mindspore/tests/st/pynative/test_pynative_hook.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.
# ============================================================================

import pytest
import numpy as np
import mindspore.nn as nn
import mindspore.common.dtype as mstype

from mindspore import Tensor
from mindspore import context
from mindspore import ParameterTuple
from mindspore.nn import Momentum
from mindspore.nn import WithLossCell
from mindspore.ops import composite as C
from mindspore.ops import operations as P
from mindspore.common.initializer import TruncatedNormal

context.set_context(mode=context.PYNATIVE_MODE, device_target="Ascend")


grad_all = C.GradOperation(get_all=True)


def weight_variable():
    """weight initial"""
    return TruncatedNormal(0.02)


def conv(in_channels, out_channels, kernel_size, stride=1, padding=0):
    """weight initial for conv layer"""
    weight = weight_variable()
    return nn.Conv2d(in_channels, out_channels,
                     kernel_size=kernel_size, stride=stride, padding=padding,
                     weight_init=weight, has_bias=False, pad_mode="valid")


def fc_with_initialize(input_channels, out_channels):
    """weight initial for fc layer"""
    weight = weight_variable()
    bias = weight_variable()
    return nn.Dense(input_channels, out_channels, weight, bias)


class test_custom_hook_function_base():
    def __init__(self):
        pass

    def test_custom_hook_function(self, hook_function, cell_hook_function):
        return hook_function, cell_hook_function


def cell_hook_function_print_grad(cell_id, grad_input, grad_output):
    assert grad_output[0].asnumpy().shape == (32, 6, 14, 14)
    assert grad_input[0].asnumpy().shape == (32, 16, 10, 10)


def custom_hook_function_print_and_save_grad(grad_out):
    assert grad_out[0].asnumpy().shape == (32, 6, 28, 28)


class LeNet5(nn.Cell):
    def __init__(self, hook_function, cell_hook_function, num_class=10):
        super(LeNet5, self).__init__()
        self.num_class = num_class
        self.batch_size = 32
        self.conv1 = conv(1, 6, 5)
        self.conv2 = conv(6, 16, 5)
        self.conv1.register_backward_hook(cell_hook_function)
        self.fc1 = fc_with_initialize(16 * 5 * 5, 120)
        self.fc2 = fc_with_initialize(120, 84)
        self.fc3 = fc_with_initialize(84, self.num_class)
        self.relu = nn.ReLU()
        self.max_pool2d = nn.MaxPool2d(kernel_size=2, stride=2)
        self.reshape = P.Reshape()
        self.hook = P.HookBackward(hook_function)

    def construct(self, x):
        x = self.conv1(x)
        x = self.relu(x)
        x = self.hook(x)
        x = self.max_pool2d(x)
        x = self.conv2(x)
        x = self.relu(x)
        x = self.max_pool2d(x)
        x = self.reshape(x, (self.batch_size, -1))
        x = self.fc1(x)
        x = self.relu(x)
        x = self.fc2(x)
        x = self.relu(x)
        x = self.fc3(x)
        return x


class GradWrap(nn.Cell):
    """ GradWrap definition """
    def __init__(self, network):
        super(GradWrap, self).__init__(auto_prefix=False)
        self.network = network
        self.weights = ParameterTuple(filter(lambda x: x.requires_grad, network.get_parameters()))

    def construct(self, x, label):
        weights = self.weights
        return C.GradOperation(get_by_list=True)(self.network, weights)(x, label)


class test_custom_cell_base():
    def __init__(self):
        pass

    def test_custom_cell_function(self, cell):
        return cell


class MulAdd(nn.Cell):
    def construct(self, x, y):
        return 2 * x + y

    def bprop(self, x, y, out, dout):
        assert x.asnumpy() == 1.0
        assert y.asnumpy() == 2.0
        assert out.asnumpy() == 4.0
        assert dout.asnumpy() == 1.0
        return dout, y


class Ms_Cell(nn.Cell):
    def __init__(self):
        super(Ms_Cell, self).__init__()
        self.relu = P.ReLU()

    def construct(self, x):
        return self.relu(x)

    def bprop(self, x, out, dout):
        dout = Tensor(np.ones([5, 5]).astype(np.float32))
        assert dout.shape == (5, 5)
        return dout


@pytest.mark.level0
@pytest.mark.platform_arm_ascend_training
@pytest.mark.platform_x86_ascend_training
@pytest.mark.env_onecard
def test_pynative_lenet_train_hook_function_print_and_save_grad():
    hook = test_custom_hook_function_base()
    function = hook.test_custom_hook_function(custom_hook_function_print_and_save_grad,
                                              cell_hook_function_print_grad)
    net = LeNet5(hook_function=function[0], cell_hook_function=function[1])
    optimizer = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()), 0.1, 0.9)
    criterion = nn.SoftmaxCrossEntropyWithLogits(is_grad=False, sparse=False)
    net_with_criterion = WithLossCell(net, criterion)
    train_network = GradWrap(net_with_criterion)
    train_network.set_train()

    input_data = Tensor(np.ones([net.batch_size, 1, 32, 32]).astype(np.float32) * 0.01)
    label = Tensor(np.ones([net.batch_size, net.num_class]).astype(np.float32))
    output = net(Tensor(input_data))
    criterion(output, label)
    grads = train_network(input_data, label)
    success = optimizer(grads)
    assert success


@pytest.mark.level0
@pytest.mark.platform_arm_ascend_training
@pytest.mark.platform_x86_ascend_training
@pytest.mark.env_onecard
def test_pynative_custom_bprop_and_Cell_MulAdd():
    custom_cell = test_custom_cell_base()
    mul_add = custom_cell.test_custom_cell_function(MulAdd())
    mul_add.bprop_debug = True
    grad_all(mul_add)(Tensor(1, mstype.float32), Tensor(2, mstype.float32))
    assert grad_all(mul_add)(Tensor(1, mstype.float32), Tensor(2, mstype.float32)) == \
           (Tensor(1.0, mstype.float32), Tensor(2.0, mstype.float32))


@pytest.mark.level0
@pytest.mark.platform_arm_ascend_training
@pytest.mark.platform_x86_ascend_training
@pytest.mark.env_onecard
def test_pynative_custom_bprop_and_Cell_Ms_Cell():
    custom_cell = test_custom_cell_base()
    ms_Cell = custom_cell.test_custom_cell_function(Ms_Cell())
    ms_Cell.bprop_debug = True
    assert grad_all(ms_Cell)(Tensor(1, mstype.float32)) == (Tensor(1.0, mstype.float32),)