mindspore 2.2.10__cp38-none-any.whl → 2.2.14__cp38-none-any.whl

mindspore/scipy/optimize/linear_sum_assignment.py

@@ -1,4 +1,4 @@
-# Copyright 2022 Huawei Technologies Co., Ltd
+# Copyright 2023 Huawei Technologies Co., Ltd
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -15,59 +15,64 @@
 """Linear Sum Assignment"""
 import sys
 from ..ops import LinearSumAssignment
-from ...common import dtype as mstype
-from ..utils import _mstype_check, _dtype_check
 from ... import Tensor
 
 
-def _linear_sum_assignment(cost_matrix, maximize, dimension_limit=Tensor(sys.maxsize)):
-    """
+def linear_sum_assignment(cost_matrix, maximize, dimension_limit=Tensor(sys.maxsize)):
+    r"""
     Solve the linear sum assignment problem.
 
+    The assignment problem is represented as follows:
+
+    .. math::
+        min\sum_{i}^{} \sum_{j}^{} C_{i,j} X_{i,j}
+
+    where :math:`C` is cost matrix, :math:`X_{i,j} = 1` means column :math:`j` is assigned to row :math:`i` .
+
     Args:
-        cost_matrix: 2-D Input Tensor.
-            The cost matrix of the bipartite graph.
-        maximize: 0-D Input bool Tensor.
-            Calculates a maximum weight matching if true.
-        dimension_limit: 0-D Input Tensor.
-            A scalar used to limit the actual size of the 1st dimension. Optimized for
-            padding scenes. Default means no dimension limit.
+        cost_matrix (Tensor): 2-D cost matrix. Tensor of shape :math:`(M, N)` .
+        maximize (bool): Calculate a maximum weight matching if true, otherwise calculate a minimum weight matching.
+        dimension_limit (Tensor, optional): A scalar used to limit the actual size of the 2nd dimension of
+            ``cost_matrix``. Default is ``Tensor(sys.maxsize)``, which means no limitation. The type is 0-D int64
+            Tensor.
 
     Returns:
-        1-D Output Tensors with 'row_indx' and 'col_idx'. An array of row indices and
-        one of corresponding column indices giving the optimal assignment. If specified
-        dimension_limit, padding value at the end would be -1.
+        A tuple of tensors containing 'row_idx' and 'col_idx'.
+
+        - **row_idx** (Tensor) - Row indices of the problem. If `dimension_limit` is given, -1 would be padded at the
+          end. The shape is  :math:`(N, )` , where :math:`N` is the minimum value of `cost_matrix` dimension.
+        - **col_idx** (Tensor) - Column indices of the problem. If `dimension_limit` is given, -1 would be padded at
+          the end. The shape is  :math:`(N, )` , where :math:`N` is the minimum value of `cost_matrix` dimension.
+
+    Raises:
+        TypeError: If the data type of `cost_matrix` is not the type in [float16, float32, float64,
+                   int8, int16, int32, int64, uint8, uint16, uint32, uint64, bool]
+        TypeError: If the type of `maximize` is not bool.
+        TypeError: If the data type of `dimension_limit` is not int64.
+        ValueError: If the rank of `cost_matrix` is not 2.
+
 
     Supported Platforms:
-        ``CPU``
+        ``Ascend`` ``CPU``
 
-     Examples:
+    Examples:
         >>> import mindspore as ms
         >>> import numpy as np
         >>> from mindspore import Tensor
-        >>> from mindspore.scipy.optimize.linear_sum_assignment import _linear_sum_assignment as lsap
-        >>> cost_matrix = Tensor(np.array([[2, 3, 3], [3, 2, 3], [3, 3, 2]])).astype("float64")
+        >>> import mindspore.scipy.optimize.linear_sum_assignment as lsap
+        >>> cost_matrix = Tensor(np.array([[2, 3, 3], [3, 2, 3], [3, 3, 2]])).astype(ms.float64)
         >>> dimension_limit = Tensor(2)
-        >>> maximize = Tensor(False)
+        >>> maximize = False
         >>> a, b = lsap(cost_matrix, maximize, dimension_limit)
         >>> print(a)
-        [[0 1 -1]]
+        [0 1 -1]
         >>> print(b)
-        [[0 1 -1]]
+        [0 1 -1]
         >>> a, b = lsap(cost_matrix, maximize)
         >>> print(a)
-        [[0 1 2]]
+        [0 1 2]
         >>> print(b)
-        [[0 1 2]]
+        [0 1 2]
     """
-    func_name = 'linear_sum_assignment'
-    _mstype_check(func_name, cost_matrix, mstype.TensorType, 'cost_matrix')
-    _mstype_check(func_name, dimension_limit,
-                  mstype.TensorType, 'dimension_limit')
-    _mstype_check(func_name, maximize, mstype.TensorType, 'maximize')
-    _dtype_check(func_name, cost_matrix, [mstype.float32, mstype.float64])
-    _dtype_check(func_name, dimension_limit, [mstype.int64])
-    _dtype_check(func_name, maximize, [mstype.bool_])
-
     solve = LinearSumAssignment()
     return solve(cost_matrix, dimension_limit, maximize)

mindspore/train/callback/_checkpoint.py

@@ -205,7 +205,7 @@ class CheckpointConfig:
         Get the value of steps to save checkpoint.
 
