!35394 add get_mindir_input api

Merge pull request !35394 from changzherui/add_get_mindir_input

docs/api/api_python/mindspore.rst

@@ -115,17 +115,18 @@ mindspore
 .. mscnautosummary::
     :toctree: mindspore
 
-    mindspore.save_checkpoint
-    mindspore.load_checkpoint
-    mindspore.load_param_into_net
+    mindspore.async_ckpt_thread_status
+    mindspore.build_searched_strategy
+    mindspore.convert_model
     mindspore.export
     mindspore.load
-    mindspore.parse_print
-    mindspore.build_searched_strategy
-    mindspore.merge_sliced_parameter
+    mindspore.load_checkpoint
     mindspore.load_distributed_checkpoint
-    mindspore.async_ckpt_thread_status
+    mindspore.load_param_into_net
+    mindspore.merge_sliced_parameter
+    mindspore.parse_print
     mindspore.restore_group_info_list
+    mindspore.save_checkpoint
 
 调试调优
 ----------

docs/api/api_python/mindspore/mindspore.convert_model.rst

@@ -0,0 +1,21 @@
+mindspore.conver_model
+======================
+
+.. py:class:: mindspore.convert_model(mindir_file, convert_file, file_format)
+
+    将MindIR模型转化为其他格式的模型文件。当前版本仅支持转化成ONNX模型。
+
+    .. note::
+        这是一个实验特性，未来API可能会发生的变化。
+
+    **参数：**
+
+    - **mindir_file** (str) - MindIR模型文件名称。
+    - **convert_file** (str) - 转化后的模型文件名称。
+    - **file_format** (str) - 需要转化的文件格式，当前版本仅支持"ONNX"。
+
+    **异常：**
+
+    - **TypeError** - `mindir_file` 参数不是str类型。
+    - **TypeError** - `convert_file` 参数不是str类型。
+    - **ValueError** - `file_format` 参数的值不是"ONNX"。

docs/api/api_python_en/mindspore.rst

@@ -236,17 +236,18 @@ Serialization
     :nosignatures:
     :template: classtemplate.rst
 
-    mindspore.save_checkpoint
-    mindspore.load_checkpoint
-    mindspore.load_param_into_net
+    mindspore.async_ckpt_thread_status
+    mindspore.build_searched_strategy
+    mindspore.convert_model
     mindspore.export
     mindspore.load
-    mindspore.parse_print
-    mindspore.build_searched_strategy
-    mindspore.merge_sliced_parameter
+    mindspore.load_checkpoint
     mindspore.load_distributed_checkpoint
-    mindspore.async_ckpt_thread_status
+    mindspore.load_param_into_net
+    mindspore.merge_sliced_parameter
+    mindspore.parse_print
     mindspore.restore_group_info_list
+    mindspore.save_checkpoint
 
 JIT
 ---

mindspore/python/mindspore/_checkparam.py

@@ -483,7 +483,7 @@ class Validator:
         """Check whether file name is legitimate."""
         if not isinstance(target, str):
             prim_name = f"For '{prim_name}', the" if prim_name else "The"
-            raise ValueError("{} '{}' must be string, but got {}.".format(prim_name, target, type(target)))
+            raise TypeError("{} '{}' must be string, but got {}.".format(prim_name, target, type(target)))
         if target.endswith("\\") or target.endswith("/"):
             prim_name = f"For '{prim_name}', the" if prim_name else "The"
             raise ValueError(f"{prim_name} '{target}' cannot be a directory path.")

mindspore/python/mindspore/train/__init__.py

@@ -24,7 +24,7 @@ from .amp import build_train_network
 from .loss_scale_manager import LossScaleManager, FixedLossScaleManager, DynamicLossScaleManager
 from .serialization import save_checkpoint, load_checkpoint, load_param_into_net, export, load, parse_print,\
     build_searched_strategy, merge_sliced_parameter, load_distributed_checkpoint, async_ckpt_thread_status,\
-    restore_group_info_list
+    restore_group_info_list, convert_model
 from .callback import Callback, LossMonitor, TimeMonitor, ModelCheckpoint, SummaryCollector, CheckpointConfig, \
     RunContext, LearningRateScheduler, SummaryLandscape, FederatedLearningManager, History, LambdaCallback, \
     ReduceLROnPlateau, EarlyStopping
@@ -35,7 +35,7 @@ from .train_thor import ConvertNetUtils, ConvertModelUtils
 __all__ = ["Model", "DatasetHelper", "amp", "connect_network_with_dataset", "build_train_network", "LossScaleManager",
            "FixedLossScaleManager", "DynamicLossScaleManager", "save_checkpoint", "load_checkpoint",
            "load_param_into_net", "export", "load", "parse_print", "build_searched_strategy", "merge_sliced_parameter",
-           "load_distributed_checkpoint", "async_ckpt_thread_status", "restore_group_info_list"]
+           "load_distributed_checkpoint", "async_ckpt_thread_status", "restore_group_info_list", "convert_model"]
 __all__.extend(callback.__all__)
 __all__.extend(summary.__all__)
 __all__.extend(train_thor.__all__)

