mindspore 2.2.10__cp37-none-any.whl → 2.2.14__cp37-none-any.whl

mindspore/lib/plugin/ascend/custom_aicpu_ops/op_impl/cpu/config/cust_aicpu_kernel.json

@@ -2913,6 +2913,30 @@
             "type":"DT_INT32,DT_INT64"
         }
     },
+    "CustGenerateEodMask":{
+        "input0":{
+            "name":"inputs_ids",
+            "type":"DT_INT32,DT_DOUBLE,DT_FLOAT,DT_FLOAT16,DT_BF16"
+        },
+        "opInfo":{
+            "computeCost":"100",
+            "engine":"DNN_VM_AICPU",
+            "flagAsync":"False",
+            "flagPartial":"False",
+            "formatAgnostic":"False",
+            "functionName":"RunCpuKernel",
+            "kernelSo":"libcust_cpu_kernels.so",
+            "opKernelLib":"CUSTAICPUKernel",
+            "opsFlag":"OPS_FLAG_CLOSE",
+            "subTypeOfInferShape":"1",
+            "userDefined":"True",
+            "workspaceSize":"1024"
+        },
+        "output0":{
+            "name":"position_ids",
+            "type":"DT_INT32,DT_DOUBLE,DT_FLOAT,DT_FLOAT16,DT_BF16"
+        }
+    },
     "CustGeqrf":{
         "input0":{
             "name":"x",
@@ -3292,6 +3316,41 @@
             "type":"DT_DOUBLE,DT_FLOAT,DT_FLOAT16"
         }
     },
+    "CustLinearSumAssignment":{
+        "input0":{
+            "name":"cost_matrix",
+            "type":"DT_FLOAT16,DT_FLOAT,DT_DOUBLE,DT_INT16,DT_INT32,DT_INT64,DT_INT8,DT_UINT16,DT_UINT32,DT_UINT64,DT_UINT8,DT_BOOL"
+        },
+        "input1":{
+            "name":"dimension_limit",
+            "type":"DT_INT64"
+        },
+        "input2":{
+            "name":"maximize",
+            "type":"DT_BOOL"
+        },
+        "opInfo":{
+            "computeCost":"100",
+            "engine":"DNN_VM_AICPU",
+            "flagAsync":"False",
+            "flagPartial":"False",
+            "formatAgnostic":"False",
+            "functionName":"RunCpuKernel",
+            "kernelSo":"libcust_cpu_kernels.so",
+            "opKernelLib":"CUSTAICPUKernel",
+            "opsFlag":"OPS_FLAG_OPEN",
+            "subTypeOfInferShape":"1",
+            "userDefined":"True"
+        },
+        "output0":{
+            "name":"row_ind",
+            "type":"DT_INT64"
+        },
+        "output1":{
+            "name":"col_ind",
+            "type":"DT_INT64"
+        }
+    },
     "CustListDiff":{
         "input0":{
             "name":"x",
@@ -6225,6 +6284,30 @@
             "type":"DT_INT32,DT_INT64"
         }
     },
+    "GenerateEodMask":{
+        "input0":{
+            "name":"inputs_ids",
+            "type":"DT_INT32,DT_DOUBLE,DT_FLOAT,DT_FLOAT16,DT_BF16"
+        },
+        "opInfo":{
+            "computeCost":"100",
+            "engine":"DNN_VM_AICPU",
+            "flagAsync":"False",
+            "flagPartial":"False",
+            "formatAgnostic":"False",
+            "functionName":"RunCpuKernel",
+            "kernelSo":"libcust_cpu_kernels.so",
+            "opKernelLib":"CUSTAICPUKernel",
+            "opsFlag":"OPS_FLAG_CLOSE",
+            "subTypeOfInferShape":"1",
+            "userDefined":"True",
+            "workspaceSize":"1024"
+        },
+        "output0":{
+            "name":"position_ids",
+            "type":"DT_INT32,DT_DOUBLE,DT_FLOAT,DT_FLOAT16,DT_BF16"
+        }
+    },
     "Geqrf":{
         "input0":{
             "name":"x",
@@ -6604,6 +6687,41 @@
             "type":"DT_DOUBLE,DT_FLOAT,DT_FLOAT16"
         }
     },
+    "LinearSumAssignment":{
+        "input0":{
+            "name":"cost_matrix",
+            "type":"DT_FLOAT16,DT_FLOAT,DT_DOUBLE,DT_INT16,DT_INT32,DT_INT64,DT_INT8,DT_UINT16,DT_UINT32,DT_UINT64,DT_UINT8,DT_BOOL"
+        },
+        "input1":{
+            "name":"dimension_limit",
+            "type":"DT_INT64"
+        },
+        "input2":{
+            "name":"maximize",
+            "type":"DT_BOOL"
+        },
+        "opInfo":{
+            "computeCost":"100",
+            "engine":"DNN_VM_AICPU",
+            "flagAsync":"False",
+            "flagPartial":"False",
+            "formatAgnostic":"False",
+            "functionName":"RunCpuKernel",
+            "kernelSo":"libcust_cpu_kernels.so",
+            "opKernelLib":"CUSTAICPUKernel",
+            "opsFlag":"OPS_FLAG_OPEN",
+            "subTypeOfInferShape":"1",
+            "userDefined":"True"
+        },
+        "output0":{
+            "name":"row_ind",
+            "type":"DT_INT64"
+        },
+        "output1":{
+            "name":"col_ind",
+            "type":"DT_INT64"
+        }
+    },
     "ListDiff":{
         "input0":{
             "name":"x",

