Spaces:
Paused
Paused
| # coding=utf-8 | |
| # Copyright 2022 The HuggingFace Inc. 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. | |
| """ Testing suite for the PyTorch Blip model. """ | |
| import unittest | |
| import numpy as np | |
| from transformers import BlipTextConfig | |
| from transformers.testing_utils import require_torch, slow, torch_device | |
| from transformers.utils import is_torch_available | |
| from ...test_configuration_common import ConfigTester | |
| from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask | |
| if is_torch_available(): | |
| import torch | |
| from transformers import BlipTextModel | |
| from transformers.models.blip.modeling_blip import BLIP_PRETRAINED_MODEL_ARCHIVE_LIST | |
| class BlipTextModelTester: | |
| def __init__( | |
| self, | |
| parent, | |
| batch_size=12, | |
| seq_length=7, | |
| is_training=True, | |
| use_input_mask=True, | |
| use_labels=True, | |
| vocab_size=99, | |
| hidden_size=32, | |
| projection_dim=32, | |
| num_hidden_layers=2, | |
| num_attention_heads=4, | |
| intermediate_size=37, | |
| dropout=0.1, | |
| attention_dropout=0.1, | |
| max_position_embeddings=512, | |
| initializer_range=0.02, | |
| bos_token_id=0, | |
| scope=None, | |
| ): | |
| self.parent = parent | |
| self.batch_size = batch_size | |
| self.seq_length = seq_length | |
| self.is_training = is_training | |
| self.use_input_mask = use_input_mask | |
| self.use_labels = use_labels | |
| self.vocab_size = vocab_size | |
| self.hidden_size = hidden_size | |
| self.projection_dim = projection_dim | |
| self.num_hidden_layers = num_hidden_layers | |
| self.num_attention_heads = num_attention_heads | |
| self.intermediate_size = intermediate_size | |
| self.dropout = dropout | |
| self.attention_dropout = attention_dropout | |
| self.max_position_embeddings = max_position_embeddings | |
| self.initializer_range = initializer_range | |
| self.scope = scope | |
| self.bos_token_id = bos_token_id | |
| def prepare_config_and_inputs(self): | |
| input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size) | |
| input_mask = None | |
| if self.use_input_mask: | |
| input_mask = random_attention_mask([self.batch_size, self.seq_length]) | |
| if input_mask is not None: | |
| batch_size, seq_length = input_mask.shape | |
| rnd_start_indices = np.random.randint(1, seq_length - 1, size=(batch_size,)) | |
| for batch_idx, start_index in enumerate(rnd_start_indices): | |
| input_mask[batch_idx, :start_index] = 1 | |
| input_mask[batch_idx, start_index:] = 0 | |
| config = self.get_config() | |
| return config, input_ids, input_mask | |
| def get_config(self): | |
| return BlipTextConfig( | |
| vocab_size=self.vocab_size, | |
| hidden_size=self.hidden_size, | |
| projection_dim=self.projection_dim, | |
| num_hidden_layers=self.num_hidden_layers, | |
| num_attention_heads=self.num_attention_heads, | |
| intermediate_size=self.intermediate_size, | |
| dropout=self.dropout, | |
| attention_dropout=self.attention_dropout, | |
| max_position_embeddings=self.max_position_embeddings, | |
| initializer_range=self.initializer_range, | |
| bos_token_id=self.bos_token_id, | |
| ) | |
| def create_and_check_model(self, config, input_ids, input_mask): | |
| model = BlipTextModel(config=config) | |
| model.to(torch_device) | |
| model.eval() | |
| with torch.no_grad(): | |
| result = model(input_ids, attention_mask=input_mask) | |
| result = model(input_ids) | |
| self.parent.assertEqual(result.last_hidden_state.shape, (self.batch_size, self.seq_length, self.hidden_size)) | |
| self.parent.assertEqual(result.pooler_output.shape, (self.batch_size, self.hidden_size)) | |
| def prepare_config_and_inputs_for_common(self): | |
| config_and_inputs = self.prepare_config_and_inputs() | |
| config, input_ids, input_mask = config_and_inputs | |
| inputs_dict = {"input_ids": input_ids, "attention_mask": input_mask} | |
| return config, inputs_dict | |
| class BlipTextModelTest(ModelTesterMixin, unittest.TestCase): | |
| all_model_classes = (BlipTextModel,) if is_torch_available() else () | |
| fx_compatible = False | |
| test_pruning = False | |
| test_head_masking = False | |
| def setUp(self): | |
| self.model_tester = BlipTextModelTester(self) | |
| self.config_tester = ConfigTester(self, config_class=BlipTextConfig, hidden_size=37) | |
| def test_config(self): | |
| self.config_tester.run_common_tests() | |
| def test_model(self): | |
| config_and_inputs = self.model_tester.prepare_config_and_inputs() | |
| self.model_tester.create_and_check_model(*config_and_inputs) | |
| def test_training(self): | |
| pass | |
| def test_training_gradient_checkpointing(self): | |
| pass | |
| def test_inputs_embeds(self): | |
| pass | |
| def test_save_load_fast_init_from_base(self): | |
| pass | |
| def test_save_load_fast_init_to_base(self): | |
| pass | |
| def test_model_from_pretrained(self): | |
| for model_name in BLIP_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: | |
| model = BlipTextModel.from_pretrained(model_name) | |
| self.assertIsNotNone(model) | |
| def test_pt_tf_model_equivalence(self): | |
| super().test_pt_tf_model_equivalence(allow_missing_keys=True) | |