mindspore/python/mindspore/train/_utils.py

@@ -22,6 +22,7 @@ from mindspore.common.tensor import Tensor
 from mindspore.common.dtype import dtype_to_nptype, pytype_to_dtype
 from mindspore.common import dtype as mstype
 from mindspore import log as logger
+from mindspore._checkparam import Validator
 from mindspore.common.api import _cell_graph_executor
 from mindspore.train.mind_ir_pb2 import ModelProto as mindir_model
 from mindspore.train.checkpoint_pb2 import Checkpoint
@@ -238,7 +239,8 @@ def read_proto(file_name, proto_format="MINDIR", display_data=False):
     Returns:
         Object, proto object.
     """
-
+    Validator.check_file_name_by_regular(file_name)
+    file_name = os.path.realpath(file_name)
     if proto_format == "MINDIR":
         model = mindir_model()
     elif proto_format == "CKPT":
@@ -253,7 +255,8 @@ def read_proto(file_name, proto_format="MINDIR", display_data=False):
             pb_content = f.read()
             model.ParseFromString(pb_content)
     except BaseException as e:
-        logger.critical("Failed to read the file `%s`, please check the correct of the file.", file_name)
+        logger.critical(f"Failed to phase the file: {file_name} as format: {proto_format},"
+                        f" please check the correct file and format.")
         raise ValueError(e.__str__())
     finally:
         pass

mindspore/python/mindspore/train/callback/_history.py

@@ -52,6 +52,7 @@ class History(Callback):
     def __init__(self):
         super(History, self).__init__()
         self.history = {}
+        self.epoch = None
 
     def begin(self, run_context):
         """

mindspore/python/mindspore/train/serialization.py

@@ -44,6 +44,7 @@ from mindspore.common.api import _cell_graph_executor as _executor
 from mindspore.common.initializer import initializer
 from mindspore.common.parameter import Parameter
 from mindspore.common.tensor import Tensor
+from mindspore.common.initializer import One
 from mindspore.communication.management import get_rank, get_group_size
 from mindspore.compression.export import quant_export
 from mindspore.parallel._cell_wrapper import get_allgather_cell
@@ -51,8 +52,10 @@ from mindspore.parallel._tensor import _load_tensor, _get_tensor_strategy, _get_
 from mindspore.parallel._tensor import _reshape_param_data
 from mindspore.parallel._tensor import _reshape_param_data_with_weight
 from mindspore.parallel._utils import _infer_rank_list, _remove_repeated_slices
+from mindspore.train._utils import read_proto
 from .._c_expression import load_mindir, _encrypt, _decrypt, _is_cipher_file
 
+
 tensor_to_ms_type = {"Int8": mstype.int8, "UInt8": mstype.uint8, "Int16": mstype.int16, "UInt16": mstype.uint16,
                      "Int32": mstype.int32, "UInt32": mstype.uint32, "Int64": mstype.int64, "UInt64": mstype.uint64,
                      "Float16": mstype.float16, "Float32": mstype.float32, "Float64": mstype.float64,
@@ -62,6 +65,10 @@ tensor_to_np_type = {"Int8": np.int8, "UInt8": np.uint8, "Int16": np.int16, "UIn
                      "Int32": np.int32, "UInt32": np.uint32, "Int64": np.int64, "UInt64": np.uint64,
                      "Float16": np.float16, "Float32": np.float32, "Float64": np.float64, "Bool": np.bool_, "str": "U"}
 
+mindir_to_tensor_type = {1: mstype.float32, 2: mstype.uint8, 3: mstype.int8, 4: mstype.uint16,
+                         5: mstype.int16, 6: mstype.int32, 7: mstype.int64, 10: mstype.float16,
+                         11: mstype.float64, 12: mstype.uint32, 13: mstype.uint64}
+
 _ckpt_mutex = Lock()
 
