mindspore 2.3.0__cp310-cp310-win_amd64.whl → 2.4.1__cp310-cp310-win_amd64.whl

mindspore/ops/function/debug_func.py

@@ -16,6 +16,8 @@
 """Operators for debug function."""
 
 from mindspore.ops.operations.debug_ops import Print
+from mindspore.common.tensor import Tensor
+from mindspore.ops import operations as P
 from .._primitive_cache import _get_cache_prim
 
 
@@ -26,6 +28,8 @@ def print_(*input_x):
     Once set, the output will be saved in the file specified by print_file_path.
     :func:`mindspore.parse_print` can be employed to reload the data.
     For more information, please refer to :func:`mindspore.set_context` and :func:`mindspore.parse_print`.
+    In Ascend platform with graph mode, can set environment variables `MS_DUMP_SLICE_SIZE` and `MS_DUMP_WAIT_TIME`
+    to solve operator execution failure when outputting big tensor or outputting tensor intensively.
 
     Note:
         In pynative mode, please use python print function.
@@ -64,6 +68,114 @@ def print_(*input_x):
     return print_op(*input_x)
 
 
-__all__ = ['print_']
+
+def tensordump(file_name, tensor, mode='out'):
+    """
+    Save Tensor in numpy's npy format.
+
+    In Parallel situation, tensordump will dump slice of data at each rank.
+
+    In Ascend platform with graph mode,
+    Your code OpA --> OpB may compiled as OpA --> RedistributionOps --> OpB.
+
+    Note: The redistribution operator is introduced,
+    Due to inter-device communication and shard strategies in the static graph parallel scenario.
+
+    In case of OpA --> OpB, the dump data of OpA's output is equal to OpB's input.
+
+    But in case of OpA --> RedistributionOps --> OpB,
+    The dump data of OpA's output is not equal to OpB's input (Due to the redistribution operators).
+    So the parameter mode is to handle this situation.
+
+    Assuming OpA's output is used as both tensordump's input parameter and OpB's input parameter.
+    Different requirements of saving dump data can be achieved by configuring parameter mode:
+
+    - If the mode is 'out', the dump data contains only OpA's output slice.
+    - If the mode is 'all', the dump data contains both OpA's output slice and OpB's input slice.
+    - If the mode is 'in', the dump data contains only OpB's input slice.
+
+    For mode 'all' or 'in', the input slice npy file format is: id_fileName_cNodeID_dumpMode_rankID.npy.
+
+    For mode 'out' or 'all' the output slice npy file format is: id_filename.npy.
+
+    - id: An auto increment ID.
+    - fileName: Value of the parameter file_name
+      (if parameter file_name is a user-specified path, the value of fileName is the last level of the path).
+    - cNodeID: The cnode ID in ir graph of step_parallel_end.ir.
+    - dumpMode: Value of the parameter mode.
+    - rankID: Logical device id.
+
+    Note:
+        - The operator of tensordump doesn't support in control flow.
+        - If current parallel mode is STAND_ALONE, mode should only be 'out'.
+        - Parameter mode will be set to 'out' if user doesn't configure it.
+        - This function is used for debugging.
+
+    Args:
+        file_name (str): The path of the npy file saves.
+        tensor (Tensor): The tensor that user want to dump.
+        mode (str, optional): Used to control tensordump behavior, available value is one of ['in', 'out', 'all'].
+            Default value is ``out``.
+
+    Raises:
+        TypeError: If `file_name` is not str.
+        TypeError: If `tensor` is not Tensor.
+        TypeError: If `mode` is not str.
+        ValueError: If `mode` is not in one of ['in', 'out', 'all'].
+
+    Supported Platforms:
+        ``Ascend``
+
+    Examples:
+        .. note::
+            Using msrun command to run below example: msrun --worker_num=2 --local_worker_num=2 --master_port=11450
+            --log_dir=msrun_log --join=True --cluster_time_out=300 tensordump_example.py
+
+        >>> import numpy as np
+        >>> import mindspore as ms
+        >>> from mindspore import nn, Tensor, ops, context
+        >>> from mindspore.ops import operations as P
+        >>> from mindspore.communication import init, get_rank
+        >>> init()
+        >>> rank_id = get_rank()
+        >>> dump_path = f'dumps/rank_{rank_id}/mul1_mul2.npy'
+        >>> class Net(nn.Cell):
+        ...     def __init__(self, strategy1, strategy2):
+        ...         super(Net, self).__init__()
+        ...         self.matmul1 = P.MatMul().shard(strategy1)
+        ...         self.matmul2 = P.MatMul().shard(strategy2)
+        ...
+        ...     def construct(self, x, y, b):
+        ...         out1 = self.matmul1(x, y)
+        ...         ops.tensordump(dump_path, out1, 'all')
+        ...         out2 = self.matmul2(out1, b)
+        ...         return out2
+        ...
+        >>> ms.set_context(mode=ms.GRAPH_MODE, save_graphs=2)
+        >>> context.set_auto_parallel_context(parallel_mode='semi_auto_parallel', full_batch=True)
+        >>> strategy1 = ((1, 2), (2, 1))
+        >>> strategy2 = ((1, 2), (2, 1))
+        >>> net = Net(strategy1, strategy2)
+        >>> x = Tensor(0.1 * np.random.randn(64, 64).astype(np.float32))
+        >>> y = Tensor(0.1 * np.random.randn(64, 64).astype(np.float32))
+        >>> b = Tensor(0.1 * np.random.randn(64, 64).astype(np.float32))
+        >>> out = net(x, y, b)
+        >>> print(f"out shape is: {out.shape}")
+        >>> matmul1_output_slice = np.load('0_mul1_mul2.npy')                       # load matmul1's output slice
+        >>> matmul2_input_slice = np.load('1_mul1_mul2_CNode_64_all_rank_0.npy')    # load matmul2's input slice
+    """
+    if not isinstance(file_name, str):
+        raise TypeError(f"Parameter file_name should only be build_in str type but got: {type(file_name)}")
+    if not isinstance(tensor, Tensor):
+        raise TypeError(f"Parameter tensor should only be Tensor type, but got: {type(tensor)}")
+    if not isinstance(mode, str):
+        raise TypeError(f"Parameter mode should only be build_in str type, but got: {type(mode)}")
+    mode_list = ['out', 'in', 'all']
+    if mode not in mode_list:
+        raise ValueError(f"Parameter mode should in {mode_list}, but got {mode}")
+    _tensordump = _get_cache_prim(P.TensorDump)(input_output=mode)
+    return _tensordump(file_name, tensor)
+
+__all__ = ['print_', 'tensordump']
 
