Pytorch - optimize model parameters

Reference resources ：OPTIMIZING MODEL PARAMETERS

Gradient back propagation algorithm , Update parameters

SGD -> Adam

dataset -> dataloader,train_dataloader Training ,test_dataloader test , iterator

Model inheritance Module class ,__init__(self), Definition layer ,Flatten() Flattening , and Sequential(), An orderly container

forward() function , Forward calculation ,logits Number of output categories ,10 Categories

Hyperparameters, Hyperparameters , Not involved in optimization , Affect the effect of the model

Loss Function, Loss function 、 Objective function , Classification function CrossEntropyLoss, Regression function MSELoss

Optimizer, Optimizer ,SGD, Update the parameters ,torch.optim.SGD(), Update model parameters ,model.parameters()

Before optimization , call optimizer.zero_grad(), Calculate the gradient loss.backward(),optimizer.step() Update all parameters

optimizer.zero_grad()
loss.backward()
optimizer.step()

torch.no_grad() Reasoning , Calculate the accuracy correct

Data set section , To write dataset User defined classes for ,Transformer function ,Collection function ,dataset -> dataloader

model Can replace timm class , call require_grad(False), Freeze parameters

Simple classification or regression tasks ,seq2seq, Training incoming real value , Testing is an autoregressive task

AI Mission , Data sets + Model + Training , When predicting , There is no need to optimize parameters

PyTorch Of torch.autograd(), Automatic differentiation

Don't use zero_grad(), The learning rate needs to be reduced , Actual projects require custom datasets

Sequence modeling , need padding operation , There are some illegal values ,Python Classes and class inheritance , Learning rate + Optimizer

Embedding class , Floating point type , Input one-hot vector , The weight is embedding vector

Source code ：

• Use timm Model replaces the base model
• Convert the grayscale image , Convert to color image

import timm
import torch
from torch import nn
from torch.utils.data import DataLoader
from torchvision import datasets
from torchvision.transforms import ToTensor, Lambda

training_data = datasets.FashionMNIST(
    root="data",
    train=True,
    download=True,
    transform=ToTensor()
)

test_data = datasets.FashionMNIST(
    root="data",
    train=False,
    download=True,
    transform=ToTensor()
)

train_dataloader = DataLoader(training_data, batch_size=64)
test_dataloader = DataLoader(test_data, batch_size=64)

class NeuralNetwork(nn.Module):
    def __init__(self):
        super(NeuralNetwork, self).__init__()
        self.flatten = nn.Flatten()
# self.linear_relu_stack = nn.Sequential(
# nn.Linear(28*28, 512),
# nn.ReLU(),
# nn.Linear(512, 512),
# nn.ReLU(),
# nn.Linear(512, 10),
# )
        self.mobilenetv3 = timm.create_model('mobilenetv3_large_100', num_classes=10, pretrained=True)

    def forward(self, x):
        x = torch.cat([x, x, x], dim=1)  #  Convert gray image to color image 
        logits = self.mobilenetv3(x)
        return logits

model = NeuralNetwork()

learning_rate = 1e-3
batch_size = 64
epochs = 5

# Initialize the loss function
loss_fn = nn.CrossEntropyLoss()

optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)

def train_loop(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)
    for batch, (X, y) in enumerate(dataloader):
        pred = model(X)
        loss = loss_fn(pred, y)

        # Backpropagation
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        if batch % 100 == 0:
            loss, current = loss.item(), batch * len(X)
            print(f"loss: {
      loss:>7f} [{
      current:>5d}/{
      size:>5d}]")


def test_loop(dataloader, model, loss_fn):
    size = len(dataloader.dataset)
    num_batches = len(dataloader)
    test_loss, correct = 0, 0

    with torch.no_grad():
        for X, y in dataloader:
            pred = model(X)
            test_loss += loss_fn(pred, y).item()
            correct += (pred.argmax(1) == y).type(torch.float).sum().item()

    test_loss /= num_batches
    correct /= size
    print(f"Test Error: \n Accuracy: {
      (100*correct):>0.1f}%, Avg loss: {
      test_loss:>8f} \n")

loss_fn = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)

epochs = 10
for t in range(epochs):
    print(f"Epoch {
      t+1}\n-------------------------------")
    train_loop(train_dataloader, model, loss_fn, optimizer)
    test_loop(test_dataloader, model, loss_fn)
print("Done!")