mindspore/mindrecord/tools/cifar100_to_mr.py

@@ -18,7 +18,6 @@ Cifar100 convert tool for MindRecord.
 
 from importlib import import_module
 import os
-import numpy as np
 
 from mindspore import log as logger
 from .cifar100 import Cifar100
@@ -26,10 +25,6 @@ from ..common.exceptions import PathNotExistsError
 from ..filewriter import FileWriter
 from ..shardutils import check_filename, ExceptionThread, SUCCESS, FAILED
 
-try:
-    cv_import = import_module("cv2")
-except ModuleNotFoundError:
-    cv_import = None
 
 __all__ = ['Cifar100ToMR']
 
@@ -57,6 +52,8 @@ class Cifar100ToMR:
     """
 
     def __init__(self, source, destination):
+        self.cv_import = import_module("cv2")
+
         check_filename(source)
         self.source = source
 
@@ -100,12 +97,12 @@ class Cifar100ToMR:
         test_coarse_labels = cifar100_data.Test.coarse_labels
         logger.info("test images coarse label: {}".format(coarse_labels.shape))
 
-        data_list = _construct_raw_data(images, fine_labels, coarse_labels)
-        test_data_list = _construct_raw_data(test_images, test_fine_labels, test_coarse_labels)
+        data_list = self._construct_raw_data(images, fine_labels, coarse_labels)
+        test_data_list = self._construct_raw_data(test_images, test_fine_labels, test_coarse_labels)
 
-        if _generate_mindrecord(self.destination, data_list, fields, "img_train") != SUCCESS:
+        if self._generate_mindrecord(self.destination, data_list, fields, "img_train") != SUCCESS:
             return FAILED
-        if _generate_mindrecord(self.destination + "_test", test_data_list, fields, "img_test") != SUCCESS:
+        if self._generate_mindrecord(self.destination + "_test", test_data_list, fields, "img_test") != SUCCESS:
             return FAILED
         return SUCCESS
 
@@ -142,57 +139,52 @@ class Cifar100ToMR:
             raise t.exception
         return t.res
 
+    def _construct_raw_data(self, images, fine_labels, coarse_labels):
+        """
+        Construct raw data from cifar100 data.
 
-def _construct_raw_data(images, fine_labels, coarse_labels):
-    """
-    Construct raw data from cifar100 data.
+        Args:
+            images (list): image list from cifar100.
+            fine_labels (list): fine label list from cifar100.
+            coarse_labels (list): coarse label list from cifar100.
 
-    Args:
-        images (list): image list from cifar100.
-        fine_labels (list): fine label list from cifar100.
-        coarse_labels (list): coarse label list from cifar100.
+        Returns:
+            list[dict], data dictionary constructed from cifar100.
+        """
+        raw_data = []
+        for i, img in enumerate(images):
+            fine_label = fine_labels[i][0]
+            coarse_label = coarse_labels[i][0]
+            _, img = self.cv_import.imencode(".jpeg", img[..., [2, 1, 0]])
+            row_data = {"id": int(i),
+                        "data": img.tobytes(),
+                        "fine_label": int(fine_label),
+                        "coarse_label": int(coarse_label)}
+            raw_data.append(row_data)
+        return raw_data
+
+    def _generate_mindrecord(self, file_name, raw_data, fields, schema_desc):
+        """
+        Generate MindRecord file from raw data.
 
