Location>code7788 >text

Learning Artificial Intelligence from Zero - Python-Pytorch Learning (XI)

Popularity:441 ℃/2024-12-12 23:47:21

preamble

This article focuses on the use of tensorboard.
tensorboard is a visualization that supports artificial intelligence learning.
The official address of tensorboard:/tensorboard
The content of this article is from the video tutorial 16 lessons, I personally feel that for tensorboard talk very good.

Use of Tensorboard

Use the code below:

import torch
import as nn
import torchvision
import as transforms
import as plt
import sys
import as F
from import SummaryWriter

# pip install tensorboard 安装 tensorboard
# 启动 tensorboard 启动成功的话,The address ishttp://localhost:6006/
# logdirbe tantamount to SummaryWriter('runs/mnist1')input address
# tensorboard --logdir=C:\Project\python_test\github\PythonTest\PythonTest\PythonTest\pytorchTutorial\runs
# tensorboardofficial address:/tensorboard

############## TENSORBOARD ########################
writer = SummaryWriter('runs/mnist1')
###################################################

# Device configuration
device = ('cuda' if .is_available() else 'cpu')

# Hyper-parameters
input_size = 784 # 28x28
hidden_size = 500
num_classes = 10
num_epochs = 1
batch_size = 64
learning_rate = 0.001

# MNIST dataset
train_dataset = (root='./data',
                                           train=True,
                                           transform=(),
                                           download=True)

test_dataset = (root='./data',
                                          train=False,
                                          transform=())

# Data loader
train_loader = (dataset=train_dataset,
                                           batch_size=batch_size,
                                           shuffle=True)

test_loader = (dataset=test_dataset,
                                          batch_size=batch_size,
                                          shuffle=False)

examples = iter(test_loader)
example_data, example_targets = next(examples)

for i in range(6):
    (2,3,i+1)
    (example_data[i][0], cmap='gray')
#()

############## TENSORBOARD ########################
img_grid = .make_grid(example_data)
writer.add_image('mnist_images', img_grid)
#()
#()
###################################################

# Fully connected neural network with one hidden layer
class NeuralNet():
    def __init__(self, input_size, hidden_size, num_classes):
        super(NeuralNet, self).__init__()
        self.input_size = input_size
        self.l1 = (input_size, hidden_size)
         = ()
        self.l2 = (hidden_size, num_classes)
    
    def forward(self, x):
        out = self.l1(x)
        out = (out)
        out = self.l2(out)
        # no activation and no softmax at the end
        return out

model = NeuralNet(input_size, hidden_size, num_classes).to(device)

# Loss and optimizer
criterion = ()
optimizer = ((), lr=learning_rate)

############## TENSORBOARD ########################
writer.add_graph(model, example_data.reshape(-1, 28*28).to(device))
#()
#()
###################################################

# Train the model
running_loss = 0.0
running_correct = 0
n_total_steps = len(train_loader)
for epoch in range(num_epochs):
    for i, (images, labels) in enumerate(train_loader):  
        # origin shape: [100, 1, 28, 28]
        # resized: [100, 784]
        images = (-1, 28*28).to(device)
        labels = (device)
        
        # Forward pass
        outputs = model(images)
        loss = criterion(outputs, labels)
        
        # Backward and optimize
        optimizer.zero_grad()
        ()
        ()
        
        running_loss += ()

        _, predicted = (, 1)
        running_correct += (predicted == labels).sum().item()
        if (i+1) % 100 == 0:
            print (f'Epoch [{epoch+1}/{num_epochs}], Step [{i+1}/{n_total_steps}], Loss: {():.4f}')
            ############## TENSORBOARD ########################
            writer.add_scalar('training loss', running_loss / 100, epoch * n_total_steps + i)
            running_accuracy = running_correct / 100 / (0)
            writer.add_scalar('accuracy', running_accuracy, epoch * n_total_steps + i)
            running_correct = 0
            running_loss = 0.0
            ###################################################

# Test the model
# In test phase, we don't need to compute gradients (for memory efficiency)
class_labels = []
class_preds = []
with torch.no_grad():
    n_correct = 0
    n_samples = 0
    for images, labels in test_loader:
        images = (-1, 28*28).to(device)
        labels = (device)
        outputs = model(images)
        # max returns (value ,index)
        values, predicted = (, 1)
        n_samples += (0)
        n_correct += (predicted == labels).sum().item()

        class_probs_batch = [(output, dim=0) for output in outputs]

        class_preds.append(class_probs_batch)
        class_labels.append(labels)

    # 10000, 10, and 10000, 1
    # stack concatenates tensors along a new dimension
    # cat concatenates tensors in the given dimension
    class_preds = ([(batch) for batch in class_preds])
    class_labels = (class_labels)

    acc = 100.0 * n_correct / n_samples
    print(f'Accuracy of the network on the 10000 test images: {acc} %')

    ############## TENSORBOARD ########################
    classes = range(10)
    for i in classes:
        labels_i = class_labels == i
        preds_i = class_preds[:, i]
        writer.add_pr_curve(str(i), labels_i, preds_i, global_step=0)
        ()
    ###################################################

(of a computer) runhttp://localhost:6006 , you can get the following graph, and you can analyze the learning results based on the curves and other information in the graph.

image
image


Portal:
Learning Artificial Intelligence from Zero - Python-Pytorch Learning - Full Episode


Note: This post is original, please contact the author for authorization and attribution for any form of reproduction!



If you think this article is still good, please click [Recommend] below, thank you very much!

/kiba/p/18601612