         Returns:
-            Int, steps to save checkpoint.
+            int, steps to save checkpoint.
         """
         return self._save_checkpoint_steps
 
@@ -214,7 +214,7 @@ class CheckpointConfig:
         """Get the value of _save_checkpoint_seconds.
 
         Returns:
-            Int, seconds to save the checkpoint file.
+            int, seconds to save the checkpoint file.
         """
         return self._save_checkpoint_seconds
 
@@ -224,7 +224,7 @@ class CheckpointConfig:
         Get the value of maximum number of checkpoint files can be saved.
 
         Returns:
-            Int, Maximum number of checkpoint files can be saved.
+            int, Maximum number of checkpoint files can be saved.
         """
         return self._keep_checkpoint_max
 
@@ -244,7 +244,7 @@ class CheckpointConfig:
         Get the value of whether to merge and save the split Tensor in the automatic parallel scenario.
 
         Returns:
-            Bool, whether to merge and save the split Tensor in the automatic parallel scenario.
+            bool, whether to merge and save the split Tensor in the automatic parallel scenario.
         """
         return self._integrated_save
 
@@ -254,7 +254,7 @@ class CheckpointConfig:
         Get the value of whether asynchronous execution saves the checkpoint to a file.
 
         Returns:
-            Bool, whether asynchronous execution saves the checkpoint to a file.
+            bool, whether asynchronous execution saves the checkpoint to a file.
         """
         return self._async_save
 
@@ -294,7 +294,7 @@ class CheckpointConfig:
         Get the value of information dict saved to checkpoint file.
 
         Returns:
-            Dict, the information saved to checkpoint file.
+            dict, the information saved to checkpoint file.
         """
         return self._append_dict
 
@@ -304,7 +304,7 @@ class CheckpointConfig:
         Get the value of whether to save map Parameter incrementally.
 
         Returns:
-            Bool, whether to save map Parameter incrementally.
+            bool, whether to save map Parameter incrementally.
         """
         return self._map_param_inc
 
@@ -313,7 +313,7 @@ class CheckpointConfig:
         Get the policy of checkpoint.
 
         Returns:
-            Dict, the information of checkpoint policy.
+            dict, the information of checkpoint policy.
         """
         checkpoint_policy = {'save_checkpoint_steps': self.save_checkpoint_steps,
                              'save_checkpoint_seconds': self.save_checkpoint_seconds,

mindspore/train/callback/_landscape.py

@@ -1,4 +1,4 @@
-# Copyright 2021-2022 Huawei Technologies Co., Ltd
+# Copyright 2021-2023 Huawei Technologies Co., Ltd
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -170,8 +170,7 @@ class SummaryLandscape:
     It can create landscape in PCA direction or random direction by calculating loss.
 
     Note:
-        1. When using SummaryLandscape, you need to run the code in `if __name__ == "__main__"` .
-        2. SummaryLandscape only supports Linux systems.
+        1. SummaryLandscape only supports Linux systems.
 
     Args:
         summary_dir (str): The path of summary is used to save the model weight,

mindspore/train/callback/_summary_collector.py

@@ -1,4 +1,4 @@
-# Copyright 2020-2022 Huawei Technologies Co., Ltd
+# Copyright 2020-2023 Huawei Technologies Co., Ltd
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -77,12 +77,11 @@ class SummaryCollector(Callback):
     SummaryCollector also enables the summary operator to collect data to summary files.
 
     Note:
-        1. When using SummaryCollector, you need to run the code in `if __name__ == "__main__"` .
-        2. Multiple SummaryCollector instances in callback list are not allowed.
-        3. Not all information is collected at the training phase or at the eval phase.
-        4. SummaryCollector always record the data collected by the summary operator.
-        5. SummaryCollector only supports Linux systems.
-        6. The Summary is not supported when compile source with `-s on` option.
+        1. Multiple SummaryCollector instances in callback list are not allowed.
+        2. Not all information is collected at the training phase or at the eval phase.
+        3. SummaryCollector always record the data collected by the summary operator.
+        4. SummaryCollector only supports Linux systems.
+        5. The Summary is not supported when compile source with `-s on` option.
 
