Commit ·
53f2ce4
1
Parent(s): 4c71d43
Update README.md
Browse files
README.md
CHANGED
|
@@ -13,26 +13,17 @@ model-index:
|
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
|
| 16 |
-
#
|
| 17 |
-
|
| 18 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 19 |
-
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.1513
|
| 21 |
-
- Accuracy: 0.9649
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 24 |
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
## Intended uses & limitations
|
| 28 |
-
|
| 29 |
-
More information needed
|
| 30 |
|
| 31 |
## Training and evaluation data
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
## Training procedure
|
| 36 |
|
| 37 |
### Training hyperparameters
|
| 38 |
|
|
@@ -48,28 +39,6 @@ The following hyperparameters were used during training:
|
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
- num_epochs: 20
|
| 50 |
|
| 51 |
-
### Training results
|
| 52 |
-
|
| 53 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
-
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 55 |
-
| No log | 0.8 | 3 | 0.6639 | 0.7193 |
|
| 56 |
-
| No log | 1.87 | 7 | 0.5238 | 0.8421 |
|
| 57 |
-
| 0.6164 | 2.93 | 11 | 0.4318 | 0.8246 |
|
| 58 |
-
| 0.6164 | 4.0 | 15 | 0.3080 | 0.9474 |
|
| 59 |
-
| 0.6164 | 4.8 | 18 | 0.3269 | 0.9123 |
|
| 60 |
-
| 0.3439 | 5.87 | 22 | 0.2105 | 0.9649 |
|
| 61 |
-
| 0.3439 | 6.93 | 26 | 0.1825 | 0.9649 |
|
| 62 |
-
| 0.2371 | 8.0 | 30 | 0.1349 | 0.9649 |
|
| 63 |
-
| 0.2371 | 8.8 | 33 | 0.1607 | 0.9649 |
|
| 64 |
-
| 0.2371 | 9.87 | 37 | 0.1251 | 0.9825 |
|
| 65 |
-
| 0.2196 | 10.93 | 41 | 0.1570 | 0.9825 |
|
| 66 |
-
| 0.2196 | 12.0 | 45 | 0.1602 | 0.9649 |
|
| 67 |
-
| 0.2196 | 12.8 | 48 | 0.1101 | 0.9825 |
|
| 68 |
-
| 0.1576 | 13.87 | 52 | 0.1103 | 1.0 |
|
| 69 |
-
| 0.1576 | 14.93 | 56 | 0.1405 | 0.9649 |
|
| 70 |
-
| 0.1634 | 16.0 | 60 | 0.1513 | 0.9649 |
|
| 71 |
-
|
| 72 |
-
|
| 73 |
### Framework versions
|
| 74 |
|
| 75 |
- Transformers 4.33.2
|
|
|
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
|
| 16 |
+
# Crop vs Full Body
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
+
This model predicts whether the person in the image is **cropped** or **full body**. It is trained on [Lucy in the Sky](https://www.lucyinthesky.com/shop) images.
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k).
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
## Training and evaluation data
|
| 24 |
+
It achieves the following results on the evaluation set:
|
| 25 |
+
- Loss: 0.1513
|
| 26 |
+
- Accuracy: 0.9649
|
|
|
|
| 27 |
|
| 28 |
### Training hyperparameters
|
| 29 |
|
|
|
|
| 39 |
- lr_scheduler_warmup_ratio: 0.1
|
| 40 |
- num_epochs: 20
|
| 41 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
### Framework versions
|
| 43 |
|
| 44 |
- Transformers 4.33.2
|