
||||
from transformers.models.t5.modeling_flax_t5 import shift_tokens_right
|
||||
|
||||
|
||||
@require_torch
|
||||
@require_sentencepiece
|
||||
@require_tokenizers
|
||||
@require_flax
|
||||
class MT5IntegrationTest(unittest.TestCase):
|
||||
@slow
|
||||
def test_small_integration_test(self):
|
||||
"""
|
||||
For comparision run:
|
||||
>>> import t5 # pip install t5==0.7.1
|
||||
>>> from t5.data.sentencepiece_vocabulary import SentencePieceVocabulary
|
||||
|
||||
>>> path_to_mtf_small_mt5_checkpoint = '<fill_in>'
|
||||
>>> path_to_mtf_small_mt5_spm_model_path = '<fill_in>'
|
||||
>>> t5_model = t5.models.MtfModel(model_dir=path_to_mtf_small_mt5_checkpoint, batch_size=1, tpu=None)
|
||||
>>> vocab = SentencePieceVocabulary(path_to_mtf_small_mt5_spm_model_path)
|
||||
>>> score = t5_model.score(inputs=["Hello there"], targets=["Hi I am"], vocabulary=vocab)
|
||||
"""
|
||||
|
||||
model = FlaxMT5ForConditionalGeneration.from_pretrained("google/mt5-small")
|
||||
tokenizer = AutoTokenizer.from_pretrained("google/mt5-small")
|
||||
|
||||
input_ids = tokenizer("Hello there", return_tensors="np").input_ids
|
||||
labels = tokenizer("Hi I am", return_tensors="np").input_ids
|
||||
|
||||
decoder_input_ids = shift_tokens_right(labels, model.config.pad_token_id, model.config.decoder_start_token_id)
|
||||
|
||||
logits = model(input_ids, decoder_input_ids=decoder_input_ids).logits
|
||||
loss = optax.softmax_cross_entropy(logits, onehot(labels, logits.shape[-1])).mean()
|
||||
|
||||
mtf_score = -(labels.shape[-1] * loss.item())
|
||||
|
||||
EXPECTED_SCORE = -84.9127
|
||||
self.assertTrue(abs(mtf_score - EXPECTED_SCORE) < 1e-4)
|
||||
53
tests/models/mt5/test_modeling_mt5.py
Normal file
53
tests/models/mt5/test_modeling_mt5.py
Normal file
@@ -0,0 +1,53 @@
|
||||
# Copyright 2020 The HuggingFace Team. All rights reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import unittest
|
||||
|
||||
from transformers import is_torch_available
|
||||
from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device
|
||||
|
||||
|
||||
if is_torch_available():
|
||||
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
||||
|
||||
|
||||
@require_torch
|
||||
@require_sentencepiece
|
||||
@require_tokenizers
|
||||
class MT5IntegrationTest(unittest.TestCase):
|
||||
@slow
|
||||
def test_small_integration_test(self):
|
||||
"""
|
||||
For comparision run:
|
||||
>>> import t5 # pip install t5==0.7.1
|
||||
>>> from t5.data.sentencepiece_vocabulary import SentencePieceVocabulary
|
||||
|
||||
>>> path_to_mtf_small_mt5_checkpoint = '<fill_in>'
|
||||
>>> path_to_mtf_small_mt5_spm_model_path = '<fill_in>'
|
||||
>>> t5_model = t5.models.MtfModel(model_dir=path_to_mtf_small_mt5_checkpoint, batch_size=1, tpu=None)
|
||||
>>> vocab = SentencePieceVocabulary(path_to_mtf_small_mt5_spm_model_path)
|
||||
>>> score = t5_model.score(inputs=["Hello there"], targets=["Hi I am"], vocabulary=vocab)
|
||||
"""
|
||||
|
||||
model = AutoModelForSeq2SeqLM.from_pretrained("google/mt5-small", return_dict=True).to(torch_device)
|
||||
tokenizer = AutoTokenizer.from_pretrained("google/mt5-small")
|
||||
|
||||
input_ids = tokenizer("Hello there", return_tensors="pt").input_ids
|
||||
