Instructions to use shaanzeeeee/vit_base_patch16_pc_parts_classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- fastai
How to use shaanzeeeee/vit_base_patch16_pc_parts_classifier with fastai:
from huggingface_hub import from_pretrained_fastai learn = from_pretrained_fastai("shaanzeeeee/vit_base_patch16_pc_parts_classifier") - timm
How to use shaanzeeeee/vit_base_patch16_pc_parts_classifier with timm:
import timm model = timm.create_model("hf_hub:shaanzeeeee/vit_base_patch16_pc_parts_classifier", pretrained=True) - Notebooks
- Google Colab
- Kaggle
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,3 +1,68 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language: en
|
| 3 |
license: mit
|
| 4 |
+
library_name: fastai
|
| 5 |
+
pipeline_tag: image-classification
|
| 6 |
+
tags:
|
| 7 |
+
- fastai
|
| 8 |
+
- timm
|
| 9 |
+
- vision-transformer
|
| 10 |
+
- image-classification
|
| 11 |
+
- pc-parts
|
| 12 |
+
widget:
|
| 13 |
+
- src: https://images.unsplash.com/photo-1587202372775-a457f4ad61b9
|
| 14 |
+
example_title: PC build example
|
| 15 |
---
|
| 16 |
+
|
| 17 |
+
# vit_base_patch16_pc_parts_classifier
|
| 18 |
+
|
| 19 |
+
Vision Transformer image classifier for 11 PC component and cable-management classes.
|
| 20 |
+
|
| 21 |
+
## Model Details
|
| 22 |
+
|
| 23 |
+
- Architecture: ViT-Base Patch16 224 (`vit_base_patch16_224`)
|
| 24 |
+
- Framework: FastAI + timm + PyTorch
|
| 25 |
+
- Input size: 224x224 RGB
|
| 26 |
+
- Classes: 11
|
| 27 |
+
- Epochs: 15
|
| 28 |
+
- Batch size: 16
|
| 29 |
+
- Test accuracy: 0.7389
|
| 30 |
+
- Training date: 2026-04-17
|
| 31 |
+
|
| 32 |
+
## Labels
|
| 33 |
+
|
| 34 |
+
1. AIO_Liquid_Cooler
|
| 35 |
+
2. Air_Cooler
|
| 36 |
+
3. Bad_Cable_Management
|
| 37 |
+
4. CPU
|
| 38 |
+
5. Good_Cable_Management
|
| 39 |
+
6. Graphics_Card
|
| 40 |
+
7. M2_NVMe_Drive
|
| 41 |
+
8. Motherboard
|
| 42 |
+
9. PC_Case
|
| 43 |
+
10. Power_Supply
|
| 44 |
+
11. RAM_Stick
|
| 45 |
+
|
| 46 |
+
## Files
|
| 47 |
+
|
| 48 |
+
- `best_model_export.pkl`: FastAI export for direct inference.
|
| 49 |
+
- `best_model_state_dict.pth`: PyTorch state dict.
|
| 50 |
+
- `best_model_metadata.json`: Training and class metadata.
|
| 51 |
+
|
| 52 |
+
## Inference (FastAI)
|
| 53 |
+
|
| 54 |
+
```python
|
| 55 |
+
from fastai.learner import load_learner
|
| 56 |
+
from pathlib import Path
|
| 57 |
+
|
| 58 |
+
learn = load_learner("best_model_export.pkl")
|
| 59 |
+
pred_class, pred_idx, probs = learn.predict(Path("sample.jpg"))
|
| 60 |
+
print(pred_class)
|
| 61 |
+
print({learn.dls.vocab[i]: float(probs[i]) for i in range(len(probs))})
|
| 62 |
+
```
|
| 63 |
+
|
| 64 |
+
## Live Demo
|
| 65 |
+
|
| 66 |
+
A live inference demo is available on Hugging Face Spaces:
|
| 67 |
+
|
| 68 |
+
- https://huggingface.co/spaces/shaanzeeeee/vit-base-pc-parts-inference
|