100-Days-Of-ML-Code: A Raw Learning Log, Not a Tutorial Repo
GitHub describes it as 100 Days of ML Coding. The metadata lists the MIT license. This article stays within the project description and details documented in the GitHub repository README.
At a glance
- What is it?
- Avik-Jain's 100-Days-Of-ML-Code is a public diary of one developer's daily machine learning studies, mixing code snippets, course links, and infographics. It is a useful reference for beginners who want a curated path, but it is not a structured curriculum or a maintained codebase.
- Who is it for?
- Adopt this repo if you are a beginner who wants a day-by-day list of topics and links to external resources, and you are comfortable with the fact that the code is minimal and the documentation is uneven. Do not adopt it if you need a structured tutorial with consistent explanations or a maintained codebase.
- Can I use it commercially?
- Yes. MIT is a permissive licence: you can use, modify and sell software built on it, as long as you keep its copyright and licence notices.
- Is it still maintained?
- Probably not. The repository last received commits 35 months ago, on December 29, 2023.
- What is it written in?
- GitHub does not report a main language for this repository.
Answers come from the project's GitHub data, last synced on November 15, 2026, and from our analysis. They are not legal advice.
Star history
- Avik-Jain/100-Days-Of-ML-Code52,069 ★ · +421
| Repository | 2026-08-13 | 2026-11-15 |
|---|---|---|
| Avik-Jain/100-Days-Of-ML-Code | 51.6k | 52.1k |
421 Stars in the last 94 days, measured from the daily GitHub snapshot. Compare with other projects
Editorial analysis
A Learning Diary, Not a Codebase
The intended audience is a beginner who wants a structured sequence of machine learning topics, from data preprocessing to deep learning. The repo provides a day-by-day list, which can serve as a checklist. It is not for someone who wants a polished, self-contained course, because the material is a mix of links, notes, and occasional code. The author himself asks for help with documentation, which shows the informal nature of the project.
The Daily Structure and What It Covers
The code files are separate Markdown files in the Code directory, not a single script. For example, the README links to 'Day 1_Data PreProcessing.md' and 'Day2_Simple_Linear_Regression.md'. The datasets are stored in a separate 'datasets' folder. This organization is simple, but it means you have to navigate multiple files to follow along. The repository also has an Info-graphs directory, and the README says infographics are posted on alternate days.
How the Code Is Presented and What It Assumes
The README also assumes you are willing to watch external videos and read external articles. Day 8 points to a Towards Data Science article on logistic regression. Day 19 starts Caltech's CS 156 course. Day 26 to 30 are entirely about 3Blue1Brown's linear algebra and calculus playlists. So the repo is not self-contained. It is a hub of links, and the actual learning happens elsewhere. This is a strength if you want curated resources, but a weakness if you expect the repo to teach you directly.
The Maintenance Problem and the Truncated README
You should not treat this repo as a reliable reference for production code. It is a learning log, and its maintenance status reflects that.
What You Can Actually Do With It
The repo can also be used as a reading list. Each day's entry points to a topic and often to an external resource. For instance, Day 21 mentions web scraping with Beautiful Soup, and Day 24 introduces statistical learning theory from a Bloomberg course. You could follow the day-by-day sequence and use the linked resources as your primary material. The repo then becomes a syllabus, not a textbook. This is a valid use case, but it requires self-discipline and an ability to find and follow external resources.
A Genuine Limitation: No Math, No Depth, No Consistency
Another limitation is that the code is minimal. The README does not show the code for many days, only links to it. For example, Day 6 links to 'Day 6 Logistic Regression.md', but the README does not give a preview. This means you have to click through to see the content, and the quality of that code is unknown. The author is a learner, not an expert, so the code may contain errors or suboptimal practices. Nothing in the README shows the code has been reviewed by others. This is a risk if you want to learn best practices.
Alternatives and How They Differ
The key difference is that this repo is a diary, not a course.
Licence and Upgrade Cost
The license does not affect the learning content, but it does mean you can legally reuse the code snippets in your own tutorials or projects. That is a positive, but it does not compensate for the lack of maintenance. If you are a contributor, you could fork the repo and update the code, and the repository is not archived, but its last push was on 2023-12-29. This is a factor to consider if you plan to rely on it.
Editorial conclusion
Adopt this repo if you are a beginner who wants a day-by-day list of topics and links to external resources, and you are comfortable with the fact that the code is minimal and the documentation is uneven. Do not adopt it if you need a structured tutorial with consistent explanations or a maintained codebase. Before relying on it, verify each day's code against current versions of scikit-learn and Python, because the repository has not been updated recently. The value here is the curated sequence of topics, not the code itself, so treat the repo as a study plan, not a reference.
Frequently asked questions
What is 100-Days-Of-ML-Code?
It is a personal learning journal following a hundred days of machine learning study, as proposed by Siraj Raval. The readme is a day-by-day log with links to Markdown code files for some days, covering data preprocessing, linear and logistic regression, k-nearest neighbours, support vector machines including the kernel trick, naive Bayes, decision trees, web scraping, and material from online courses.
How do you code for machine learning with this repository?
You follow the day-by-day readme and open the Markdown code files it links, with the datasets committed in a directory inside the repository. The named tool for the machine learning work is scikit-learn, and a web scraping library is used for the data collection entry.
Is 100-Days-Of-ML-Code still being updated?
No. The last push to the master branch was on 2023-12-29 and the repository has no GitHub releases, so there is no tag to anchor to.
Do all 100 days in 100-Days-Of-ML-Code contain code?
No. Some entries link to a Markdown code file, others are a title with an image, and several consecutive entries are notes about video lectures rather than code, including four in a row covering a linear algebra playlist. One entry headed as the SVM implementation also links to the previous day's file, so the day numbers are labels rather than a reliable index.
Official sources
Was this useful?