-    Returns:
-        list[dict], data dictionary constructed from cifar100.
-    """
-    if not cv_import:
-        raise ModuleNotFoundError("opencv-python module not found, please use pip install it.")
-
-    raw_data = []
-    for i, img in enumerate(images):
-        fine_label = np.int(fine_labels[i][0])
-        coarse_label = np.int(coarse_labels[i][0])
-        _, img = cv_import.imencode(".jpeg", img[..., [2, 1, 0]])
-        row_data = {"id": int(i),
-                    "data": img.tobytes(),
-                    "fine_label": int(fine_label),
-                    "coarse_label": int(coarse_label)}
-        raw_data.append(row_data)
-    return raw_data
-
-
-def _generate_mindrecord(file_name, raw_data, fields, schema_desc):
-    """
-    Generate MindRecord file from raw data.
+        Args:
+            file_name (str): File name of MindRecord File.
+            fields (list[str]): Fields would be set as index which
+            could not belong to blob fields and type could not be 'array' or 'bytes'.
+            raw_data (dict): Dict of raw data.
+            schema_desc (str): String of schema description.
 
-    Args:
-        file_name (str): File name of MindRecord File.
-        fields (list[str]): Fields would be set as index which
-          could not belong to blob fields and type could not be 'array' or 'bytes'.
-        raw_data (dict): Dict of raw data.
-        schema_desc (str): String of schema description.
-
-    Returns:
-        MSRStatus, SUCCESS or FAILED.
-    """
-    schema = {"id": {"type": "int64"}, "fine_label": {"type": "int64"},
-              "coarse_label": {"type": "int64"}, "data": {"type": "bytes"}}
+        Returns:
+            MSRStatus, SUCCESS or FAILED.
+        """
+        schema = {"id": {"type": "int64"}, "fine_label": {"type": "int64"},
+                  "coarse_label": {"type": "int64"}, "data": {"type": "bytes"}}
 
-    logger.info("transformed MindRecord schema is: {}".format(schema))
+        logger.info("transformed MindRecord schema is: {}".format(schema))
 
-    writer = FileWriter(file_name, 1)
-    writer.add_schema(schema, schema_desc)
-    if fields and isinstance(fields, list):
-        writer.add_index(fields)
-    writer.write_raw_data(raw_data)
-    return writer.commit()
+        writer = FileWriter(file_name, 1)
+        writer.add_schema(schema, schema_desc)
+        if fields and isinstance(fields, list):
+            writer.add_index(fields)
+        writer.write_raw_data(raw_data)
+        return writer.commit()

mindspore/mindrecord/tools/cifar10_to_mr.py

@@ -18,7 +18,6 @@ Cifar10 convert tool for MindRecord.
 
 from importlib import import_module
 import os
-import numpy as np
 
 from mindspore import log as logger
 from .cifar10 import Cifar10
@@ -26,10 +25,6 @@ from ..common.exceptions import PathNotExistsError
 from ..filewriter import FileWriter
 from ..shardutils import check_filename, ExceptionThread, SUCCESS, FAILED
 
-try:
-    cv_import = import_module("cv2")
-except ModuleNotFoundError:
-    cv_import = None
 
 __all__ = ['Cifar10ToMR']
 
@@ -57,6 +52,8 @@ class Cifar10ToMR:
     """
 
     def __init__(self, source, destination):
+        self.cv_import = import_module("cv2")
+
         check_filename(source)
         self.source = source
 
@@ -96,12 +93,12 @@ class Cifar10ToMR:
         test_labels = cifar10_data.Test.labels
         logger.info("test images label: {}".format(test_labels.shape))
 
-        data_list = _construct_raw_data(images, labels)
-        test_data_list = _construct_raw_data(test_images, test_labels)
+        data_list = self._construct_raw_data(images, labels)
+        test_data_list = self._construct_raw_data(test_images, test_labels)
 
-        if _generate_mindrecord(self.destination, data_list, fields, "img_train") != SUCCESS:
+        if self._generate_mindrecord(self.destination, data_list, fields, "img_train") != SUCCESS:
             return FAILED
-        if _generate_mindrecord(self.destination + "_test", test_data_list, fields, "img_test") != SUCCESS:
+        if self._generate_mindrecord(self.destination + "_test", test_data_list, fields, "img_test") != SUCCESS:
             return FAILED
         return SUCCESS
 
@@ -137,56 +134,50 @@ class Cifar10ToMR:
             raise t.exception
         return t.res
 
+    def _construct_raw_data(self, images, labels):
+        """
+        Construct raw data from cifar10 data.
 