     Args:
         summary_dir (str): The collected data will be persisted to this directory.

mindspore/train/dataset_helper.py

@@ -444,6 +444,11 @@ class DatasetHelper:
         # Generally, it works in dynamic shape scenarios.
         return self.iter.get_data_info()
 
+    # pylint: disable=missing-docstring
+    def get_mbuf_queue_size(self):
+        # In sink mode, it returns the element numbers inside mbuf channel.
+        return self.iter.get_mbuf_queue_size()
+
     # pylint: disable=missing-docstring
     def get_send_info(self, run_context):
         # In sink mode, it returns the send information of dataset at this moment.
@@ -526,6 +531,7 @@ class _DatasetIter:
         self.release = dataset.__transfer_dataset__.release
         self.continue_send = dataset.__transfer_dataset__.continue_send
         self.get_data_info = dataset.__transfer_dataset__.get_data_info
+        self.get_mbuf_queue_size = dataset.__transfer_dataset__.get_mbuf_queue_size
         self.get_send_info = dataset.__transfer_dataset__.get_send_info
         if hasattr(dataset.__transfer_dataset__, "_reset"):
             self._reset = dataset.__transfer_dataset__._reset  # pylint: disable=protected-access

mindspore/train/model.py

@@ -22,6 +22,7 @@ import os
 import math
 import copy
 import importlib
+import time
 import numpy as np
 
 import mindspore
@@ -523,6 +524,16 @@ class Model:
                                                                         dataset_sink_mode=True,
                                                                         sink_size=sink_size)
             self._warmup_dataset(epoch, train_dataset, sink_size)
+            if train_dataset.get_init_step() > 0:
+                mbuf_size = train_dataset.__transfer_dataset__.get_mbuf_queue_size()
+                while mbuf_size == 0:
+                    time.sleep(10)
+                    mbuf_size = train_dataset.__transfer_dataset__.get_mbuf_queue_size()
+                    if mbuf_size != 0:
+                        break
+                    logger.warning(f"Failover mode, waiting for dataset recover to specify step, "
+                                   f"current device queue size: {mbuf_size}")
+
             if context.get_auto_parallel_context("pipeline_stages") > 1 and valid_dataset:
                 train_network.add_flags_recursive(is_first_iteration=True)
             for inputs in train_dataset_helper:
@@ -967,7 +978,7 @@ class Model:
             of data will be transferred one by one. The limitation of data transmission per time is 256M.
 
             When dataset_sink_mode is True, the `step_end` method of the instance of Callback will be called at the end
-            of step in PyNative mode， or will be called at the end of epoch in Graph mode.
+            of step in PyNative mode, or will be called at the end of epoch in Graph mode.
 
             If dataset_sink_mode is True, dataset will be bound to this model and cannot be used by other models.
 
@@ -1004,6 +1015,7 @@ class Model:
                                  Default: 0.
 
         Examples:
+            >>> import mindspore as ms
             >>> from mindspore import nn
             >>> from mindspore.train import Model
             >>>
@@ -1515,8 +1527,8 @@ class Model:
                     [ascend_context]
                     rank_table_file = [path_a](storage initial path of the rank table file)
                     [execution_plan]
-                    [op_name1] = data_type:float16 (operator named op_name1 is set to data type Float16)
-                    [op_name2] = data_type:float32 (operator named op_name2 is set to data type Float32)
+                    [op_name1] = data_type:float16 (operator named op_name1 is set to data type float16)
+                    [op_name2] = data_type:float32 (operator named op_name2 is set to data type float32)
 
                 When only the config_path is configured, it is done as follows:
 
@@ -1649,8 +1661,8 @@ class Model:
                     [ascend_context]
                     rank_table_file = [path_a](storage initial path of the rank table file)
                     [execution_plan]
-                    [op_name1] = data_type:float16 (operator named op_name1 is set to data type Float16)
-                    [op_name2] = data_type:float32 (operator named op_name2 is set to data type Float32)
+                    [op_name1] = data_type:float16 (operator named op_name1 is set to data type float16)
+                    [op_name2] = data_type:float32 (operator named op_name2 is set to data type float32)
 
                 When only the config_path is configured, it is done as follows:
 

mindspore/train/serialization.py

@@ -176,7 +176,7 @@ def _update_param(param, new_param, strict_load):
 
 def _type_convert(param, new_param, strict_load):
     """Whether to convert parameter's type during load checkpoint into network."""
-    float_type = (mstype.float16, mstype.float32, mstype.float64)
+    float_type = (mstype.float16, mstype.float32, mstype.float64, mstype.bfloat16)
     int_type = (mstype.int8, mstype.int16, mstype.int32, mstype.int64)
     if not strict_load and ({param.data.dtype, new_param.data.dtype}.issubset(float_type) or
                             {param.data.dtype, new_param.data.dtype}.issubset(int_type)):
@@ -1560,6 +1560,8 @@ def export(net, *inputs, file_name, file_format, **kwargs):
         3. Exporting functions decorated with :func:`mindspore.jit` to mindir format is supported.
         4. When exporting a function decorated with :func:`mindspore.jit`, the function should not involve
            class properties in calculations.
+        5. AIR format is deprecated, and will be removed in a future version, please use other format or use
+           MindSpore Lite to do offline inference.
 
