diff --git a/docs/api/api_python/dataset/mindspore.dataset.OBSMindDataset.rst b/docs/api/api_python/dataset/mindspore.dataset.OBSMindDataset.rst index c897e09ae2e..18c382949af 100644 --- a/docs/api/api_python/dataset/mindspore.dataset.OBSMindDataset.rst +++ b/docs/api/api_python/dataset/mindspore.dataset.OBSMindDataset.rst @@ -40,8 +40,6 @@ - 如果线下训练，建议为每次训练设置 `BATCH_JOB_ID` 环境变量。 - 分布式训练中，假如使用多个节点（服务器），则必须使用每个节点全部的8张卡。如果只有一个节点（服务器），则没有这样的限制。 - .. include:: mindspore.dataset.Dataset.add_sampler.rst - .. include:: mindspore.dataset.Dataset.rst .. include:: mindspore.dataset.Dataset.b.rst @@ -50,6 +48,4 @@ .. include:: mindspore.dataset.Dataset.d.rst - .. include:: mindspore.dataset.Dataset.use_sampler.rst - .. include:: mindspore.dataset.Dataset.zip.rst diff --git a/docs/api/api_python/mindrecord/mindspore.mindrecord.FileWriter.rst b/docs/api/api_python/mindrecord/mindspore.mindrecord.FileWriter.rst index 54c3a2351f4..8d17ade2d95 100644 --- a/docs/api/api_python/mindrecord/mindspore.mindrecord.FileWriter.rst +++ b/docs/api/api_python/mindrecord/mindspore.mindrecord.FileWriter.rst @@ -9,7 +9,7 @@ **参数：** - **file_name** (str) - 转换生成的MindRecord文件路径。 - - **shard_num** (int，可选) - 生成MindRecord的文件个数。默认值：1。取值范围为[1, 1000]。 + - **shard_num** (int，可选) - 生成MindRecord的文件个数。取值范围为[1, 1000]。默认值：1。 - **overwrite** (bool，可选) - 当指定目录存在同名文件时是否覆盖写。默认值：False。 **异常：** @@ -52,7 +52,7 @@ **参数：** - **content** (dict) - schema内容的字典。 - - **desc** (str，可选) - schema的描述字符。默认值：None。 + - **desc** (str，可选) - schema的描述。默认值：None。 **返回：** @@ -87,7 +87,7 @@ .. py:method:: open_and_set_header() - 打开MindRecord文件准备写入并且设置header。该函数仅用于并行写入，并在 `write_raw_data` 函数之前调用。 + 打开MindRecord文件准备写入并且设置描述其meta信息的头部，该函数仅用于并行写入，并在 `write_raw_data` 函数之前调用。 **返回：** diff --git a/docs/api/api_python/mindrecord/mindspore.mindrecord.MnistToMR.rst b/docs/api/api_python/mindrecord/mindspore.mindrecord.MnistToMR.rst index 118f3b50c03..2cfc6af327a 100644 --- a/docs/api/api_python/mindrecord/mindspore.mindrecord.MnistToMR.rst +++ b/docs/api/api_python/mindrecord/mindspore.mindrecord.MnistToMR.rst @@ -5,7 +5,7 @@ **参数：** - - **source** (str) - 包含t10k-images-idx3-ubyte.gz、train-images-idx3-ubyte.gz、t10k-labels-idx1-ubyte.gz和train-labels-idx1-ubyte.gz数据集文件的目录路径。 + - **source** (str) - 数据集目录路径，其包含t10k-images-idx3-ubyte.gz、train-images-idx3-ubyte.gz、t10k-labels-idx1-ubyte.gz和train-labels-idx1-ubyte.gz数据集文件。 - **destination** (str) - 转换生成的MindRecord文件路径，需提前创建目录并且目录下不能存在同名文件。 - **partition_number** (int，可选) - 生成MindRecord的文件个数。默认值：1。 diff --git a/mindspore/python/mindspore/dataset/engine/datasets_standard_format.py b/mindspore/python/mindspore/dataset/engine/datasets_standard_format.py index da134576b9f..1569c5d1ac3 100644 --- a/mindspore/python/mindspore/dataset/engine/datasets_standard_format.py +++ b/mindspore/python/mindspore/dataset/engine/datasets_standard_format.py @@ -421,20 +421,8 @@ class OBSMindDataset(GeneratorDataset): def add_sampler(self, new_sampler): - """ - Add a sampler for current dataset, it's not supported for OBSMindDataset. - - Args: - new_sampler (Sampler): The sampler to be added as the parent sampler for current dataset. - """ raise NotImplementedError("add_sampler is not supported for OBSMindDataset.") def use_sampler(self, new_sampler): - """ - Make the current dataset use the new_sampler provided by other API, it's not supported for OBSMindDataset. - - Args: - new_sampler (Sampler): The sampler to use for the current dataset. - """ raise NotImplementedError("use_sampler is not supported for OBSMindDataset.") diff --git a/mindspore/python/mindspore/dataset/engine/datasets_user_defined.py b/mindspore/python/mindspore/dataset/engine/datasets_user_defined.py index 3f93c0e37bf..a5cc35e5951 100644 --- a/mindspore/python/mindspore/dataset/engine/datasets_user_defined.py +++ b/mindspore/python/mindspore/dataset/engine/datasets_user_defined.py @@ -135,7 +135,7 @@ def _convert_row(row): """ if isinstance(row, dict): raise TypeError("Input data is expected to be " \ - "int, float, str, bytes, Numpy array, Tensor or list-or-tuple of them, but got dict.") + "int, float, str, bytes, numpy.ndarray, Tensor or list/tuple of them, but got dict.") # convert single item to np.array prim_type = (int, float, str, bytes, np.ndarray, Tensor) @@ -147,7 +147,7 @@ def _convert_row(row): else: item = np.array(row, copy=False) if item.dtype == 'object': - raise TypeError("Data type of the input or converted Numpy array is expected to be " \ + raise TypeError("Data type of the input or its converted Numpy array is expected to be " \ "int or float or str, but got {}.".format(item.dtype)) return tuple([item]) @@ -161,12 +161,12 @@ def _convert_row(row): elif isinstance(x, Tensor): # mindspore.Tensor value.append(x.asnumpy()) elif isinstance(x, dict): - raise TypeError("The {}st item of input data is expected to be " \ - "int, float, str, bytes, Numpy array, Tensor, but got dict.".format(idx)) + raise TypeError("The {}th item of input data is expected to be " \ + "int, float, str, bytes, numpy.ndarray, Tensor, but got dict.".format(idx)) else: item = np.array(x, copy=False) if item.dtype == 'object': - raise TypeError("Data type of {}st item of the input or converted Numpy array is expected to be " \ + raise TypeError("Data type of {}th item of the input or its converted Numpy array is expected to be " \ "int or float or str, but got {}.".format(idx, item.dtype)) value.append(item) return tuple(value) diff --git a/mindspore/python/mindspore/dataset/engine/obs/config_loader.py b/mindspore/python/mindspore/dataset/engine/obs/config_loader.py index 4d5b1fb1cfe..27667406a09 100644 --- a/mindspore/python/mindspore/dataset/engine/obs/config_loader.py +++ b/mindspore/python/mindspore/dataset/engine/obs/config_loader.py @@ -47,7 +47,7 @@ class _Config: return self._convert_type(key) if key in self.config: return self.config[key] - raise RuntimeError("Variable {} can not found in configuration file or environment variables.".format(key)) + raise RuntimeError("Variable {} can not be found in configuration file or environment variables.".format(key)) def __setattr__(self, key, value): if key == 'config': diff --git a/mindspore/python/mindspore/mindrecord/filewriter.py b/mindspore/python/mindspore/mindrecord/filewriter.py index ec8b0761752..c5becd75672 100644 --- a/mindspore/python/mindspore/mindrecord/filewriter.py +++ b/mindspore/python/mindspore/mindrecord/filewriter.py @@ -42,8 +42,8 @@ class FileWriter: Args: file_name (str): File name of MindRecord file. - shard_num (int, optional): The Number of MindRecord files. Default: 1. - It should be between [1, 1000]. + shard_num (int, optional): The Number of MindRecord files. + It should be between [1, 1000]. Default: 1. overwrite (bool, optional): Whether to overwrite if the file already exists. Default: False. Raises: @@ -273,7 +273,7 @@ class FileWriter: def open_and_set_header(self): """ - Open writer and set header. The function is only used for parallel \ + Open writer and set header which stores meta information. The function is only used for parallel \ writing and is called before the `write_raw_data`. Returns: diff --git a/tests/ut/python/dataset/test_datasets_generator.py b/tests/ut/python/dataset/test_datasets_generator.py index f57cc7ddc83..ccdc7939033 100644 --- a/tests/ut/python/dataset/test_datasets_generator.py +++ b/tests/ut/python/dataset/test_datasets_generator.py @@ -641,7 +641,7 @@ def test_generator_error_1(): data1 = ds.GeneratorDataset(generator_np, ["data"]) for _ in data1: pass - assert "Data type of 1st item of the input or converted Numpy array is expected" in str(info.value) + assert "Data type of 1th item of the input or its converted Numpy array is expected" in str(info.value) def test_generator_error_2(): @@ -653,7 +653,7 @@ def test_generator_error_2(): data1 = ds.GeneratorDataset(generator_np, ["data"]) for _ in data1: pass - assert "Data type of 1st item of the input or converted Numpy array is expected" in str(info.value) + assert "Data type of 1th item of the input or its converted Numpy array is expected" in str(info.value) def test_generator_error_3(): @@ -1360,7 +1360,7 @@ def test_generator_single_input_6(): data1 = ds.GeneratorDataset(data, ["data"], shuffle=False) for _ in data1.create_dict_iterator(num_epochs=1, output_numpy=True): # each data is a dictionary pass - assert " Data type of the input or converted Numpy array is expected" in str(info.value) + assert " Data type of the input or its converted Numpy array is expected" in str(info.value) assert_generator_single_input_6(generator_nested_np) assert_generator_single_input_6(RandomAccessDatasetInner())