 # unit is KB
@@ -162,7 +169,7 @@ def _save_weight(checkpoint_dir, model_name, iteration, params):
     exist_ckpt_file_list = []
     if os.path.exists(checkpoint_dir):
         for exist_ckpt_name in os.listdir(checkpoint_dir):
-            file_prefix = model_name + "_iteration_"
+            file_prefix = os.path.join(model_name, "_iteration_")
             if exist_ckpt_name.startswith(file_prefix):
                 exist_ckpt_file_list.append(exist_ckpt_name)
 
@@ -1113,7 +1120,7 @@ def _save_mindir(net, file_name, *inputs, **kwargs):
     model.ParseFromString(mindir_stream)
 
     if kwargs.get('dataset'):
-        check_input_data(kwargs['dataset'], data_class=mindspore.dataset.Dataset)
+        check_input_data(kwargs.get('dataset'), data_class=mindspore.dataset.Dataset)
         dataset = kwargs.get('dataset')
         _save_dataset_to_mindir(model, dataset)
 
@@ -1871,3 +1878,85 @@ def _calculation_net_size(net):
         data_total += sys.getsizeof(net_dict[name].data.asnumpy().tobytes()) / 1024
 
     return data_total
+
+
+def _get_mindir_inputs(file_name):
+    """
+    Get MindIR file's inputs.
+
+    Note:
+        1. Parsing encrypted MindIR file is not supported.
+        2. Parsing dynamic shape MindIR file is not supported.
+
+    Args:
+        file_name (str): MindIR file name.
+
+    Returns:
+        Tensor, list(Tensor), the input of MindIR file.
+
+    Raises:
+        TypeError: If the parameter file_name is not `str`.
+        RuntimeError: MindIR's input is not tensor type or has no dims.
+
+    Examples:
+        >>> input_tensor = get_mindir_inputs("lenet.mindir")
+    """
+    Validator.check_file_name_by_regular(file_name)
+    file_name = os.path.realpath(file_name)
+    model = read_proto(file_name)
+    input_tensor = []
+
+    for ele_input in model.graph.input:
+        input_shape = []
+        if not hasattr(ele_input, "tensor") or not hasattr(ele_input.tensor[0], "dims"):
+            raise RuntimeError("MindIR's inputs has no tensor or tensor has no dims, please check MindIR file.")
+
+        for ele_shape in ele_input.tensor[0].dims:
+            input_shape.append(ele_shape)
+        if -1 in input_shape:
+            raise RuntimeError(f"MindIR input's shape is: {input_shape}, dynamic shape is not supported.")
+
+        mindir_type = ele_input.tensor[0].data_type
+        if mindir_type not in mindir_to_tensor_type:
+            raise RuntimeError(f"MindIR input's type: {mindir_type} is not supported.")
+
+        input_type = mindir_to_tensor_type.get(mindir_type)
+        input_tensor.append(Tensor(shape=input_shape, dtype=input_type, init=One()))
+
+    if not input_tensor:
+        logger.warning("The MindIR model has no input, return None.")
+        return None
+    return input_tensor[0] if len(input_tensor) == 1 else input_tensor
+
+
+def convert_model(mindir_file, convert_file, file_format):
+    """
+    Convert mindir model to other format model. Current version only support convert to "ONNX" format.
+
+    Note:
+        This is an experimental function that is subject to change or deletion.
+
+    Args:
+        mindir_file (str): MindIR file name.
+        convert_file (str): Convert model file name.
+        file_format (str): Convert model's format, current version only supports "ONNX".
+
+    Raises:
+        TypeError: If the parameter `mindir_file` is not `str`.
+        TypeError: If the parameter `convert_file` is not `str`.
+        ValueError: If the parameter `file_format` is not "ONNX".
+
+    Examples:
+        >>> convert_model("lenet.mindir", "lenet.onnx", "ONNX")
+    """
+    Validator.check_file_name_by_regular(mindir_file)
+    Validator.check_file_name_by_regular(convert_file)
+    if file_format != "ONNX":
+        raise ValueError(f"For 'convert_model', 'file_format' must be 'ONNX', but got {file_format}.")
+    net_input = _get_mindir_inputs(mindir_file)
+    graph = load(mindir_file)
+    net = nn.GraphCell(graph)
+    if isinstance(net_input, Tensor):
+        export(net, net_input, file_name=convert_file, file_format=file_format)
+    else:
+        export(net, *net_input, file_name=convert_file, file_format=file_format)

