seyviour/paligemma_ocr_final

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/paligemma-3b-mix-448](https://huggingface.co/google/paligemma-3b-mix-448) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5112
+- Loss: 0.3263
 
 ## Model description
 
@@ -48,10 +48,10 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 0.6137        | 1.0   | 1250 | 0.5314          |
-| 0.4361        | 2.0   | 2500 | 0.5112          |
+| Training Loss | Epoch  | Step  | Validation Loss |
+|:-------------:|:------:|:-----:|:---------------:|
+| 0.3228        | 0.9999 | 6926  | 0.3505          |
+| 0.2699        | 1.9999 | 13852 | 0.3263          |
 
 
 ### Framework versions