hvass-labs/tensorflow-tutorials

★ 9,268Jupyter NotebookAudience · dataComplexity · 2/5Setup · easy

Mindmap

mindmap
  root((tensorflow-tutorials))
    What it does
      Beginner ML tutorials
      YouTube video paired
      Runnable notebooks
    Tech Stack
      Python
      TensorFlow 2
      Jupyter Notebook
      Google Colab
    Topics Covered
      Image recognition
      NLP and text
      Reinforcement learning
      Time-series
    Setup
      Run free on Colab
      No install needed
      TF1 set separate

mindmap root((tensorflow-tutorials)) What it does Beginner ML tutorials YouTube video paired Runnable notebooks Tech Stack Python TensorFlow 2 Jupyter Notebook Google Colab Topics Covered Image recognition NLP and text Reinforcement learning Time-series Setup Run free on Colab No install needed TF1 set separate

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Run beginner machine learning code on Google Colab without installing anything locally, following along with YouTube videos.

USE CASE 2

Learn how to build image recognition models by reading and running the paired notebook for that tutorial.

USE CASE 3

Use the time-series prediction tutorial as a starting point for forecasting your own data that changes over time.

Tech stack

PythonTensorFlowJupyter NotebookGoogle Colab

Getting it running

Difficulty · easy Time to first run · 5min

Older TF1 tutorials require installing TensorFlow 1 separately, the main TF2 tutorials run free on Google Colab with no local setup.

In plain English

This repository is a collection of beginner-friendly tutorials for TensorFlow, a popular library used to build and train machine learning models. Each tutorial focuses on a single topic and is paired with a YouTube video, so you can read through the code or watch it explained. The tutorials are written as Jupyter Notebooks, which means they combine explanatory text, runnable code, and output in a single document you can open in a browser. The topics covered span a wide range of machine learning techniques. Some tutorials introduce basic ideas like fitting a simple mathematical model to data or recognizing handwritten digits. Others go further into image recognition, natural language processing (understanding and generating text), machine translation, and reinforcement learning, which is a technique where a program learns by trial and error to achieve a goal. There is also a tutorial on time-series prediction, useful for forecasting data that changes over time. The collection is split into two groups. The main set of tutorials has been updated to work with TensorFlow 2, which is the current version of the library. A second group of older tutorials only works with TensorFlow 1 and requires installing that older version separately. The author notes these older tutorials have not been converted because of the effort involved. Each notebook can be run directly on Google Colab, which is a free online environment that does not require installing anything on your computer. This makes it straightforward to try the code without setting up a local environment. The tutorials have also been translated into Chinese by the community. The author invites contributions for additional translations of the remaining tutorials.

Copy-paste prompts

Prompt 1

Using the hvass-labs TensorFlow 2 tutorial on handwritten digit recognition, explain each code cell in plain terms and show me how to adapt it to classify my own images instead.

Prompt 2

I want to run the reinforcement learning tutorial from hvass-labs/tensorflow-tutorials on Google Colab. Walk me through opening the notebook, running each cell, and understanding what the agent is learning.

Prompt 3

Using the time-series prediction tutorial from hvass-labs/tensorflow-tutorials, help me adapt the code to forecast weekly sales numbers from a CSV file I have.

Open on GitHub → Explain another repo

← hvass-labs on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.