dl-d2l 1.0.26

dl_d2l

dl_d2l

This repository contains PyTorch implementations of the Deep Learning book "Dive into Deep Learning" (D2L).
It provides code examples, tutorials, and exercises to help you learn deep learning concepts using PyTorch.

Installation

pip install dl_d2l

Code Examples

You can find the code examples in the dl_d2l package. Here is a simple example of how to use it:

from dl_d2l import d2l_torch as d2l

x = d2l.arange(4)
print(x)

# enable_matplotlib_chinese
from dl_d2l.util import matplotlib_util

matplotlib_util.enable_chinese()

# example plot with Chinese characters
import numpy as np
import matplotlib.pyplot as plt

x = np.linspace(-10, 10, 100)
y = x

plt.plot(x, y)
plt.title("中文标题：y = x")
plt.xlabel("横轴（x）")
plt.ylabel("纵轴（y）")
plt.show()

# get_available_device
from dl_d2l.util import device_util

device = device_util.get_available_device()

print(f"Available device: {device}")

# Check if running in Google Colab
from dl_d2l.util import colab_util

if colab_util.is_colab():
    print("Running in Google Colab")
else:
    print("Not running in Google Colab")

# Get base data directory in Google Colab or local environment
import os
from dl_d2l.util import colab_util

base_data_dir = colab_util.get_base_data_dir()
print(f'base data dir: {base_data_dir}')

datasets_dir = os.path.join(base_data_dir, 'ML', 'Datasets')
os.makedirs(datasets_dir, exist_ok=True)
print(f'datasets dir: {datasets_dir}')

# flush_drive in Google Colab
from dl_d2l.util import colab_util

colab_util.flush_drive()

Build & upload to pypi (For Developers)

prerequirement: twine is installed. If not, run the following command to install it:

pip install -U twine

build and upload:

## package
python setup.py sdist

## upload
twine upload dist/*

D2L

For more information about the "Dive into Deep Learning" book, visit the official website.