     Args:
         net (Union[Cell, function]): MindSpore network.
@@ -1633,6 +1635,9 @@ def export(net, *inputs, file_name, file_format, **kwargs):
     supported_formats = ['AIR', 'ONNX', 'MINDIR']
     if file_format not in supported_formats:
         raise ValueError(f"For 'export', 'file_format' must be one of {supported_formats}, but got {file_format}.")
+    if file_format == 'AIR':
+        logger.warning("AIR format is deprecated, and will be removed in a future version, please use other format or "
+                       "use MindSpore Lite to do offline inference")
     Validator.check_file_name_by_regular(file_name)
     logger.info("exporting model file:%s format:%s.", file_name, file_format)
 

mindspore/train/summary/_writer_pool.py

@@ -31,7 +31,7 @@ from mindspore.train.summary.writer import LineageWriter, SummaryWriter, ExportW
 
 try:
     from multiprocessing import get_context
-    ctx = get_context('forkserver')
+    ctx = get_context('fork')
 except ValueError:
     import multiprocessing as ctx
 

mindspore/train/summary/summary_record.py

@@ -1,4 +1,4 @@
-# Copyright 2020-2022 Huawei Technologies Co., Ltd
+# Copyright 2020-2023 Huawei Technologies Co., Ltd
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -141,12 +141,11 @@ class SummaryRecord:
     can be added by calling add_value.
 
     Note:
-        1. When using SummaryRecord, you need to run the code in `if __name__ == "__main__"` .
-        2. Make sure to close the SummaryRecord at the end, otherwise the process will not exit.
+        1. Make sure to close the SummaryRecord at the end, otherwise the process will not exit.
            Please see the Example section below to learn how to close properly in two ways.
-        3. Only one SummaryRecord instance is allowed at a time, otherwise it will cause data writing problems.
-        4. SummaryRecord only supports Linux systems.
-        5. The Summary is not supported when compile source with `-s on` option.
+        2. Only one SummaryRecord instance is allowed at a time, otherwise it will cause data writing problems.
+        3. SummaryRecord only supports Linux systems.
+        4. The Summary is not supported when compile source with `-s on` option.
 
     Args:
         log_dir (str): The log_dir is a directory location to save the summary.

mindspore/version.py

@@ -1 +1 @@
-__version__ = '2.2.10'
+__version__ = '2.2.14'

{mindspore-2.2.10.dist-info → mindspore-2.2.14.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mindspore
-Version: 2.2.10
+Version: 2.2.14
 Summary: MindSpore is a new open source deep learning training/inference framework that could be used for mobile, edge and cloud scenarios.
 Home-page: https://www.mindspore.cn
 Download-URL: https://github.com/mindspore-ai/mindspore/tags
@@ -318,10 +318,11 @@ Project stable branches will be in one of the following states:
 
 | **Branch** | **Status**   | **Initial Release Date** | **Next Phase**                         | **EOL Date**|
 |------------|--------------|--------------------------|----------------------------------------|-------------|
+| **r2.2**   | Maintained   | 2023-10-18               | Unmaintained <br> 2024-10-18 estimated |             |
 | **r2.1**   | Maintained   | 2023-07-29               | Unmaintained <br> 2024-07-29 estimated |             |
 | **r2.0**   | Maintained   | 2023-06-15               | Unmaintained <br> 2024-06-15 estimated |             |
 | **r1.10**  | Maintained   | 2023-02-02               | Unmaintained <br> 2024-02-02 estimated |             |
-| **r1.9**   | Maintained   | 2022-10-26               | Unmaintained <br> 2023-10-26 estimated |             |
+| **r1.9**   | End Of Life  | 2022-10-26               |                                        | 2023-10-26  |
 | **r1.8**   | End Of Life  | 2022-07-29               |                                        | 2023-07-29  |
 | **r1.7**   | End Of Life  | 2022-04-29               |                                        | 2023-04-29  |
 | **r1.6**   | End Of Life  | 2022-01-29               |                                        | 2023-01-29  |