-def _construct_raw_data(images, labels):
-    """
-    Construct raw data from cifar10 data.
-
-    Args:
-        images (list): image list from cifar10.
-        labels (list): label list from cifar10.
-
-    Returns:
-        list[dict], data dictionary constructed from cifar10.
-    """
-
-    if not cv_import:
-        raise ModuleNotFoundError("opencv-python module not found, please use pip install it.")
-
-    raw_data = []
-    for i, img in enumerate(images):
-        label = np.int(labels[i][0])
-        _, img = cv_import.imencode(".jpeg", img[..., [2, 1, 0]])
-        row_data = {"id": int(i),
-                    "data": img.tobytes(),
-                    "label": int(label)}
-        raw_data.append(row_data)
-    return raw_data
+        Args:
+            images (list): image list from cifar10.
+            labels (list): label list from cifar10.
 
+        Returns:
+            list[dict], data dictionary constructed from cifar10.
+        """
+        raw_data = []
+        for i, img in enumerate(images):
+            label = labels[i][0]
+            _, img = self.cv_import.imencode(".jpeg", img[..., [2, 1, 0]])
+            row_data = {"id": int(i),
+                        "data": img.tobytes(),
+                        "label": int(label)}
+            raw_data.append(row_data)
+        return raw_data
+
+    def _generate_mindrecord(self, file_name, raw_data, fields, schema_desc):
+        """
+        Generate MindRecord file from raw data.
 
-def _generate_mindrecord(file_name, raw_data, fields, schema_desc):
-    """
-    Generate MindRecord file from raw data.
+        Args:
+            file_name (str): File name of MindRecord File.
+            fields (list[str]): Fields would be set as index which
+            could not belong to blob fields and type could not be 'array' or 'bytes'.
+            raw_data (dict): dict of raw data.
+            schema_desc (str): String of schema description.
 
-    Args:
-        file_name (str): File name of MindRecord File.
-        fields (list[str]): Fields would be set as index which
-          could not belong to blob fields and type could not be 'array' or 'bytes'.
-        raw_data (dict): dict of raw data.
-        schema_desc (str): String of schema description.
-
-    Returns:
-        MSRStatus, SUCCESS or FAILED.
-    """
+        Returns:
+            MSRStatus, SUCCESS or FAILED.
+        """
 
-    schema = {"id": {"type": "int64"}, "label": {"type": "int64"},
-              "data": {"type": "bytes"}}
+        schema = {"id": {"type": "int64"}, "label": {"type": "int64"},
+                  "data": {"type": "bytes"}}
 
-    logger.info("transformed MindRecord schema is: {}".format(schema))
+        logger.info("transformed MindRecord schema is: {}".format(schema))
 
-    writer = FileWriter(file_name, 1)
-    writer.add_schema(schema, schema_desc)
-    if fields and isinstance(fields, list):
-        writer.add_index(fields)
-    writer.write_raw_data(raw_data)
-    return writer.commit()
+        writer = FileWriter(file_name, 1)
+        writer.add_schema(schema, schema_desc)
+        if fields and isinstance(fields, list):
+            writer.add_index(fields)
+        writer.write_raw_data(raw_data)
+        return writer.commit()

mindspore/mindrecord/tools/csv_to_mr.py

@@ -22,10 +22,6 @@ from mindspore import log as logger
 from ..filewriter import FileWriter
 from ..shardutils import check_filename, ExceptionThread
 
-try:
-    pd = import_module("pandas")
-except ModuleNotFoundError:
-    pd = None
 
 __all__ = ['CsvToMR']
 
@@ -55,8 +51,7 @@ class CsvToMR:
     """
 
     def __init__(self, source, destination, columns_list=None, partition_number=1):
-        if not pd:
-            raise Exception("Module pandas is not found, please use pip install it.")
+        self.pd = import_module("pandas")
         if isinstance(source, str):
             check_filename(source)
             self.source = source
@@ -135,8 +130,8 @@ class CsvToMR:
         if not os.path.exists(self.source):
             raise IOError("Csv file {} do not exist.".format(self.source))
 