 __all__.sort()

mindspore/ops/function/fft_func.py

@@ -13,7 +13,8 @@
 # limitations under the License.
 # ============================================================================
 """Defines Fast Fourier Transform operators with functional form."""
-from mindspore.ops.auto_generate import fft, fft2, fftn, ifft, ifft2, ifftn, fftshift, ifftshift, rfft, irfft
+from mindspore.ops.auto_generate import fft, fft2, fftn, ifft, ifft2, ifftn, fftshift, ifftshift, \
+    rfft, irfft, rfft2, irfft2, rfftn, irfftn, hfft, ihfft, hfft2, ihfft2, hfftn, ihfftn, fftfreq, rfftfreq
 
 __all__ = [
     'fftshift',
@@ -25,7 +26,19 @@ __all__ = [
     'ifft2',
     'ifftn',
     'rfft',
-    'irfft'
+    'irfft',
+    'rfft2',
+    'irfft2',
+    'rfftn',
+    'irfftn',
+    'hfft',
+    'ihfft',
+    'hfft2',
+    'ihfft2',
+    'hfftn',
+    'ihfftn',
+    'fftfreq',
+    'rfftfreq'
 ]
 
 __all__.sort()

mindspore/ops/function/grad/grad_func.py

@@ -22,6 +22,7 @@ from mindspore.common import Tensor
 from mindspore.common import dtype as mstype
 from mindspore.nn.cell import Cell
 from mindspore.nn.grad.cell_grad import _LinearizeInner
+from mindspore.ops.operations.other_ops import stop_gradient_
 from mindspore.ops.primitive import constexpr, _primexpr
 from mindspore.ops.function.array_func import ones, expand_dims, size, reshape, broadcast_to, transpose, zeros
 from mindspore.ops.composite import _Vmap, _Grad, _TaylorOperation, GradOperation
@@ -658,7 +659,7 @@ def _check_jvp_input_v_len(inputs_len, v_len):
 
 def jvp(fn, inputs, v, has_aux=False):
     """
-    Compute the jacobian-vector-product of the given network. `jvp` matches
+    Compute the jacobian-vector-product of the given network. The calculation procedure of JVP can be found in
     `forward-mode differentiation
     <https://www.mindspore.cn/docs/en/master/design/programming_paradigm.html#forward-mode-ad>`_.
 
@@ -1405,7 +1406,7 @@ def stop_gradient(value):
         [[1.4100001 1.6       6.5999994]
          [1.4100001 1.6       6.5999994]]
     """
-    return P.StopGradient()(value)
+    return stop_gradient_(value)
 
 
 __all__ = [