# course-machine-learning

**Repository Path**: gitconomy/course-machine-learning

## Basic Information

- **Project Name**: course-machine-learning
- **Description**: No description available
- **Primary Language**: Python
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-10-26
- **Last Updated**: 2026-10-29

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Machine Learning: Theory, Algorithms, and Python Practice

An undergraduate course resource repository for students in computer science, data science, artificial intelligence, and related disciplines. The course focuses on classical machine learning, integrates Python programming and experimental validation, and introduces selected topics in deep learning and machine learning engineering.

## 1. Course at a Glance

| Item | Description |
| --- | --- |
| Duration | 6 weeks, 24 chapters; 2 hours per chapter |
| Total workload | 60 hours: 48 hours of lectures and case-based instruction, plus 12 hours of laboratory work |
| Organization | Four chapters and one 2-hour integrated lab each week |
| Main tools | Python 3, Jupyter Notebook, NumPy, Pandas, Matplotlib, Scikit-learn, and PyTorch |
| Recurring context | E-commerce customer and order analysis, using clearly sourced datasets suited to each topic; one dataset is not expected to cover every task |

The course begins with data preparation, mathematical foundations, and model training, then moves through supervised and unsupervised learning, deep learning, and applications such as time-series forecasting, recommendation systems, and model deployment.

## 2. Learning Outcomes

By the end of the course, students will be able to:

1. Explain common machine learning tasks, training workflows, generalization, overfitting, and data leakage, and apply appropriate dataset-splitting practices.
2. Describe the core principles and appropriate use cases of linear models, KNN, Naive Bayes, decision trees, SVMs, and ensemble methods.
3. Use Python for preprocessing, feature engineering, model training, cross-validation, metric computation, and hyperparameter optimization.
4. Implement and evaluate clustering, PCA, and probabilistic mixture models, and explain the findings and limitations of unsupervised learning.
5. Explain the basic mechanisms of neural network training, backpropagation, CNNs, and Attention/Transformer architectures, and complete introductory PyTorch experiments.
6. Build a reproducible machine learning workflow for a business problem and report model performance, limitations, and basic ethical risks.

## 3. Laboratories and Assessment

The course plan includes six integrated 2-hour laboratories aligned with the weekly chapters. Use datasets with clear provenance and permission for educational use. For data involving customer identifiers, purchase records, or review text, use anonymized or publicly de-identified data.

| Assessment | Weight | Main criteria |
| --- | ---: | --- |
| Weekly Python laboratories (6) | 40% | Task completion, reproducible code, model results, analytical interpretation, and sound experimental practice |
| Algorithm principles and progress tests | 20% | Core concepts, principles, algorithm suitability, and evaluation methods; two tests worth 10% each are recommended |
| Week 6 capstone project | 40% | Problem and data design, models and experiments, evaluation and reflection, code, and presentation |

Assessment emphasizes students’ ability to explain why an algorithm was selected, how it was evaluated, and what limitations remain. Accuracy alone is not a substitute for comprehensive evaluation.

## 4. Repository Navigation

| Path | Contents |
| --- | --- |
| [`01-syllabus/spec/en/machine-learning-syllabus-en.md`](01-syllabus/spec/en/machine-learning-syllabus-en.md) | English course syllabus |
| [`01-syllabus/spec/cn/machine-learning-syllabus-cn.md`](01-syllabus/spec/cn/machine-learning-syllabus-cn.md) | Chinese course syllabus |
| [`03-labs/README.md`](03-labs/README.md) | Laboratory resource overview and environment setup links |
| [`03-labs/en/ENVIRONMENT_SETUP.md`](03-labs/en/ENVIRONMENT_SETUP.md) | English environment setup guide |
| [`03-labs/cn/ENVIRONMENT_SETUP.md`](03-labs/cn/ENVIRONMENT_SETUP.md) | Chinese environment setup guide |

Chapter resources are organized below. In this repository, “Colab” refers to local Jupyter lab materials.

| Chapter     | Topic                                                   | Lecture Notes                                                                              | Slides                                                                                      | Colab                                                                                                         |
| ----------- | ------------------------------------------------------- | ------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------- |
| Ch01        | Giving Computers the Ability to Learn from Data         | [Notes](02-lectures/notes/ch01-giving-computers-the-ability-to-learning-from-data.md)      | [PPT](02-lectures/slides/ch01-giving-computers-the-ability-to-learning-from-data.pptx)      | [Lab Guide](03-labs/ch01-colab/README.md) · [Notebook](03-labs/ch01-colab/ch01-learning-from-data.ipynb)      |
| Ch02        | From Artificial Neurons to Trainable Linear Classifiers | [Notes](02-lectures/notes/ch02-from-artificial-neurons-to-trainable-linear-classifiers.md) | [PPT](02-lectures/slides/ch02-from-artificial-neurons-to-trainable-linear-classifiers.pptx) | —                                                                                                             |
| Ch03        | A Tour of Machine Learning Classifiers                  | [Notes](02-lectures/notes/ch03-ml-classifiers-principles-decisions-and-scikit-learn.md)    | [PPT](02-lectures/slides/ch03-ml-classifiers-principles-decisions-and-scikit-learn.pptx)    | —                                                                                                             |
| Ch04        | Building Good Training Datasets                         | [Notes](02-lectures/notes/ch04-building-good-training-datasets.md)                         | [PPT](02-lectures/slides/ch04-building-good-training-datasets.pptx)                         | —                                                                                                             |

The repository currently includes lecture notes and PPT slides for Ch01–Ch04, along with Week 1 Lab01 materials in English and Chinese. The syllabus covers all 24 chapters; additional chapter and lab resources will be added to the repository as they become available.

## 5. Getting Started

1. Read the course syllabus in your preferred language to review the roadmap and prerequisites.
2. Set up Python and the course environment using the guides linked from [`03-labs/README.md`](03-labs/README.md).
3. Start with the Ch01–Ch04 lecture notes and matching PPT slides, then use the Lab01 resources for practice.
4. Before running a Notebook, check the chapter README, data documentation, and requirements file. Dependencies may vary by chapter.

Recommended prerequisites include basic Python syntax, functions, and data structures, plus matrix operations, derivatives, and introductory probability and statistics. Week 1 reviews or supplements essential concepts but does not replace a full foundational mathematics course.

## License

The repository code is licensed under the terms in [LICENSE](LICENSE). Course materials may have separate licensing terms; follow the notices in the relevant files.