|
--- |
|
license: gpl-3.0 |
|
--- |
|
# Fish Disease Classification Model |
|
|
|
This repository contains a model trained for classifying various fish diseases using Hugging Face's Transformers library. The model is trained to classify the following fish diseases: |
|
|
|
|
|
1. **Tumor and deformity attack** |
|
2. **Polydactyly** |
|
3. **Hydrocephalus and swim bladder** |
|
4. **Holes in the head** |
|
5. **Nopez (I couldn't find a direct translation, it seems like a term specific to a certain context)** |
|
6. **Parasite in the mouth** |
|
7. **Catfish burn** |
|
8. **Gills** |
|
9. **Lernea worm** |
|
10. **Fungi** |
|
11. **Ich white spot** |
|
12. **Blind eye** |
|
13. **Fin rot** |
|
14. **Healthy** |
|
|
|
## Dataset |
|
|
|
The model is trained on a diverse dataset containing images of fish affected by different diseases as well as healthy fish samples. The dataset is labeled with the aforementioned disease categories to facilitate supervised training. |
|
|
|
[dataset base](https://huggingface.co/datasets/jero98772/CuraPeces_Background) |
|
[better dataset(removed background)](https://huggingface.co/datasets/jero98772/CuraPeces_Removed_background) |
|
|
|
## Model Architecture |
|
|
|
The classification model is built upon state-of-the-art deep learning architecture, leveraging the power of convolutional neural networks (CNNs) for image classification tasks. It utilizes transfer learning techniques, starting with a pre-trained backbone network (e.g., ResNet, EfficientNet) and fine-tuning it on the fish disease dataset to adapt to the specific classification task. |
|
|
|
|
|
|
|
|
|
Happy classifying fish diseases! 🐟🔍 |
