mindspore2022/tests/ut/python/train/dataset.py

# Copyright 2020 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.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
"""dataset base and LeNet."""
import os

from mindspore import dataset as ds
from mindspore.common import dtype as mstype
from mindspore.dataset.transforms import c_transforms as C
from mindspore.dataset.vision import Inter
from mindspore.dataset.vision import c_transforms as CV
from mindspore import nn, Tensor
from mindspore.common.initializer import Normal
from mindspore.ops import operations as P

def create_mnist_dataset(mode='train', num_samples=2, batch_size=2):
    """create dataset for train or test"""
    mnist_path = '/home/workspace/mindspore_dataset/mnist'
    num_parallel_workers = 1

    # define dataset
    mnist_ds = ds.MnistDataset(os.path.join(mnist_path, mode), num_samples=num_samples, shuffle=False)

    resize_height, resize_width = 32, 32

    # define map operations
    resize_op = CV.Resize((resize_height, resize_width), interpolation=Inter.LINEAR)  # Bilinear mode
    rescale_nml_op = CV.Rescale(1 / 0.3081, -1 * 0.1307 / 0.3081)
    rescale_op = CV.Rescale(1.0 / 255.0, shift=0.0)
    hwc2chw_op = CV.HWC2CHW()
    type_cast_op = C.TypeCast(mstype.int32)

    # apply map operations on images
    mnist_ds = mnist_ds.map(operations=type_cast_op, input_columns="label", num_parallel_workers=num_parallel_workers)
    mnist_ds = mnist_ds.map(operations=resize_op, input_columns="image", num_parallel_workers=num_parallel_workers)
    mnist_ds = mnist_ds.map(operations=rescale_op, input_columns="image", num_parallel_workers=num_parallel_workers)
    mnist_ds = mnist_ds.map(operations=rescale_nml_op, input_columns="image", num_parallel_workers=num_parallel_workers)
    mnist_ds = mnist_ds.map(operations=hwc2chw_op, input_columns="image", num_parallel_workers=num_parallel_workers)

    # apply DatasetOps
    mnist_ds = mnist_ds.batch(batch_size=batch_size, drop_remainder=True)

    return mnist_ds


class LeNet5(nn.Cell):
    """
    Lenet network

    Args:
        num_class (int): Number of classes. Default: 10.
        num_channel (int): Number of channels. Default: 1.

    Returns:
        Tensor, output tensor
    Examples:
        >>> LeNet(num_class=10)

    """

    def __init__(self, num_class=10, num_channel=1, include_top=True):
        super(LeNet5, self).__init__()
        self.conv1 = nn.Conv2d(num_channel, 6, 5, pad_mode='valid')
        self.conv2 = nn.Conv2d(6, 16, 5, pad_mode='valid')
        self.relu = nn.ReLU()
        self.max_pool2d = nn.MaxPool2d(kernel_size=2, stride=2)
        self.include_top = include_top
        if self.include_top:
            self.flatten = nn.Flatten()
            self.fc1 = nn.Dense(16 * 5 * 5, 120, weight_init=Normal(0.02))
            self.fc2 = nn.Dense(120, 84, weight_init=Normal(0.02))
            self.fc3 = nn.Dense(84, num_class, weight_init=Normal(0.02))

        self.scalar_summary = P.ScalarSummary()
        self.image_summary = P.ImageSummary()
        self.histogram_summary = P.HistogramSummary()
        self.tensor_summary = P.TensorSummary()
        self.channel = Tensor(num_channel)

    def construct(self, x):
        """construct."""
        self.image_summary('image', x)
        x = self.conv1(x)
        self.histogram_summary('histogram', x)
        x = self.relu(x)
        self.tensor_summary('tensor', x)
        x = self.relu(x)
        x = self.max_pool2d(x)
        self.scalar_summary('scalar', self.channel)
        x = self.conv2(x)
        x = self.relu(x)
        x = self.max_pool2d(x)
        if not self.include_top:
            return x
        x = self.flatten(x)
        x = self.relu(self.fc1(x))
        x = self.relu(self.fc2(x))
        x = self.fc3(x)
        return x