guanjihuan/py.guanjihuan.com
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

0.1.56

Browse Source
This commit is contained in:
guanjihuan
2023-11-29 00:57:24 +08:00
parent 5c1bad3589
commit 4a4873339f
3 changed files with 12 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
PyPI/setup.cfg
View File

@@ -1,7 +1,7 @@
[metadata] [metadata]
# replace with your username: # replace with your username:
name = guan name = guan
version = 0.1.55 version = 0.1.56
author = guanjihuan author = guanjihuan
author_email = guanjihuan@163.com author_email = guanjihuan@163.com
description = An open source python package description = An open source python package

2
PyPI/src/guan.egg-info/PKG-INFO
View File

@@ -1,6 +1,6 @@
Metadata-Version: 2.1 Metadata-Version: 2.1
Name: guan Name: guan
Version: 0.1.55 Version: 0.1.56
Summary: An open source python package Summary: An open source python package
Home-page: https://py.guanjihuan.com Home-page: https://py.guanjihuan.com
Author: guanjihuan Author: guanjihuan

20
PyPI/src/guan/machine_learning.py
View File

@@ -1,13 +1,13 @@
# Module: machine_learning # Module: machine_learning
import guan import guan
model_class = None # 把类定义成全局的，防止保存完整模型时，无法访问函数中的类
# 全连接神经网络模型（包含一个隐藏层） # 全连接神经网络模型（包含一个隐藏层）
@guan.function_decorator @guan.function_decorator
def fully_connected_neural_network_with_one_hidden_layer(input_size=1, hidden_size=10, output_size=1, activation='relu'): def fully_connected_neural_network_with_one_hidden_layer(input_size=1, hidden_size=10, output_size=1, activation='relu'):
import torch import torch
global model_class # 如果在函数中定义模型类，尽量定义成全局的，这样可以防止在保存完整模型到文件时，无法访问函数中的模型类。
class model_class(torch.nn.Module): global model_class_of_fully_connected_neural_network_with_one_hidden_layer
class model_class_of_fully_connected_neural_network_with_one_hidden_layer(torch.nn.Module):
def __init__(self): def __init__(self):
super().__init__() super().__init__()
self.hidden_layer = torch.nn.Linear(input_size, hidden_size) self.hidden_layer = torch.nn.Linear(input_size, hidden_size)
@@ -25,15 +25,15 @@ def fully_connected_neural_network_with_one_hidden_layer(input_size=1, hidden_si
hidden_output = self.hidden_layer(x) hidden_output = self.hidden_layer(x)
output = self.output_layer(hidden_output) output = self.output_layer(hidden_output)
return output return output
model = model_class() model = model_class_of_fully_connected_neural_network_with_one_hidden_layer()
return model return model
# 全连接神经网络模型（包含两个隐藏层） # 全连接神经网络模型（包含两个隐藏层）
@guan.function_decorator @guan.function_decorator
def fully_connected_neural_network_with_two_hidden_layers(input_size=1, hidden_size_1=10, hidden_size_2=10, output_size=1, activation_1='relu', activation_2='relu'): def fully_connected_neural_network_with_two_hidden_layers(input_size=1, hidden_size_1=10, hidden_size_2=10, output_size=1, activation_1='relu', activation_2='relu'):
import torch import torch
global model_class global model_class_of_fully_connected_neural_network_with_two_hidden_layers
class model_class(torch.nn.Module): class model_class_of_fully_connected_neural_network_with_two_hidden_layers(torch.nn.Module):
def __init__(self): def __init__(self):
super().__init__() super().__init__()
self.hidden_layer_1 = torch.nn.Linear(input_size, hidden_size_1) self.hidden_layer_1 = torch.nn.Linear(input_size, hidden_size_1)
@@ -64,15 +64,15 @@ def fully_connected_neural_network_with_two_hidden_layers(input_size=1, hidden_s
output = self.output_layer(hidden_output_2) output = self.output_layer(hidden_output_2)
return output return output
model = model_class() model = model_class_of_fully_connected_neural_network_with_two_hidden_layers()
return model return model
# 全连接神经网络模型（包含三个隐藏层） # 全连接神经网络模型（包含三个隐藏层）
@guan.function_decorator @guan.function_decorator
def fully_connected_neural_network_with_three_hidden_layers(input_size=1, hidden_size_1=10, hidden_size_2=10, hidden_size_3=10, output_size=1, activation_1='relu', activation_2='relu', activation_3='relu'): def fully_connected_neural_network_with_three_hidden_layers(input_size=1, hidden_size_1=10, hidden_size_2=10, hidden_size_3=10, output_size=1, activation_1='relu', activation_2='relu', activation_3='relu'):
import torch import torch
global model_class global model_class_of_fully_connected_neural_network_with_three_hidden_layers
class model_class(torch.nn.Module): class model_class_of_fully_connected_neural_network_with_three_hidden_layers(torch.nn.Module):
def __init__(self): def __init__(self):
super().__init__() super().__init__()
self.hidden_layer_1 = torch.nn.Linear(input_size, hidden_size_1) self.hidden_layer_1 = torch.nn.Linear(input_size, hidden_size_1)
@@ -115,7 +115,7 @@ def fully_connected_neural_network_with_three_hidden_layers(input_size=1, hidden
output = self.output_layer(hidden_output_3) output = self.output_layer(hidden_output_3)
return output return output
model = model_class() model = model_class_of_fully_connected_neural_network_with_three_hidden_layers()
return model return model
# 使用优化器训练模型 # 使用优化器训练模型