Update README.md
Browse files
README.md
CHANGED
|
@@ -117,6 +117,8 @@ Corrected:
|
|
| 117 |
|
| 118 |

|
| 119 |
|
|
|
|
|
|
|
| 120 |
## β οΈ Limitations
|
| 121 |
|
| 122 |
- Rotationally symmetric images (balls, textures, patterns) are inherently ambiguous β no model can reliably classify these.
|
|
@@ -140,6 +142,10 @@ Trained from scratch on 50k images β a pretrained backbone would likely yield
|
|
| 140 |
|
| 141 |
> Yesterday, I was sorting photos and like every photo was rotated wrong! This inspired me to make this tool π
|
| 142 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 143 |
## π License
|
| 144 |
|
| 145 |
Apache 2.0
|
|
|
|
| 117 |
|
| 118 |

|
| 119 |
|
| 120 |
+
*Original Image Source: [Link to Pexels](https://www.pexels.com/de-de/foto/ruhiger-schlaf-einer-getigerten-hauskatze-auf-einem-sofa-32441547/)*
|
| 121 |
+
|
| 122 |
## β οΈ Limitations
|
| 123 |
|
| 124 |
- Rotationally symmetric images (balls, textures, patterns) are inherently ambiguous β no model can reliably classify these.
|
|
|
|
| 142 |
|
| 143 |
> Yesterday, I was sorting photos and like every photo was rotated wrong! This inspired me to make this tool π
|
| 144 |
|
| 145 |
+
## π» Training code
|
| 146 |
+
|
| 147 |
+
The full training code can be found in `train.py`. Have fun π
|
| 148 |
+
|
| 149 |
## π License
|
| 150 |
|
| 151 |
Apache 2.0
|