Update README.md
Browse files
README.md
CHANGED
|
@@ -114,13 +114,6 @@ This repository does **not train models** but exports pre-trained models to opti
|
|
| 114 |
- **Export**: ExecuTorch with dynamic quantization
|
| 115 |
- **Categories**: 50+ predefined audio event categories
|
| 116 |
|
| 117 |
-
### Quantization Impact
|
| 118 |
-
|
| 119 |
-
| Model | Original Size | Quantized Size | Quality Impact |
|
| 120 |
-
|-------|---------------|----------------|----------------|
|
| 121 |
-
| Decoder | ~17MB | ~15MB | Minimal (<2% caption quality) |
|
| 122 |
-
| Categorizer | ~90MB | ~23MB | Minimal (<1% accuracy) |
|
| 123 |
-
|
| 124 |
## Project Structure
|
| 125 |
|
| 126 |
```
|
|
|
|
| 114 |
- **Export**: ExecuTorch with dynamic quantization
|
| 115 |
- **Categories**: 50+ predefined audio event categories
|
| 116 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 117 |
## Project Structure
|
| 118 |
|
| 119 |
```
|