PyTorch——P1 Getting started with PyTorch

Realization of two-layer neural network by Torch

import torch

N, D_in, H, D_out = 64, 1000, 100, 10

# create train date of random
x = torch.randn(N, D_in).to("cuda:0")
y = torch.randn(N, D_out).to("cuda:0")

w1 = torch.randn(D_in, H).to("cuda:0")
w2 = torch.randn(H, D_out).to("cuda:0")

learning_rate = 1e-6
for it in range(5000):
    # Forward pass
    h = x.mm(w1)   # N * H
    h_relu = h.clamp(min=0)    # N * H
    y_pred = h_relu.mm(w2)     # N * D_out

    # compute loss
    loss = (y_pred - y).pow(2).sum().item()
    print(it, loss)

    # backward pass
    # compute the gradient
    grad_y_pred = 2.0 * (y_pred - y)
    grad_w2 = h_relu.t().mm(grad_y_pred)
    grad_h_relu = grad_y_pred.mm(w2.T)
    grad_h = grad_h_relu.clone()
    grad_h[h<0] = 0
    grad_w1 = x.t().mm(grad_h)

    # update weights of w1 and w2
    w1 -= learning_rate * grad_w1
    w2 -= learning_rate * grad_w2

Gradient descent with Torch

import torch
import numpy as np

N, D_in, H, D_out = 64, 1000, 100, 10

# create train date of random
x = torch.randn(N, D_in)
y = torch.randn(N, D_out)

w1 = torch.randn(D_in, H, requires_grad=True)
w2 = torch.randn(H, D_out, requires_grad=True)

learning_rate = 1e-6
for it in range(5000):
    # Forward pass
    y_pred = x.mm(w1).clamp(min=0).mm(w2)

    # compute loss
    loss = (y_pred - y).pow(2).sum()    # computation graph
    print(it, loss.item())

    # backward pass
    loss.backward()

    # update weights of w1 and w2
    with torch.no_grad():
        w1 -= learning_rate * w1.grad
        w2 -= learning_rate * w2.grad
        w1.grad.zero_()
        w2.grad.zero_()

chage optimizer

import torch

N, D_in, H, D_out = 64, 1000, 100, 10

# create train date of random
x = torch.randn(N, D_in)
y = torch.randn(N, D_out)

model = torch.nn.Sequential(
    torch.nn.Linear(D_in, H),
    torch.nn.ReLU(),
    torch.nn.Linear(H, D_out)
)

loss_fn = torch.nn.MSELoss(reduction="sum")
learning_rate = 1e-4
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

for it in range(5000):
    # Forward pass
    y_pred = model(x)

    # compute loss
    loss = loss_fn(y_pred, y)   # computation graph
    print(it, loss.item())

    optimizer.zero_grad()
    # backward pass
    loss.backward()

    # update model parameters
    optimizer.step()

import torch

N, D_in, H, D_out = 64, 1000, 100, 10

# create train date of random
x = torch.randn(N, D_in)
y = torch.randn(N, D_out)

class TwoLayerNet(torch.nn.Module):
    def __init__(self, D_in, H, D_out):
        super(TwoLayerNet, self).__init__()
        #define the model architecture
        self.linear1 = torch.nn.Linear(D_in, H)
        self.linear2 = torch.nn.Linear(H, D_out)

    def forward(self, x):
        y_pred = self.linear2(self.linear1(x).clamp(min=0))
        return y_pred

model = TwoLayerNet(D_in, H, D_out)

loss_fn = torch.nn.MSELoss(reduction="sum")
learning_rate = 1e-4
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

for it in range(5000):
    # Forward pass
    y_pred = model(x)

    # compute loss
    loss = loss_fn(y_pred, y)   # computation graph
    print(it, loss.item())

    optimizer.zero_grad()
    # backward pass
    loss.backward()

    # update model parameters
    optimizer.step()

define input and output
define model
define loss function
define optimizer
use input and model predict output
compute loss
backward pass
up model parameters

最后编辑于：2020.10.30 11:03:30