tests/ut/python/utils/test_export.py

@@ -17,6 +17,7 @@ import os
 from io import BytesIO
 import numpy as np
 
+import mindspore
 import mindspore.nn as nn
 import mindspore.dataset as ds
 import mindspore.dataset.vision as CV
@@ -29,7 +30,7 @@ from mindspore.common.initializer import TruncatedNormal
 from mindspore.common.parameter import ParameterTuple
 from mindspore.ops import operations as P
 from mindspore.ops import composite as C
-from mindspore.train.serialization import export
+from mindspore.train.serialization import export, _get_mindir_inputs, convert_model
 
 
 def weight_variable():
@@ -108,6 +109,16 @@ class LeNet5(nn.Cell):
         return x
 
 
+class InputNet1(nn.Cell):
+    def construct(self, x):
+        return x
+
+
+class InputNet2(nn.Cell):
+    def construct(self, x, y):
+        return x, y
+
+
 class WithLossCell(nn.Cell):
     def __init__(self, network):
         super(WithLossCell, self).__init__(auto_prefix=False)
@@ -160,6 +171,77 @@ def test_export_lenet_grad_mindir():
     os.remove(verify_name)
 
 
+def test_get_mindir_inputs1():
+    """
+    Feature: Get MindIR input.
+    Description: Test get mindir input.
+    Expectation: Successfully
+    """
+    context.set_context(mode=context.GRAPH_MODE, device_target="Ascend")
+    net = InputNet1()
+    input1 = Tensor(np.zeros([32, 10]).astype(np.float32))
+    file_name = "input1.mindir"
+    export(net, input1, file_name=file_name, file_format='MINDIR')
+    input_tensor = _get_mindir_inputs(file_name)
+    assert os.path.exists(file_name)
+    assert input_tensor.shape == (32, 10)
+    assert input_tensor.dtype == mindspore.float32
+    os.remove(file_name)
+
+
+def test_get_mindir_inputs2():
+    """
+    Feature: Get MindIR input.
+    Description: Test get mindir input.
+    Expectation: Successfully
+    """
+    context.set_context(mode=context.GRAPH_MODE, device_target="Ascend")
+    net = InputNet2()
+    input1 = Tensor(np.zeros(1).astype(np.float16))
+    input2 = Tensor(np.zeros([10, 20]), dtype=mstype.int32)
+    file_name = "input2.mindir"
+    export(net, input1, input2, file_name=file_name, file_format='MINDIR')
+    input_tensor = _get_mindir_inputs(file_name)
+    assert os.path.exists(file_name)
+    assert len(input_tensor) == 2
+    assert input_tensor[0].shape == (1,)
+    assert input_tensor[0].dtype == mindspore.float16
+    assert input_tensor[1].shape == (10, 20)
+    assert input_tensor[1].dtype == mindspore.int32
+    os.remove(file_name)
+
+
+def test_convert_model():
+    """
+    Feature: Convert mindir to onnx.
+    Description: Test convert.
+    Expectation: Successfully
+    """
+    context.set_context(mode=context.GRAPH_MODE, device_target="Ascend")
+    net1 = InputNet1()
+    input1 = Tensor(np.ones([1, 32, 32]).astype(np.float32))
+    mindir_name1 = "lenet1.mindir"
+    export(net1, input1, file_name=mindir_name1, file_format='MINDIR')
+    onnx_name1 = "lenet1.onnx"
+    convert_model(mindir_name1, onnx_name1, "ONNX")
+    assert os.path.exists(mindir_name1)
+    assert os.path.exists(onnx_name1)
+    os.remove(mindir_name1)
+    os.remove(onnx_name1)
+
+    net2 = InputNet2()
+    input1 = Tensor(np.ones(32).astype(np.float32))
+    input2 = Tensor(np.ones([32, 32]).astype(np.float32))
+    mindir_name2 = "lenet2.mindir"
+    export(net2, input1, input2, file_name=mindir_name2, file_format='MINDIR')
+    onnx_name2 = "lenet2.onnx"
+    convert_model(mindir_name2, onnx_name2, "ONNX")
+    assert os.path.exists(mindir_name2)
+    assert os.path.exists(onnx_name2)
+    os.remove(mindir_name2)
+    os.remove(onnx_name2)
+
+
 def test_export_lenet_with_dataset():
     """
     Feature: Export LeNet with data preprocess to MindIR