diff --git a/tests/core/callbacks/test_checkpoint_callback_torch.py b/tests/core/callbacks/test_checkpoint_callback_torch.py index c700fa79..ca2a3292 100644 --- a/tests/core/callbacks/test_checkpoint_callback_torch.py +++ b/tests/core/callbacks/test_checkpoint_callback_torch.py @@ -16,7 +16,7 @@ from fastNLP.envs import FASTNLP_LAUNCH_TIME, FASTNLP_DISTRIBUTED_CHECK from tests.helpers.utils import magic_argv_env_context from fastNLP.core import rank_zero_rm from tests.helpers.models.torch_model import TorchNormalModel_Classification_1 -from tests.helpers.datasets.torch_data import TorchArgMaxDatset +from tests.helpers.datasets.torch_data import TorchArgMaxDataset from torchmetrics import Accuracy from fastNLP.core.log import logger @@ -53,7 +53,7 @@ def model_and_optimizers(request): feature_dimension=ArgMaxDatasetConfig.feature_dimension ) trainer_params.optimizers = SGD(trainer_params.model.parameters(), lr=0.001) - dataset = TorchArgMaxDatset( + dataset = TorchArgMaxDataset( feature_dimension=ArgMaxDatasetConfig.feature_dimension, data_num=ArgMaxDatasetConfig.data_num, seed=ArgMaxDatasetConfig.seed diff --git a/tests/core/callbacks/test_load_best_model_callback_torch.py b/tests/core/callbacks/test_load_best_model_callback_torch.py index 31933347..0bc63bd5 100644 --- a/tests/core/callbacks/test_load_best_model_callback_torch.py +++ b/tests/core/callbacks/test_load_best_model_callback_torch.py @@ -19,7 +19,7 @@ from fastNLP.core import Evaluator from fastNLP.core.utils.utils import safe_rm from fastNLP.core.drivers.torch_driver import TorchSingleDriver from tests.helpers.models.torch_model import TorchNormalModel_Classification_1 -from tests.helpers.datasets.torch_data import TorchArgMaxDatset +from tests.helpers.datasets.torch_data import TorchArgMaxDataset from tests.helpers.utils import magic_argv_env_context @@ -55,7 +55,7 @@ def model_and_optimizers(request): feature_dimension=ArgMaxDatasetConfig.feature_dimension ) trainer_params.optimizers = optim.SGD(trainer_params.model.parameters(), lr=0.01) - dataset = TorchArgMaxDatset( + dataset = TorchArgMaxDataset( feature_dimension=ArgMaxDatasetConfig.feature_dimension, data_num=ArgMaxDatasetConfig.data_num, seed=ArgMaxDatasetConfig.seed diff --git a/tests/core/callbacks/test_more_evaluate_callback.py b/tests/core/callbacks/test_more_evaluate_callback.py index 1c24ea9a..16ee3e17 100644 --- a/tests/core/callbacks/test_more_evaluate_callback.py +++ b/tests/core/callbacks/test_more_evaluate_callback.py @@ -24,7 +24,7 @@ from fastNLP.envs import FASTNLP_LAUNCH_TIME, FASTNLP_DISTRIBUTED_CHECK from tests.helpers.utils import magic_argv_env_context from fastNLP.core import rank_zero_rm from tests.helpers.models.torch_model import TorchNormalModel_Classification_1 -from tests.helpers.datasets.torch_data import TorchArgMaxDatset +from tests.helpers.datasets.torch_data import TorchArgMaxDataset from torchmetrics import Accuracy from fastNLP.core.metrics import Metric from fastNLP.core.log import logger @@ -64,7 +64,7 @@ def model_and_optimizers(request): feature_dimension=ArgMaxDatasetConfig.feature_dimension ) trainer_params.optimizers = SGD(trainer_params.model.parameters(), lr=0.001) - dataset = TorchArgMaxDatset( + dataset = TorchArgMaxDataset( feature_dimension=ArgMaxDatasetConfig.feature_dimension, data_num=ArgMaxDatasetConfig.data_num, seed=ArgMaxDatasetConfig.seed diff --git a/tests/core/controllers/test_trainer_w_evaluator_torch.py b/tests/core/controllers/test_trainer_w_evaluator_torch.py index 2973e417..94f66403 100644 --- a/tests/core/controllers/test_trainer_w_evaluator_torch.py +++ b/tests/core/controllers/test_trainer_w_evaluator_torch.py @@ -11,7 +11,7 @@ from torchmetrics import Accuracy from fastNLP.core.controllers.trainer import Trainer from tests.helpers.models.torch_model import TorchNormalModel_Classification_1 -from tests.helpers.datasets.torch_data