3種Python查看神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)的方法小結(jié)
1. 網(wǎng)絡(luò)結(jié)構(gòu)代碼
import torch
import torch.nn as nn
# 定義Actor-Critic模型
class ActorCritic(nn.Module):
def __init__(self, state_dim, action_dim):
super(ActorCritic, self).__init__()
self.actor = nn.Sequential(
# 全連接層，輸入維度為 state_dim，輸出維度為 256
nn.Linear(state_dim, 64),
nn.ReLU(),
nn.Linear(64, action_dim),
# Softmax 函數(shù)，將輸出轉(zhuǎn)換為概率分布，dim=-1 表示在最后一個(gè)維度上應(yīng)用 Softmax
nn.Softmax(dim=-1)
)
self.critic = nn.Sequential(
nn.Linear(state_dim, 64),
nn.ReLU(),
nn.Linear(64, 1)
)
def forward(self, state):
policy = self.actor(state)
value = self.critic(state)
return policy, value
# 參數(shù)設(shè)置
state_dim = 1
action_dim = 2
model = ActorCritic(state_dim, action_dim)
2. 查看結(jié)構(gòu)
2.1 直接打印模型
print(model)
輸出：
ActorCritic(
(actor): Sequential(
(0): Linear(in_features=1, out_features=64, bias=True)
(1): ReLU()
(2): Linear(in_features=64, out_features=2, bias=True)
(3): Softmax(dim=-1)
)
(critic): Sequential(
(0): Linear(in_features=1, out_features=64, bias=True)
(1): ReLU()
(2): Linear(in_features=64, out_features=1, bias=True)
)
)
2.2 可視化網(wǎng)絡(luò)結(jié)構(gòu)（需要安裝 torchviz 包）
安裝 torchsummary 包：
$ pip install torchsummary
python 代碼：
from torchviz import make_dot
# 創(chuàng)建一個(gè)虛擬輸入
x = torch.randn(1, state_dim)
# 生成計(jì)算圖
dot = make_dot(model(x), params=dict(model.named_parameters()))
dot.render("actor_critic_model", format="png") # 保存為PNG圖片
輸出 actor_critic_model
digraph {
graph [size="12,12"]
node [align=left fontname=monospace fontsize=10 height=0.2 ranksep=0.1 shape=box style=filled]
140281544075344 [label="
(1, 2)" fillcolor=darkolivegreen1]
140281544213744 [label=SoftmaxBackward0]
140281544213840 -> 140281544213744
140281544213840 [label=AddmmBackward0]
140281544213600 -> 140281544213840
140285722327344 [label="actor.2.bias
(2)" fillcolor=lightblue]
140285722327344 -> 140281544213600
140281544213600 [label=AccumulateGrad]
140281544214032 -> 140281544213840
140281544214032 [label=ReluBackward0]
140281544213984 -> 140281544214032
140281544213984 [label=AddmmBackward0]
140281544214176 -> 140281544213984
140285722327024 [label="actor.0.bias
(64)" fillcolor=lightblue]
140285722327024 -> 140281544214176
140281544214176 [label=AccumulateGrad]
140281544214224 -> 140281544213984
140281544214224 [label=TBackward0]
140281543934832 -> 140281544214224
140285722327264 [label="actor.0.weight
(64, 1)" fillcolor=lightblue]
140285722327264 -> 140281543934832
140281543934832 [label=AccumulateGrad]
140281544213648 -> 140281544213840
140281544213648 [label=TBackward0]
140281544214080 -> 140281544213648
140285722327184 [label="actor.2.weight
(2, 64)" fillcolor=lightblue]
140285722327184 -> 140281544214080
140281544214080 [label=AccumulateGrad]
140281544213744 -> 140281544075344
140285722328704 [label="
(1, 1)" fillcolor=darkolivegreen1]
140281544213888 [label=AddmmBackward0]
140281544214368 -> 140281544213888
140285722328064 [label="critic.2.bias
(1)" fillcolor=lightblue]
140285722328064 -> 140281544214368
140281544214368 [label=AccumulateGrad]
140281544214128 -> 140281544213888
140281544214128 [label=ReluBackward0]
140281544214464 -> 140281544214128
140281544214464 [label=AddmmBackward0]
140281544214512 -> 140281544214464
140285722327424 [label="critic.0.bias
(64)" fillcolor=lightblue]
140285722327424 -> 140281544214512
140281544214512 [label=AccumulateGrad]
140281544214560 -> 140281544214464
140281544214560 [label=TBackward0]
140281544214704 -> 140281544214560
140285722327504 [label="critic.0.weight
(64, 1)" fillcolor=lightblue]
140285722327504 -> 140281544214704
140281544214704 [label=AccumulateGrad]
140281544213696 -> 140281544213888
140281544213696 [label=TBackward0]
140281544214272 -> 140281544213696
140285722327584 [label="critic.2.weight
(1, 64)" fillcolor=lightblue]
140285722327584 -> 140281544214272
140281544214272 [label=AccumulateGrad]
140281544213888 -> 140285722328704
}
輸出模型圖片：
2.3 使用 summary 方法（需要安裝 torchsummary 包）
安裝 torchsummary 包：
pip install torchsummary
代碼：
from torchsummary import summary
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
model = model.to(device)
summary(model, input_size=(state_dim,))
#查看模型參數(shù)
print("查看模型參數(shù)：")
for name, param in model.named_parameters():
print(f"Layer: {name} | Size: {param.size()} | Values: {param[:2]}...")
