Commit
•
f9dc86b
1
Parent(s):
4dcfa1b
DazMashaly/iee_class
Browse files- README.md +11 -53
- pytorch_model.bin +1 -1
README.md
CHANGED
|
@@ -2,9 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
-
metrics:
|
| 7 |
-
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: vit-base-beans-demo-v5
|
| 10 |
results: []
|
|
@@ -15,10 +14,15 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# vit-base-beans-demo-v5
|
| 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
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
-
|
| 21 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -37,59 +41,13 @@ More information needed
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
-
- learning_rate:
|
| 41 |
- train_batch_size: 16
|
| 42 |
- eval_batch_size: 8
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
-
- num_epochs:
|
| 47 |
-
|
| 48 |
-
### Training results
|
| 49 |
-
|
| 50 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 51 |
-
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 52 |
-
| 0.8306 | 0.15 | 100 | 0.7915 | 0.9085 |
|
| 53 |
-
| 0.3828 | 0.3 | 200 | 0.4295 | 0.9270 |
|
| 54 |
-
| 0.3623 | 0.45 | 300 | 0.3022 | 0.9370 |
|
| 55 |
-
| 0.2623 | 0.6 | 400 | 0.2883 | 0.9270 |
|
| 56 |
-
| 0.2251 | 0.75 | 500 | 0.2303 | 0.9463 |
|
| 57 |
-
| 0.2695 | 0.9 | 600 | 0.2465 | 0.9379 |
|
| 58 |
-
| 0.1256 | 1.04 | 700 | 0.1249 | 0.9715 |
|
| 59 |
-
| 0.0718 | 1.19 | 800 | 0.1335 | 0.9622 |
|
| 60 |
-
| 0.0436 | 1.34 | 900 | 0.0989 | 0.9790 |
|
| 61 |
-
| 0.1681 | 1.49 | 1000 | 0.1853 | 0.9496 |
|
| 62 |
-
| 0.0771 | 1.64 | 1100 | 0.1400 | 0.9647 |
|
| 63 |
-
| 0.0815 | 1.79 | 1200 | 0.0970 | 0.9765 |
|
| 64 |
-
| 0.0203 | 1.94 | 1300 | 0.1408 | 0.9647 |
|
| 65 |
-
| 0.0774 | 2.09 | 1400 | 0.1239 | 0.9664 |
|
| 66 |
-
| 0.0723 | 2.24 | 1500 | 0.0623 | 0.9824 |
|
| 67 |
-
| 0.0187 | 2.39 | 1600 | 0.0926 | 0.9731 |
|
| 68 |
-
| 0.055 | 2.54 | 1700 | 0.0910 | 0.9757 |
|
| 69 |
-
| 0.0304 | 2.69 | 1800 | 0.1278 | 0.9723 |
|
| 70 |
-
| 0.0068 | 2.84 | 1900 | 0.1283 | 0.9681 |
|
| 71 |
-
| 0.0319 | 2.99 | 2000 | 0.0590 | 0.9866 |
|
| 72 |
-
| 0.0054 | 3.13 | 2100 | 0.0761 | 0.9824 |
|
| 73 |
-
| 0.0136 | 3.28 | 2200 | 0.0759 | 0.9824 |
|
| 74 |
-
| 0.0136 | 3.43 | 2300 | 0.0448 | 0.9866 |
|
| 75 |
-
| 0.0086 | 3.58 | 2400 | 0.0816 | 0.9849 |
|
| 76 |
-
| 0.0212 | 3.73 | 2500 | 0.0574 | 0.9849 |
|
| 77 |
-
| 0.0392 | 3.88 | 2600 | 0.0765 | 0.9824 |
|
| 78 |
-
| 0.0029 | 4.03 | 2700 | 0.0740 | 0.9790 |
|
| 79 |
-
| 0.0521 | 4.18 | 2800 | 0.0349 | 0.9924 |
|
| 80 |
-
| 0.0025 | 4.33 | 2900 | 0.0639 | 0.9866 |
|
| 81 |
-
| 0.0024 | 4.48 | 3000 | 0.0509 | 0.9882 |
|
| 82 |
-
| 0.0024 | 4.63 | 3100 | 0.0468 | 0.9899 |
|
| 83 |
-
| 0.0024 | 4.78 | 3200 | 0.0524 | 0.9882 |
|
| 84 |
-
| 0.0022 | 4.93 | 3300 | 0.0504 | 0.9899 |
|
| 85 |
-
| 0.0022 | 5.07 | 3400 | 0.0488 | 0.9891 |
|
| 86 |
-
| 0.0021 | 5.22 | 3500 | 0.0484 | 0.9899 |
|
| 87 |
-
| 0.0019 | 5.37 | 3600 | 0.0480 | 0.9899 |
|
| 88 |
-
| 0.0019 | 5.52 | 3700 | 0.0477 | 0.9899 |
|
| 89 |
-
| 0.0019 | 5.67 | 3800 | 0.0475 | 0.9899 |
|
| 90 |
-
| 0.0018 | 5.82 | 3900 | 0.0474 | 0.9899 |
|
| 91 |
-
| 0.0018 | 5.97 | 4000 | 0.0473 | 0.9899 |
|
| 92 |
-
|
| 93 |
|
| 94 |
### Framework versions
|
| 95 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
|
|
|
|
|
|
| 7 |
model-index:
|
| 8 |
- name: vit-base-beans-demo-v5
|
| 9 |
results: []
|
|
|
|
| 14 |
|
| 15 |
# vit-base-beans-demo-v5
|
| 16 |
|
| 17 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the iee_clas dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- eval_loss: 0.0386
|
| 20 |
+
- eval_accuracy: 0.9916
|
| 21 |
+
- eval_runtime: 19.4296
|
| 22 |
+
- eval_samples_per_second: 61.298
|
| 23 |
+
- eval_steps_per_second: 7.669
|
| 24 |
+
- epoch: 0.45
|
| 25 |
+
- step: 300
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
|
|
| 41 |
### Training hyperparameters
|
| 42 |
|
| 43 |
The following hyperparameters were used during training:
|
| 44 |
+
- learning_rate: 8e-07
|
| 45 |
- train_batch_size: 16
|
| 46 |
- eval_batch_size: 8
|
| 47 |
- seed: 42
|
| 48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
- lr_scheduler_type: linear
|
| 50 |
+
- num_epochs: 1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
|
| 52 |
### Framework versions
|
| 53 |
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343308717
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb51ffc43df7a2b4027d38fe0f48bfd5339fff4053aed3a259b148bbb7feacb7
|
| 3 |
size 343308717
|