https://github.com/tirendazacademy/lightly-train-tutorials
This repo shows how to use the LightlyTrain framework
https://github.com/tirendazacademy/lightly-train-tutorials
ai computer-vision data-science deep-learning fine-tuning generative-ai image-classification image-processing lightly-train machine-learning object-detection self-supervised-learning
Last synced: 3 months ago
JSON representation
This repo shows how to use the LightlyTrain framework
- Host: GitHub
- URL: https://github.com/tirendazacademy/lightly-train-tutorials
- Owner: TirendazAcademy
- Created: 2025-04-12T07:57:10.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2025-04-12T08:17:31.000Z (6 months ago)
- Last Synced: 2025-04-12T09:23:38.491Z (6 months ago)
- Topics: ai, computer-vision, data-science, deep-learning, fine-tuning, generative-ai, image-classification, image-processing, lightly-train, machine-learning, object-detection, self-supervised-learning
- Language: Python
- Homepage:
- Size: 18.6 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# LightlyTrain Tutorial
**LightlyTrain** is a powerful new framework for training computer vision models on **unlabeled data**.
It allows you to get started with model training immediately using just a directory of images — no labels needed!
This repo provides guides on how to use **LightlyTrain** effectively.
## 📺 My YouTube Video:
- [Train Your Computer Vision Models in Minutes with LightlyTrain](https://youtu.be/CtvX6yjUA_k)
## ✍️ My Blog Post:
- [LightlyTrain: A New Framework for Computer Vision](https://medium.com/data-science-collective/lightlytrain-a-new-framework-for-computer-vision-1b71d988789b)
## 🚀 What is LightlyTrain?
LightlyTrain is designed to make **self-supervised learning** easy and accessible. It wraps around powerful computer vision techniques and enables fast training with just a few lines of code.
- ✅ Works with raw image folders (no annotations required)
- ✅ Supports pretrained backbones like ResNet
- ✅ Easily fine-tune for classification later
## 📦 Installation
```
pip install lightly-train
```
📁 Example Dataset
You can use the official clothing dataset:
```
git clone https://github.com/lightly-ai/dataset_clothing_images.git my_data_dir
rm -rf my_data_dir/.git
```
🧠 Training with LightlyTrain
```
import lightly_train
lightly_train.train(
out="out/my_experiment", # Output directory
data="my_data_dir", # Directory with images
model="torchvision/resnet18", # Model to train
epochs=10, # Number of epochs
batch_size=32, # Batch size
)
```
📚 Resources
- [Lightly AI](https://docs.lightly.ai/train/stable/index.html)