輸出：
cuda:0
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Linear-1 [-1, 64] 128
ReLU-2 [-1, 64] 0
Linear-3 [-1, 2] 130
Softmax-4 [-1, 2] 0
Linear-5 [-1, 64] 128
ReLU-6 [-1, 64] 0
Linear-7 [-1, 1] 65
================================================================
Total params: 451
Trainable params: 451
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.00
Forward/backward pass size (MB): 0.00
Params size (MB): 0.00
Estimated Total Size (MB): 0.00
----------------------------------------------------------------
查看模型參數(shù)：
Layer: actor.0.weight | Size: torch.Size([64, 1]) | Values: tensor([[ 0.7747],
[-0.0440]], device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: actor.0.bias | Size: torch.Size([64]) | Values: tensor([ 0.5995, -0.2155], device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: actor.2.weight | Size: torch.Size([2, 64]) | Values: tensor([[ 0.0373, 0.0851, 0.1000, 0.1060, 0.0387, 0.0479, 0.0127, 0.0696,
0.0388, 0.0033, 0.1173, -0.1195, -0.0830, 0.0186, 0.0063, -0.0863,
-0.0353, 0.0782, -0.0558, 0.0011, -0.0533, 0.1241, 0.0120, -0.0906,
-0.0551, -0.0673, -0.1070, 0.0402, -0.0662, 0.0596, -0.0811, 0.0457,
0.0349, 0.0564, -0.0155, -0.0404, 0.0843, -0.0978, 0.0459, 0.1097,
-0.0858, 0.0736, -0.0067, -0.0756, -0.0363, -0.0525, -0.0426, -0.1087,
-0.0611, 0.0420, -0.1038, 0.0402, 0.0065, -0.1217, -0.0467, 0.0383,
-0.0217, 0.0283, 0.0800, 0.0228, 0.0415, -0.0473, -0.0199, -0.0436],
[-0.1118, -0.0806, -0.0700, -0.0224, 0.0335, -0.0087, 0.0265, -0.1196,
-0.0907, -0.0360, 0.0621, -0.0471, -0.0939, -0.0912, -0.1061, 0.1051,
-0.0592, -0.0757, 0.0758, -0.1082, -0.0317, 0.1208, -0.0279, -0.0693,
0.0920, -0.0318, -0.0476, 0.0236, -0.0761, 0.0591, 0.0862, -0.0712,
0.0156, -0.1073, 0.1133, 0.0039, -0.0191, 0.0605, -0.0686, -0.1202,
0.0962, 0.0581, 0.1145, 0.0741, -0.0993, -0.0987, 0.0939, 0.1006,
0.0773, -0.0756, -0.1096, 0.0156, -0.0599, 0.0857, 0.1005, -0.0618,
0.0474, 0.0066, -0.0531, -0.0479, 0.1136, 0.0356, 0.1169, -0.0023]],
device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: actor.2.bias | Size: torch.Size([2]) | Values: tensor([-0.0039, 0.0937], device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: critic.0.weight | Size: torch.Size([64, 1]) | Values: tensor([[0.5799],
[0.0473]], device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: critic.0.bias | Size: torch.Size([64]) | Values: tensor([ 0.6507, -0.6974], device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: critic.2.weight | Size: torch.Size([1, 64]) | Values: tensor([[ 0.0738, -0.0370, -0.1010, -0.0333, -0.0595, -0.0172, 0.0928, 0.0815,
0.1221, -0.0842, 0.0511, 0.0452, -0.0386, -0.0503, -0.0964, 0.0370,
-0.0341, -0.0693, -0.0845, 0.0424, -0.0491, -0.0439, -0.0443, 0.0203,
0.0960, -0.1178, -0.0836, -0.0144, -0.0576, -0.0851, 0.0461, 0.1160,
0.0120, 0.1180, 0.0255, 0.1047, -0.0398, 0.0786, 0.1143, 0.0806,
0.1125, 0.0267, 0.0534, -0.0318, 0.1125, -0.0727, 0.1169, 0.0120,
-0.0178, -0.0845, 0.0069, 0.0194, 0.1188, 0.0481, 0.1077, -0.0840,
0.1013, 0.0586, -0.0857, -0.0974, -0.0630, 0.0359, -0.0080, -0.0926]],
device='cuda:0', grad_fn=<SliceBackward0>)...