-        pd.set_option('display.max_columns', None)
-        df = pd.read_csv(self.source)
+        self.pd.set_option('display.max_columns', None)
+        df = self.pd.read_csv(self.source)
 
         csv_schema = self._get_schema(df)
 

mindspore/mindrecord/tools/mnist_to_mr.py

@@ -25,10 +25,6 @@ from mindspore import log as logger
 from ..filewriter import FileWriter
 from ..shardutils import check_filename, ExceptionThread, SUCCESS, FAILED
 
-try:
-    cv_import = import_module("cv2")
-except ModuleNotFoundError:
-    cv_import = None
 
 __all__ = ['MnistToMR']
 
@@ -58,6 +54,8 @@ class MnistToMR:
     """
 
     def __init__(self, source, destination, partition_number=1):
+        self.cv_import = import_module("cv2")
+
         self.image_size = 28
         self.num_channels = 1
 
@@ -89,9 +87,6 @@ class MnistToMR:
 
     # pylint: disable=missing-docstring
     def run(self):
-        if not cv_import:
-            raise ModuleNotFoundError("opencv-python module not found, please use pip install it.")
-
         if self._transform_train() == FAILED:
             return FAILED
         if self._transform_test() == FAILED:
@@ -155,7 +150,7 @@ class MnistToMR:
         train_data = self._extract_images(self.train_data_filename_)
         train_labels = self._extract_labels(self.train_labels_filename_)
         for data, label in zip(train_data, train_labels):
-            _, img = cv_import.imencode(".jpeg", data)
+            _, img = self.cv_import.imencode(".jpeg", data)
             yield {"label": int(label), "data": img.tobytes()}
 
     def _mnist_test_iterator(self):
@@ -168,7 +163,7 @@ class MnistToMR:
         test_data = self._extract_images(self.test_data_filename_)
         test_labels = self._extract_labels(self.test_labels_filename_)
         for data, label in zip(test_data, test_labels):
-            _, img = cv_import.imencode(".jpeg", data)
+            _, img = self.cv_import.imencode(".jpeg", data)
             yield {"label": int(label), "data": img.tobytes()}
 
     def _transform_train(self):

mindspore/mindrecord/tools/tfrecord_to_mr.py

@@ -97,10 +97,7 @@ class TFRecordToMR:
     """
 
     def __init__(self, source, destination, feature_dict, bytes_fields=None):
-        try:
-            self.tf = import_module("tensorflow")  # just used to convert tfrecord to mindrecord
-        except ModuleNotFoundError:
-            raise Exception("Module tensorflow is not found, please use pip install it.")
+        self.tf = import_module("tensorflow")  # just used to convert tfrecord to mindrecord
 
         if self.tf.__version__ < SupportedTensorFlowVersion:
             raise Exception("Module tensorflow version must be greater or equal {}.".format(SupportedTensorFlowVersion))

mindspore/nn/layer/activation.py

@@ -782,7 +782,7 @@ class Tanhshrink(Cell):
         ``Ascend`` ``GPU`` ``CPU``
 
     Examples:
-        >>> import mindspore
+        >>> import mindspore as ms
         >>> from mindspore import Tensor, nn
         >>> import numpy as np
         >>> x = Tensor(np.array([1, 2, 3, 2, 1]), ms.float16)

mindspore/nn/layer/embedding.py

@@ -522,12 +522,12 @@ class MultiFieldEmbeddingLookup(EmbeddingLookup):
           this interface. Type is Int32, Int64.
         - **input_values** (Tensor) - The shape of tensor is :math:`(batch\_size, seq\_length)`.
           Specifies the weights of elements of the input_indices. The lookout vector will multiply with
-          the input_values. Type is Float32.
+          the input_values. Type is float32.
         - **field_ids** (Tensor)  - The shape of tensor is :math:`(batch\_size, seq\_length)`.
           Specifies the field id of elements of the input_indices. Type is Int32.
 
     Outputs:
-        Tensor, the shape of tensor is :math:`(batch\_size, field\_size, embedding\_size)`. Type is Float32.
+        Tensor, the shape of tensor is :math:`(batch\_size, field\_size, embedding\_size)`. Type is float32.
 
     Raises:
         TypeError: If `vocab_size` or `embedding_size` or `field_size` is not an int.