mleveryday/practicalai-cn

★ 6,866Jupyter NotebookAudience · dataComplexity · 2/5LicenseSetup · easy

Mindmap

mindmap
  root((practicalai-cn))
    Curriculum tiers
      Basics Python NumPy
      Deep learning PyTorch
      Advanced GANs LLMs
    Tools
      Jupyter Notebook
      Google Colab
      PyTorch
    Topics
      CNNs and RNNs
      Embeddings attention
      Recommendation systems
    Audience
      Chinese ML learners
      Beginners

mindmap root((practicalai-cn)) Curriculum tiers Basics Python NumPy Deep learning PyTorch Advanced GANs LLMs Tools Jupyter Notebook Google Colab PyTorch Topics CNNs and RNNs Embeddings attention Recommendation systems Audience Chinese ML learners Beginners

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Work through a structured ML curriculum in Chinese from Python basics to GANs and pretrained language models.

USE CASE 2

Run any notebook instantly in Google Colab without installing anything locally using a Google account.

USE CASE 3

Study production-quality, object-oriented ML code patterns rather than toy examples.

USE CASE 4

Use individual notebooks as focused references for specific topics like CNNs, RNNs, or embeddings.

Tech stack

PythonJupyter NotebookPyTorchNumPyPandasGoogle Colab

Getting it running

Difficulty · easy Time to first run · 5min

Use freely for any purpose, including commercial use, as long as you keep the copyright notice.

In plain English

practicalAI-cn is a Chinese-language translation of the practicalAI machine learning course, presented as a collection of interactive Jupyter Notebooks. The project covers a wide range of topics, from foundational programming tools like Python, NumPy, and Pandas, through to neural networks, convolutional networks, recurrent networks, and computer vision. The goal is to teach machine learning in a way that emphasizes writing production-quality, object-oriented code rather than just running toy examples. All notebooks are designed to run inside Google Colab, a free browser-based environment provided by Google. This means a learner needs no local software installation: open the notebook link, sign in with a Google account, copy it to your own Drive, and run it. The README explains exactly how to convert a GitHub notebook URL into a Colab URL, or how to use a Chrome extension to do it in one click. The curriculum is organized into three rough tiers. The basics section covers Python, NumPy, Pandas, linear regression, logistic regression, random forests, and k-means clustering. The deep learning section introduces PyTorch, multilayer networks, convolutional neural networks, recurrent neural networks, and embeddings. An advanced section covers topics like attention-based RNNs, computer vision, autoencoders, generative adversarial networks, recommendation systems, and pretrained language models. Some advanced notebooks are listed in the table but not yet translated. The project is a fork and translation effort with multiple volunteer contributors, each credited in the README. The original course was created by GokuMohandas. Contributions of new or corrected translations can be submitted through GitHub by uploading revised notebooks as pull requests. The project uses an MIT license.

Copy-paste prompts

Prompt 1

Using the practicalAI-cn PyTorch notebook, help me rewrite this NumPy neural network in PyTorch with proper object-oriented class structure.

Prompt 2

Based on the practicalAI-cn curriculum, create a 3-month study plan starting from Python basics and ending at attention-based RNNs.

Prompt 3

Show me how to convert a practicalAI-cn GitHub notebook URL into a Google Colab URL so I can run it in my browser.

Prompt 4

Help me adapt the practicalAI-cn logistic regression notebook to classify my own CSV dataset using PyTorch.

Prompt 5

Using the practicalAI-cn GAN notebook as a starting point, explain how a generative adversarial network trains and help me run the first example.

Open on GitHub → Explain another repo

← mleveryday on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.