Layer: critic.2.bias | Size: torch.Size([1]) | Values: tensor([0.0621], device='cuda:0', grad_fn=<SliceBackward0>)...
到此這篇關(guān)于3種Python查看神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)的方法小結(jié)的文章就介紹到這了,更多相關(guān)Python查看神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)內(nèi)容請搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家！
相關(guān)文章
這篇文章主要介紹了python通過openpyxl生成Excel文件的方法,實(shí)例分析了openpyxl的安裝與使用技巧,非常具有實(shí)用價(jià)值,需要的朋友可以參考下2015-05-05
Python中常用于時(shí)間的模塊有time、datetime 和 calendar，顧名思義 time 是表示時(shí)間（時(shí)、分、秒、毫秒）等，calendar 是表示日歷時(shí)間的，本章先討論 datetime 模塊,需要的朋友可以參考下2022-06-06
Python語言有許多優(yōu)點(diǎn)，常用于不同的領(lǐng)域，如數(shù)據(jù)科學(xué)、web開發(fā)、自動(dòng)化運(yùn)維等。本文將學(xué)習(xí)如何使用字符串中內(nèi)置的方法來格式化字符串，感興趣的可以了解一下2022-06-06
這篇文章主要介紹了如何在Python中使用Requests庫的相關(guān)資料,Requests庫是Python中常用的第三方庫,用于簡化HTTP請求的發(fā)送和響應(yīng)處理,文中通過代碼介紹的非常詳細(xì),需要的朋友可以參考下2025-02-02
在Numpy中,NaN表示非數(shù)值,Inf表示無窮大,NaN與任何值計(jì)算都是NaN,Inf與0相乘是NaN,其余情況下與Inf運(yùn)算仍為Inf,可以使用np.isnan(), np.isinf(), np.isneginf(), np.isposinf(), np.isfinite()等函數(shù)進(jìn)行批量判別,返回布爾值數(shù)組2024-09-09
pandas之query方法和sample隨機(jī)抽樣操作
這篇文章主要介紹了pandas之query方法和sample隨機(jī)抽樣操作，具有很好的參考價(jià)值，希望對大家有所幫助。一起跟隨小編過來看看吧2021-03-03
解決python?pip安裝第三方模塊報(bào)錯(cuò):error:legacy-install-failure
pip是python的第三方庫管理器,可以根據(jù)所開發(fā)項(xiàng)目的需要,使用pip相關(guān)命令安裝不同庫,下面這篇文章主要給大家介紹了關(guān)于解決python?pip安裝第三方模塊報(bào)錯(cuò):error:?legacy?-?install?-?failure的相關(guān)資料,需要的朋友可以參考下2023-04-04
本文主要將涉及os,glob,docx模塊的綜合應(yīng)用，幫助大家快速批量修改Word文檔樣式實(shí)現(xiàn)辦公自動(dòng)化，感興趣的朋友可以了解下2021-05-05
大家好，本篇文章主要講的是Python中BeautifulSoup模塊詳解，感興趣的同學(xué)趕緊來看一看吧，對你有幫助的話記得收藏一下2022-02-02
這篇文章主要介紹了Django重置migrations文件的方法步驟，小編覺得挺不錯(cuò)的，現(xiàn)在分享給大家，也給大家做個(gè)參考。一起跟隨小編過來看看吧2019-05-05