Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/xunino/mlp_mixer
This implementation from paper MLP Mixer
https://github.com/xunino/mlp_mixer
mlp-mixer python3 tf2
Last synced: 24 days ago
JSON representation
This implementation from paper MLP Mixer
- Host: GitHub
- URL: https://github.com/xunino/mlp_mixer
- Owner: Xunino
- Created: 2022-01-27T00:28:37.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2022-02-08T03:38:26.000Z (almost 3 years ago)
- Last Synced: 2023-03-08T22:44:16.470Z (almost 2 years ago)
- Topics: mlp-mixer, python3, tf2
- Language: Python
- Homepage:
- Size: 5.48 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# MLP Mixer
This implementation from paper [MLP Mixer](https://arxiv.org/pdf/2105.01601.pdf). Give us a star if you like this repo.
## Architecture Image
Authors:
- Github: Xunino
- Email: [email protected]
## I. Set up environment
- Step 1:
```bash
conda create -n {your_env_name} python==3.7.0
```
- Step 2:
```bash
conda env create -f environment.yml
```
- Step 3:
```bash
conda activate {your_env_name}
```
## II. Set up your dataset
- Guide user how to download your data and set the data pipeline
- Data pipeline example:
```
train/
...class_a/
......a_image_1.jpg
......a_image_2.jpg
...class_b/
......b_image_1.jpg
......b_image_2.jpg
...class_c/
......c_image_1.jpg
......c_image_2.jpg
```
```
val/
...class_a/
......a_image_1.jpg
......a_image_2.jpg
...class_b/
......b_image_1.jpg
......b_image_2.jpg
...class_c/
......c_image_1.jpg
......c_image_2.jpg
```
## III. Training process:
**Script training:**
```
python3 train.py --train-path={dataset/train} --val-path={dataset/val} --batch-size=32 --epochs=100 --n_blocks=8 --C=512 --DC=1024 --DS=256 --image-size=224 --patch-size=32 --augments=False --retrain=False
```
- [Sample Colab](https://colab.research.google.com/drive/10sCbMku9ep9PK1U55IozjOrtlIFOsRO2?usp=sharing)