目錄
- 概述
- 代碼實(shí)現(xiàn)
- 手動定義求導(dǎo)
- 計(jì)算流量
- 反向傳播計(jì)算
- 線性回歸
- 導(dǎo)包
- 構(gòu)造 x, y
- 構(gòu)造模型
- 參數(shù) 損失函數(shù)
- 訓(xùn)練模型
- 完整代碼
概述
PyTorch 干的最厲害的一件事情就是幫我們把反向傳播全部計(jì)算好了.
代碼實(shí)現(xiàn)
手動定義求導(dǎo)
import torch
# 方法一
x = torch.randn(3, 4, requires_grad=True)
# 方法二
x = torch.randn(3,4)
x.requires_grad = True
b = torch.randn(3, 4, requires_grad=True)
t = x + b
y = t.sum()
print(y)
print(y.backward())
print(b.grad)
print(x.requires_grad)
print(b.requires_grad)
print(t.requires_grad)
輸出結(jié)果:
tensor(1.1532, grad_fn=SumBackward0>)
None
tensor([[1., 1., 1., 1.],
[1., 1., 1., 1.],
[1., 1., 1., 1.]])
True
True
True
計(jì)算流量
# 計(jì)算流量
x = torch.rand(1)
w = torch.rand(1, requires_grad=True)
b = torch.rand(1, requires_grad=True)
y = w * x
z = y + b
print(x.requires_grad, w.requires_grad,b.requires_grad, z.requires_grad)
print(x.is_leaf, w.is_leaf, b.is_leaf, y.is_leaf,z.is_leaf)
輸出結(jié)果:
False True True True
True True True False False
反向傳播計(jì)算
# 反向傳播
z.backward(retain_graph= True) # 如果不清空會累加起來
print(w.grad)
print(b.grad)
輸出結(jié)果:
tensor([0.1485])
tensor([1.])
線性回歸
導(dǎo)包
import numpy as np
import torch
import torch.nn as nn
構(gòu)造 x, y
# 構(gòu)造數(shù)據(jù)
X_values = [i for i in range(11)]
X_train = np.array(X_values, dtype=np.float32)
X_train = X_train.reshape(-1, 1)
print(X_train.shape) # (11, 1)
y_values = [2 * i + 1 for i in X_values]
y_train = np.array(y_values, dtype=np.float32)
y_train = y_train.reshape(-1,1)
print(y_train.shape) # (11, 1)
輸出結(jié)果:
(11, 1)
(11, 1)
構(gòu)造模型
# 構(gòu)造模型
class LinerRegressionModel(nn.Module):
def __init__(self, input_dim, output_dim):
super(LinerRegressionModel, self).__init__()
self.liner = nn.Linear(input_dim, output_dim)
def forward(self, x):
out = self.liner(x)
return out
input_dim = 1
output_dim = 1
model = LinerRegressionModel(input_dim, output_dim)
print(model)
輸出結(jié)果:
LinerRegressionModel(
(liner): Linear(in_features=1, out_features=1, bias=True)
)
參數(shù) 損失函數(shù)
# 超參數(shù)
enpochs = 1000
learning_rate = 0.01
# 損失函數(shù)
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)
criterion = nn.MSELoss()
訓(xùn)練模型
# 訓(xùn)練模型
for epoch in range(enpochs):
# 轉(zhuǎn)成tensor
inputs = torch.from_numpy(X_train)
labels = torch.from_numpy(y_train)
# 梯度每次迭代清零
optimizer.zero_grad()
# 前向傳播
outputs = model(inputs)
# 計(jì)算損失
loss = criterion(outputs, labels)
# 反向傳播
loss.backward()
# 更新參數(shù)
optimizer.step()
if epoch % 50 == 0:
print("epoch {}, loss {}".format(epoch, loss.item()))
輸出結(jié)果:
epoch 0, loss 114.47456359863281
epoch 50, loss 0.00021522105089388788
epoch 100, loss 0.00012275540211703628
epoch 150, loss 7.001651829341426e-05
epoch 200, loss 3.9934264350449666e-05
epoch 250, loss 2.2777328922529705e-05
epoch 300, loss 1.2990592040296178e-05
epoch 350, loss 7.409254521917319e-06
epoch 400, loss 4.227155841363128e-06
epoch 450, loss 2.410347860859474e-06
epoch 500, loss 1.3751249525739695e-06
epoch 550, loss 7.844975016269018e-07
epoch 600, loss 4.4756839656656666e-07
epoch 650, loss 2.5517596213830984e-07
epoch 700, loss 1.4577410922811396e-07
epoch 750, loss 8.30393886985803e-08
epoch 800, loss 4.747753479250605e-08
epoch 850, loss 2.709844615367274e-08
epoch 900, loss 1.5436164346738224e-08
epoch 950, loss 8.783858973515635e-09
完整代碼
import numpy as np
import torch
import torch.nn as nn
# 構(gòu)造數(shù)據(jù)
X_values = [i for i in range(11)]
X_train = np.array(X_values, dtype=np.float32)
X_train = X_train.reshape(-1, 1)
print(X_train.shape) # (11, 1)
y_values = [2 * i + 1 for i in X_values]
y_train = np.array(y_values, dtype=np.float32)
y_train = y_train.reshape(-1,1)
print(y_train.shape) # (11, 1)
# 構(gòu)造模型
class LinerRegressionModel(nn.Module):
def __init__(self, input_dim, output_dim):
super(LinerRegressionModel, self).__init__()
self.liner = nn.Linear(input_dim, output_dim)
def forward(self, x):
out = self.liner(x)
return out
input_dim = 1
output_dim = 1
model = LinerRegressionModel(input_dim, output_dim)
print(model)
# 超參數(shù)
enpochs = 1000
learning_rate = 0.01
# 損失函數(shù)
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)
criterion = nn.MSELoss()
# 訓(xùn)練模型
for epoch in range(enpochs):
# 轉(zhuǎn)成tensor
inputs = torch.from_numpy(X_train)
labels = torch.from_numpy(y_train)
# 梯度每次迭代清零
optimizer.zero_grad()
# 前向傳播
outputs = model(inputs)
# 計(jì)算損失
loss = criterion(outputs, labels)
# 反向傳播
loss.backward()
# 更新參數(shù)
optimizer.step()
if epoch % 50 == 0:
print("epoch {}, loss {}".format(epoch, loss.item()))
到此這篇關(guān)于PyTorch一小時(shí)掌握之a(chǎn)utograd機(jī)制篇的文章就介紹到這了,更多相關(guān)PyTorch autograd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！
您可能感興趣的文章:- PyTorch一小時(shí)掌握之神經(jīng)網(wǎng)絡(luò)氣溫預(yù)測篇
- PyTorch一小時(shí)掌握之神經(jīng)網(wǎng)絡(luò)分類篇
- PyTorch一小時(shí)掌握之圖像識別實(shí)戰(zhàn)篇
- PyTorch一小時(shí)掌握之基本操作篇