import TorchNormalDataset_Classification, TorchArgMaxDatset +from tests.helpers.datasets.torch_data import TorchNormalDataset_Classification, TorchArgMaxDataset from tests.helpers.callbacks.helper_callbacks import RecordLossCallback, RecordMetricCallback from tests.helpers.utils import magic_argv_env_context @@ -80,7 +80,7 @@ def model_and_optimizers(request): feature_dimension=ArgMaxDatasetConfig.feature_dimension ) trainer_params.optimizers = SGD(trainer_params.model.parameters(), lr=0.001) - dataset = TorchArgMaxDatset( + dataset = TorchArgMaxDataset( feature_dimension=ArgMaxDatasetConfig.feature_dimension, data_num=ArgMaxDatasetConfig.data_num, seed=ArgMaxDatasetConfig.seed diff --git a/tests/core/drivers/torch_driver/test_single_device.py b/tests/core/drivers/torch_driver/test_single_device.py index 4290d02c..b8a8def9 100644 --- a/tests/core/drivers/torch_driver/test_single_device.py +++ b/tests/core/drivers/torch_driver/test_single_device.py @@ -6,7 +6,7 @@ from pathlib import Path from fastNLP.core.drivers.torch_driver.single_device import TorchSingleDriver from fastNLP.core.samplers import RandomBatchSampler, RandomSampler from tests.helpers.models.torch_model import TorchNormalModel_Classification_1 -from tests.helpers.datasets.torch_data import TorchNormalDataset, TorchArgMaxDatset +from tests.helpers.datasets.torch_data import TorchNormalDataset, TorchArgMaxDataset from tests.helpers.datasets.paddle_data import PaddleNormalDataset from tests.helpers.models.paddle_model import PaddleNormalModel_Classification_1 from fastNLP.core import rank_zero_rm @@ -17,7 +17,7 @@ import paddle def dataloader_with_randombatchsampler(dataset, batch_size, shuffle, drop_last): """ - 建立一个 batch_samper 为 RandomBatchSampler 的 dataloader + 建立一个 batch_sampler 为 RandomBatchSampler 的 dataloader """ if shuffle: sampler = torch.utils.data.RandomSampler(dataset) @@ -38,7 +38,7 @@ def dataloader_with_randombatchsampler(dataset, batch_size, shuffle, drop_last): def dataloader_with_randomsampler(dataset, batch_size, shuffle, drop_last, seed=0): """ - 建立一个 samper 为 RandomSampler 的 dataloader + 建立一个 sampler 为 RandomSampler 的 dataloader """ dataloader = DataLoader( dataset, @@ -531,7 +531,7 @@ def generate_random_driver(features, labels, fp16=False, device="cpu"): @pytest.fixture def prepare_test_save_load(): - dataset = TorchArgMaxDatset(10, 40) + dataset = TorchArgMaxDataset(10, 40) dataloader = DataLoader(dataset, batch_size=4) driver1, driver2 = generate_random_driver(10, 10), generate_random_driver(10, 10) return driver1, driver2, dataloader @@ -566,7 +566,7 @@ def test_save_and_load_with_randombatchsampler(only_state_dict, fp16): try: path = "model.ckp" - dataset = TorchArgMaxDatset(10, 40) + dataset = TorchArgMaxDataset(10, 40) dataloader = dataloader_with_randombatchsampler(dataset, 4, True, False) driver1, driver2 = generate_random_driver(10, 10, fp16, "cuda"), generate_random_driver(10, 10, False, "cuda") @@ -636,7 +636,7 @@ def test_save_and_load_with_randomsampler(only_state_dict, fp16): path = "model.ckp" driver1, driver2 = generate_random_driver(10, 10, fp16, "cuda"), generate_random_driver(10, 10, False, "cuda") - dataset = TorchArgMaxDatset(10, 40) + dataset = TorchArgMaxDataset(10, 40) dataloader = dataloader_with_randomsampler(dataset, 4, True, False) num_consumed_batches = 2 diff --git a/tests/helpers/datasets/torch_data.py b/tests/helpers/datasets/torch_data.py index 56648adb..9a0af019 100644 --- a/tests/helpers/datasets/torch_data.py +++ b/tests/helpers/datasets/torch_data.py @@ -38,7 +38,7 @@ class TorchNormalDataset_Classification(Dataset): return {"x": self.x[item], "y": self.y[item]} -class TorchArgMaxDatset(Dataset): +class TorchArgMaxDataset(Dataset): def __init__(self, feature_dimension=10, data_num=1000, seed=0): self.num_labels = feature_dimension self.feature_dimension = feature_dimension