paddle定义了一个简单网络，如何用paddle.summary查看网络结构？？

Question

paddle定义了一个简单网络，如何用paddle.summary查看网络结构？？

class LSTM(paddle.nn.Layer):
    def __init__(self, input_size=1, hidden_size=16): 
        super().__init__()
        self.rnn = paddle.nn.LSTM(input_size=input_size, hidden_size=hidden_size,  num_layers=3)
        self.linear = paddle.nn.Linear(hidden_size, 1)

    def forward(self, inputs):
        y, (hidden, cell) = self.rnn(inputs)
        output = self.linear(hidden[-1])
   
     
        return output         

Answer 1

import paddle

class LSTM(paddle.nn.Layer):
    def __init__(self, input_size=1, hidden_size=16):
        super().__init__()
        self.rnn = paddle.nn.LSTM(input_size=input_size, hidden_size=hidden_size, num_layers=3)
        self.linear = paddle.nn.Linear(hidden_size, 1)

    def forward(self, inputs):
        y, (hidden, cell) = self.rnn(inputs)
        output = self.linear(hidden[-1])
        return output

input_size = 12  # 任意可修改
time_steps = 2  # 任意可修改
model = LSTM(input_size)
summary = paddle.summary(model, (None, time_steps, input_size))
print(summary)

Answer 2

summary第一个参数为网络模型，第二个参数为模型输入形状，用元组表示，包括batch_size(用1就行)

import paddle
import paddle.nn as nn

class LeNet(nn.Layer):
    def __init__(self, num_classes=10):
        super(LeNet, self).__init__()
        self.num_classes = num_classes
        self.features = nn.Sequential(
            nn.Conv2D(
                1, 6, 3, stride=1, padding=1),
            nn.ReLU(),
            nn.MaxPool2D(2, 2),
            nn.Conv2D(
                6, 16, 5, stride=1, padding=0),
            nn.ReLU(),
            nn.MaxPool2D(2, 2))

        if num_classes > 0:
            self.fc = nn.Sequential(
                nn.Linear(400, 120),
                nn.Linear(120, 84),
                nn.Linear(
                    84, 10))

    def forward(self, inputs):
        x = self.features(inputs)

        if self.num_classes > 0:
            x = paddle.flatten(x, 1)
            x = self.fc(x)
        return x

lenet = LeNet()

params_info = paddle.summary(lenet, (1, 1, 28, 28))
print(params_info)

Answer 3

参考paddle官网链接：

summary-API文档-PaddlePaddle深度学习平台 summary 函数通过 input_size 或 input 打印网络 net 的基础结构和参数信息。input_size 指定网络 net 输入 Tensor 的大小，而 input 指定网络 n https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/summary_cn.html#summary

import paddle
import paddle.nn as nn

class LeNet(nn.Layer):
    def __init__(self, num_classes=10):
        super(LeNet, self).__init__()
        self.num_classes = num_classes
        self.features = nn.Sequential(
            nn.Conv2D(
                1, 6, 3, stride=1, padding=1),
            nn.ReLU(),
            nn.MaxPool2D(2, 2),
            nn.Conv2D(
                6, 16, 5, stride=1, padding=0),
            nn.ReLU(),
            nn.MaxPool2D(2, 2))

        if num_classes > 0:
            self.fc = nn.Sequential(
                nn.Linear(400, 120),
                nn.Linear(120, 84),
                nn.Linear(
                    84, 10))

    def forward(self, inputs):
        x = self.features(inputs)

        if self.num_classes > 0:
            x = paddle.flatten(x, 1)
            x = self.fc(x)
        return x

lenet = LeNet()

params_info = paddle.summary(lenet, (1, 1, 28, 28))
print(params_info)
# ---------------------------------------------------------------------------
# Layer (type)       Input Shape          Output Shape         Param #
# ===========================================================================
# Conv2D-11      [[1, 1, 28, 28]]      [1, 6, 28, 28]          60
#     ReLU-11       [[1, 6, 28, 28]]      [1, 6, 28, 28]           0
# MaxPool2D-11     [[1, 6, 28, 28]]      [1, 6, 14, 14]           0
# Conv2D-12      [[1, 6, 14, 14]]     [1, 16, 10, 10]         2,416
#     ReLU-12      [[1, 16, 10, 10]]     [1, 16, 10, 10]           0
# MaxPool2D-12    [[1, 16, 10, 10]]      [1, 16, 5, 5]            0
# Linear-16         [[1, 400]]            [1, 120]           48,120
# Linear-17         [[1, 120]]            [1, 84]            10,164
# Linear-18         [[1, 84]]             [1, 10]              850
# ===========================================================================
# Total params: 61,610
# Trainable params: 61,610
# Non-trainable params: 0
# ---------------------------------------------------------------------------
# Input size (MB): 0.00
# Forward/backward pass size (MB): 0.11
# Params size (MB): 0.24
# Estimated Total Size (MB): 0.35
# ---------------------------------------------------------------------------
# {'total_params': 61610, 'trainable_params': 61610}