Commit
·
d457b1f
1
Parent(s):
45693d6
End of training
Browse files
README.md
CHANGED
|
@@ -14,6 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
# idefics-9b-instruct-ft-instruct-compact
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [HuggingFaceM4/idefics-9b-instruct](https://huggingface.co/HuggingFaceM4/idefics-9b-instruct) on an unknown dataset.
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -44,6 +46,28 @@ The following hyperparameters were used during training:
|
|
| 44 |
- num_epochs: 3
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
### Framework versions
|
| 48 |
|
| 49 |
- Transformers 4.35.2
|
|
|
|
| 14 |
# idefics-9b-instruct-ft-instruct-compact
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [HuggingFaceM4/idefics-9b-instruct](https://huggingface.co/HuggingFaceM4/idefics-9b-instruct) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.5312
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 46 |
- num_epochs: 3
|
| 47 |
- mixed_precision_training: Native AMP
|
| 48 |
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 3.9471 | 0.18 | 25 | 3.9005 |
|
| 54 |
+
| 3.513 | 0.36 | 50 | 3.4236 |
|
| 55 |
+
| 2.8952 | 0.54 | 75 | 2.5822 |
|
| 56 |
+
| 1.9127 | 0.73 | 100 | 1.7289 |
|
| 57 |
+
| 1.2521 | 0.91 | 125 | 1.0073 |
|
| 58 |
+
| 0.7576 | 1.09 | 150 | 0.7103 |
|
| 59 |
+
| 0.6537 | 1.27 | 175 | 0.6171 |
|
| 60 |
+
| 0.5969 | 1.45 | 200 | 0.5845 |
|
| 61 |
+
| 0.5778 | 1.63 | 225 | 0.5681 |
|
| 62 |
+
| 0.5663 | 1.82 | 250 | 0.5559 |
|
| 63 |
+
| 0.558 | 2.0 | 275 | 0.5471 |
|
| 64 |
+
| 0.5595 | 2.18 | 300 | 0.5409 |
|
| 65 |
+
| 0.5417 | 2.36 | 325 | 0.5367 |
|
| 66 |
+
| 0.5422 | 2.54 | 350 | 0.5341 |
|
| 67 |
+
| 0.54 | 2.72 | 375 | 0.5321 |
|
| 68 |
+
| 0.5474 | 2.91 | 400 | 0.5312 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
### Framework versions
|
| 72 |
|
| 73 |
- Transformers 4.35.2
|