labels = tokenizer("Hi I am", return_tensors="pt").input_ids
|
||||
|
||||
loss = model(input_ids.to(torch_device), labels=labels.to(torch_device)).loss
|
||||
mtf_score = -(labels.shape[-1] * loss.item())
|
||||
|
||||
EXPECTED_SCORE = -84.9127
|
||||
self.assertTrue(abs(mtf_score - EXPECTED_SCORE) < 1e-4)
|
||||
73
tests/models/mt5/test_modeling_tf_mt5.py
Normal file
73
tests/models/mt5/test_modeling_tf_mt5.py
Normal file
@@ -0,0 +1,73 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2020 The HuggingFace Team. All rights reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import unittest
|
||||
|
||||
from transformers import is_tf_available
|
||||
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
|
||||
|
||||
|
||||
if is_tf_available():
|
||||
import tensorflow as tf
|
||||
|
||||
from transformers import AutoTokenizer, T5Tokenizer, TFAutoModelForSeq2SeqLM, TFMT5ForConditionalGeneration
|
||||
|
||||
|
||||
@require_tf
|
||||
class TFMT5ModelTest(unittest.TestCase): # no mixin with common tests -> most cases are already covered in the TF T5
|
||||
@slow
|
||||
def test_resize_embeddings(self):
|
||||
model = TFMT5ForConditionalGeneration.from_pretrained("google/mt5-small")
|
||||
original_vocab_size = model.get_input_embeddings().weight.shape[0]
|
||||
# the vocab size is defined in the model config
|
||||
self.assertEqual(original_vocab_size, model.config.vocab_size)
|
||||
|
||||
tokenizer = T5Tokenizer.from_pretrained("google/mt5-small")
|
||||
tokenizer.add_special_tokens({"bos_token": "", "eos_token": ""})
|
||||
model._resize_token_embeddings(len(tokenizer))
|
||||
# the vocab size is now resized to the length of the tokenizer, which is different from the original size
|
||||
self.assertEqual(model.get_input_embeddings().weight.shape[0], len(tokenizer))
|
||||
self.assertNotEqual(model.get_input_embeddings().weight.shape[0], original_vocab_size)
|
||||
|
||||
|
||||
@require_tf
|
||||
@require_sentencepiece
|
||||
@require_tokenizers
|
||||
class TFMT5ModelIntegrationTest(unittest.TestCase):
|
||||
@slow
|
||||
def test_small_integration_test(self):
|
||||
"""
|
||||
For comparision run:
|
||||
>>> import t5 # pip install t5==0.7.1
|
||||
>>> from t5.data.sentencepiece_vocabulary import SentencePieceVocabulary
|
||||
|
||||
>>> path_to_mtf_small_mt5_checkpoint = '<fill_in>'
|
||||
>>> path_to_mtf_small_mt5_spm_model_path = '<fill_in>'
|
||||
>>> t5_model = t5.models.MtfModel(model_dir=path_to_mtf_small_mt5_checkpoint, batch_size=1, tpu=None)
|
||||
>>> vocab = SentencePieceVocabulary(path_to_mtf_small_mt5_spm_model_path, extra_ids=100)
|
||||
>>> score = t5_model.score(inputs=["Hello there"], targets=["Hi I am"], vocabulary=vocab)
|
||||
"""
|
||||
|
||||
model = TFAutoModelForSeq2SeqLM.from_pretrained("google/mt5-small")
|
||||
tokenizer = AutoTokenizer.from_pretrained("google/mt5-small")
|
||||
|
||||
input_ids = tokenizer("Hello there", return_tensors="tf").input_ids
|
||||
labels = tokenizer("Hi I am", return_tensors="tf").input_ids
|
||||
|
||||
loss = model(input_ids, labels=labels).loss
|
||||
mtf_score = -tf.math.reduce_sum(loss).numpy()
|
||||
|
||||
EXPECTED_SCORE = -84.9127
|
||||
self.assertTrue(abs(mtf_score - EXPECTED_SCORE) < 2e-4)
|
||||
Reference in New Issue
Block a user