chendl/compositional_test

1# Copyright 2020 The HuggingFace Team. All rights reserved.2#3# Licensed under the Apache License, Version 2.0 (the "License");4# you may not use this file except in compliance with the License.5# You may obtain a copy of the License at6#7#     http://www.apache.org/licenses/LICENSE-2.08#9# Unless required by applicable law or agreed to in writing, software10# distributed under the License is distributed on an "AS IS" BASIS,11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.12# See the License for the specific language governing permissions and13# limitations under the License.14 15import os16import sys17 18SRC_DIR = os.path.join(os.path.dirname(__file__), "src")19sys.path.append(SRC_DIR)20 21 22from transformers import (23    AutoConfig,24    AutoModel,25    AutoModelForCausalLM,26    AutoModelForMaskedLM,27    AutoModelForQuestionAnswering,28    AutoModelForSequenceClassification,29    AutoTokenizer,30    add_start_docstrings,31)32 33 34dependencies = ["torch", "numpy", "tokenizers", "filelock", "requests", "tqdm", "regex", "sentencepiece", "sacremoses", "importlib_metadata", "huggingface_hub"]35 36 37@add_start_docstrings(AutoConfig.__doc__)38def config(*args, **kwargs):39    r"""40                # Using torch.hub !41                import torch42 43                config = torch.hub.load('huggingface/transformers', 'config', 'bert-base-uncased')  # Download configuration from huggingface.co and cache.44                config = torch.hub.load('huggingface/transformers', 'config', './test/bert_saved_model/')  # E.g. config (or model) was saved using `save_pretrained('./test/saved_model/')`45                config = torch.hub.load('huggingface/transformers', 'config', './test/bert_saved_model/my_configuration.json')46                config = torch.hub.load('huggingface/transformers', 'config', 'bert-base-uncased', output_attentions=True, foo=False)47                assert config.output_attentions == True48                config, unused_kwargs = torch.hub.load('huggingface/transformers', 'config', 'bert-base-uncased', output_attentions=True, foo=False, return_unused_kwargs=True)49                assert config.output_attentions == True50                assert unused_kwargs == {'foo': False}51 52            """53 54    return AutoConfig.from_pretrained(*args, **kwargs)55 56 57@add_start_docstrings(AutoTokenizer.__doc__)58def tokenizer(*args, **kwargs):59    r"""60        # Using torch.hub !61        import torch62 63        tokenizer = torch.hub.load('huggingface/transformers', 'tokenizer', 'bert-base-uncased')    # Download vocabulary from huggingface.co and cache.64        tokenizer = torch.hub.load('huggingface/transformers', 'tokenizer', './test/bert_saved_model/')  # E.g. tokenizer was saved using `save_pretrained('./test/saved_model/')`65 66    """67 68    return AutoTokenizer.from_pretrained(*args, **kwargs)69 70 71@add_start_docstrings(AutoModel.__doc__)72def model(*args, **kwargs):73    r"""74            # Using torch.hub !75            import torch76 77            model = torch.hub.load('huggingface/transformers', 'model', 'bert-base-uncased')    # Download model and configuration from huggingface.co and cache.78            model = torch.hub.load('huggingface/transformers', 'model', './test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`79            model = torch.hub.load('huggingface/transformers', 'model', 'bert-base-uncased', output_attentions=True)  # Update configuration during loading80            assert model.config.output_attentions == True81            # Loading from a TF checkpoint file instead of a PyTorch model (slower)82            config = AutoConfig.from_pretrained('./tf_model/bert_tf_model_config.json')83            model = torch.hub.load('huggingface/transformers', 'model', './tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)84 85        """86 87    return AutoModel.from_pretrained(*args, **kwargs)88 89 90@add_start_docstrings(AutoModelForCausalLM.