seyviour/paligemma_VQAv2_enel645
Browse files- README.md +35 -0
- adapter_model.safetensors +1 -1
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# paligemma_vqav2
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [google/paligemma-3b-pt-224](https://huggingface.co/google/paligemma-3b-pt-224) on an unknown dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -46,6 +48,39 @@ The following hyperparameters were used during training:
|
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
|
| 51 |
### Framework versions
|
|
|
|
| 15 |
# paligemma_vqav2
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [google/paligemma-3b-pt-224](https://huggingface.co/google/paligemma-3b-pt-224) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.2568
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 53 |
+
| 6.2845 | 0.3137 | 100 | 1.7300 |
|
| 54 |
+
| 0.9773 | 0.6275 | 200 | 0.6286 |
|
| 55 |
+
| 0.5181 | 0.9412 | 300 | 0.4802 |
|
| 56 |
+
| 0.4126 | 1.2549 | 400 | 0.4208 |
|
| 57 |
+
| 0.365 | 1.5686 | 500 | 0.3774 |
|
| 58 |
+
| 0.3369 | 1.8824 | 600 | 0.3645 |
|
| 59 |
+
| 0.3037 | 2.1961 | 700 | 0.3299 |
|
| 60 |
+
| 0.2854 | 2.5098 | 800 | 0.3164 |
|
| 61 |
+
| 0.2939 | 2.8235 | 900 | 0.3093 |
|
| 62 |
+
| 0.2547 | 3.1373 | 1000 | 0.2961 |
|
| 63 |
+
| 0.2275 | 3.4510 | 1100 | 0.2943 |
|
| 64 |
+
| 0.2456 | 3.7647 | 1200 | 0.2824 |
|
| 65 |
+
| 0.2368 | 4.0784 | 1300 | 0.2723 |
|
| 66 |
+
| 0.2148 | 4.3922 | 1400 | 0.2733 |
|
| 67 |
+
| 0.2118 | 4.7059 | 1500 | 0.2737 |
|
| 68 |
+
| 0.1991 | 5.0196 | 1600 | 0.2715 |
|
| 69 |
+
| 0.1879 | 5.3333 | 1700 | 0.2657 |
|
| 70 |
+
| 0.1841 | 5.6471 | 1800 | 0.2746 |
|
| 71 |
+
| 0.1912 | 5.9608 | 1900 | 0.2642 |
|
| 72 |
+
| 0.1509 | 6.2745 | 2000 | 0.2964 |
|
| 73 |
+
| 0.1818 | 6.5882 | 2100 | 0.2607 |
|
| 74 |
+
| 0.1736 | 6.9020 | 2200 | 0.2644 |
|
| 75 |
+
| 0.1618 | 7.2157 | 2300 | 0.2663 |
|
| 76 |
+
| 0.1563 | 7.5294 | 2400 | 0.2637 |
|
| 77 |
+
| 0.159 | 7.8431 | 2500 | 0.2561 |
|
| 78 |
+
| 0.1488 | 8.1569 | 2600 | 0.2554 |
|
| 79 |
+
| 0.1417 | 8.4706 | 2700 | 0.2589 |
|
| 80 |
+
| 0.1329 | 8.7843 | 2800 | 0.2599 |
|
| 81 |
+
| 0.1455 | 9.0980 | 2900 | 0.2589 |
|
| 82 |
+
| 0.1521 | 9.4118 | 3000 | 0.2576 |
|
| 83 |
+
| 0.1334 | 9.7255 | 3100 | 0.2568 |
|
| 84 |
|
| 85 |
|
| 86 |
### Framework versions
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 45258384
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:287be4e05447c932930cc84cb59bc214cd99470a74c0185bfe5425c22d0c705b
|
| 3 |
size 45258384
|