Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,24 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
# Classifiers Enhanced by Pre-training
|
| 6 |
+
|
| 7 |
+
This project utilizes a visual encoder from CLIP (ViT-B/32) to build image classifiers, enhanced by pre-training techniques. To use the trained models, follow the steps below to set up and run the classifiers.
|
| 8 |
+
|
| 9 |
+
## Prerequisites
|
| 10 |
+
|
| 11 |
+
Before you start, make sure you have Python and the necessary libraries installed.
|
| 12 |
+
|
| 13 |
+
## Downloading Model Weights
|
| 14 |
+
|
| 15 |
+
You need to download the following pre-trained model weights for running the `test.py` or `run_test.slurm`:
|
| 16 |
+
- `fine-tune-best.pth`: Best model weights after fine-tuning.
|
| 17 |
+
- `linear-probe-best.pth`: Best model weights after the linear probe training.
|
| 18 |
+
- `train-from-scratch-best.pth`: Model weights trained from scratch.
|
| 19 |
+
|
| 20 |
+
Please download these files and place them in the `results/` directory within the project folder.
|
| 21 |
+
|
| 22 |
+
## Installation and Usage
|
| 23 |
+
|
| 24 |
+
See https://github.com/Gengsheng-Li/Classifiers-enhanced-by-pre-training for more details.
|