End of training

README.md

@@ -16,12 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 6.4079
-- eval_runtime: 96.0342
-- eval_samples_per_second: 0.666
-- eval_steps_per_second: 0.083
-- epoch: 1.0
-- step: 46
+- Loss: 0.4558
 
 ## Model description
 
@@ -51,6 +46,27 @@ The following hyperparameters were used during training:
 - lr_scheduler_warmup_ratio: 0.1
 - num_epochs: 15
 
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| No log        | 1.0   | 46   | 6.4079          |
+| No log        | 2.0   | 92   | 3.8737          |
+| 7.1608        | 3.0   | 138  | 2.9063          |
+| 7.1608        | 4.0   | 184  | 0.9980          |
+| 2.4956        | 5.0   | 230  | 0.7675          |
+| 2.4956        | 6.0   | 276  | 0.6447          |
+| 0.7023        | 7.0   | 322  | 0.5472          |
+| 0.7023        | 8.0   | 368  | 0.4761          |
+| 0.3457        | 9.0   | 414  | 0.4361          |
+| 0.3457        | 10.0  | 460  | 0.4849          |
+| 0.1816        | 11.0  | 506  | 0.4623          |
+| 0.1816        | 12.0  | 552  | 0.4471          |
+| 0.1816        | 13.0  | 598  | 0.4514          |
+| 0.132         | 14.0  | 644  | 0.4546          |
+| 0.132         | 15.0  | 690  | 0.4558          |
+
+
 ### Framework versions
 
 - Transformers 4.52.4

preprocessor_config.json

@@ -20,7 +20,7 @@
   "resample": 2,
   "rescale_factor": 0.00392156862745098,
   "size": [
-    512,
-    384
+    1024,
+    768
   ]
 }