__doc__)91def modelForCausalLM(*args, **kwargs):92    r"""93        # Using torch.hub !94        import torch95 96        model = torch.hub.load('huggingface/transformers', 'modelForCausalLM', 'gpt2')    # Download model and configuration from huggingface.co and cache.97        model = torch.hub.load('huggingface/transformers', 'modelForCausalLM', './test/saved_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`98        model = torch.hub.load('huggingface/transformers', 'modelForCausalLM', 'gpt2', output_attentions=True)  # Update configuration during loading99        assert model.config.output_attentions == True100        # Loading from a TF checkpoint file instead of a PyTorch model (slower)101        config = AutoConfig.from_pretrained('./tf_model/gpt_tf_model_config.json')102        model = torch.hub.load('huggingface/transformers', 'modelForCausalLM', './tf_model/gpt_tf_checkpoint.ckpt.index', from_tf=True, config=config)103 104    """105    return AutoModelForCausalLM.from_pretrained(*args, **kwargs)106 107 108@add_start_docstrings(AutoModelForMaskedLM.__doc__)109def modelForMaskedLM(*args, **kwargs):110    r"""111            # Using torch.hub !112            import torch113 114            model = torch.hub.load('huggingface/transformers', 'modelForMaskedLM', 'bert-base-uncased')    # Download model and configuration from huggingface.co and cache.115            model = torch.hub.load('huggingface/transformers', 'modelForMaskedLM', './test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`116            model = torch.hub.load('huggingface/transformers', 'modelForMaskedLM', 'bert-base-uncased', output_attentions=True)  # Update configuration during loading117            assert model.config.output_attentions == True118            # Loading from a TF checkpoint file instead of a PyTorch model (slower)119            config = AutoConfig.from_pretrained('./tf_model/bert_tf_model_config.json')120            model = torch.hub.load('huggingface/transformers', 'modelForMaskedLM', './tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)121 122        """123 124    return AutoModelForMaskedLM.from_pretrained(*args, **kwargs)125 126 127@add_start_docstrings(AutoModelForSequenceClassification.__doc__)128def modelForSequenceClassification(*args, **kwargs):129    r"""130            # Using torch.hub !131            import torch132 133            model = torch.hub.load('huggingface/transformers', 'modelForSequenceClassification', 'bert-base-uncased')    # Download model and configuration from huggingface.co and cache.134            model = torch.hub.load('huggingface/transformers', 'modelForSequenceClassification', './test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`135            model = torch.hub.load('huggingface/transformers', 'modelForSequenceClassification', 'bert-base-uncased', output_attentions=True)  # Update configuration during loading136            assert model.config.output_attentions == True137            # Loading from a TF checkpoint file instead of a PyTorch model (slower)138            config = AutoConfig.from_pretrained('./tf_model/bert_tf_model_config.json')139            model = torch.hub.load('huggingface/transformers', 'modelForSequenceClassification', './tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)140 141        """142 143    return AutoModelForSequenceClassification.from_pretrained(*args, **kwargs)144 145 146@add_start_docstrings(AutoModelForQuestionAnswering.__doc__)147def modelForQuestionAnswering(*args, **kwargs):148    r"""149        # Using torch.hub !150        import torch151 152        model = torch.hub.load('huggingface/transformers', 'modelForQuestionAnswering', 'bert-base-uncased')    # Download model and configuration from huggingface.co and cache.153        model = torch.hub.load('huggingface/transformers', 'modelForQuestionAnswering', './test/bert_model/')  # E.g. model was saved using `save_pretrained('./test/saved_model/')`154        model = torch.hub.load('huggingface/transformers', 'modelForQuestionAnswering', 'bert-base-uncased', output_attentions=True)  # Update configuration during loading155        assert model.config.output_attentions == True156        # Loading from a TF checkpoint file instead of a PyTorch model (slower)157        config = AutoConfig.from_pretrained('./tf_model/bert_tf_model_config.json')158        model = torch.hub.load('huggingface/transformers', 'modelForQuestionAnswering', './tf_model/bert_tf_checkpoint.ckpt.index', from_tf=True, config=config)159 160    """161    return AutoModelForQuestionAnswering.from_pretrained(*args, **kwargs)162