Hugging Face's logo

kevinoli
/
clip-finetuned-csu-p14-336-e3l57-l

Zero-Shot Image Classification
Transformers
TensorBoard
Safetensors
clip
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
kevinoli commited on
Commit
125f709
•
1 Parent(s): 6e29500

Model save

Browse files
Files changed (2) hide show
  1. README.md +34 -34
  2. runs/Aug22_08-32-24_winter-park/events.out.tfevents.1724337165.winter-park.2640878.0 +2 -2
README.md CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [openai/clip-vit-large-patch14-336](https://huggingface.co/openai/clip-vit-large-patch14-336) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.2831
19
 
20
  ## Model description
21
 
@@ -35,7 +35,7 @@ More information needed
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 5e-07
38
- - train_batch_size: 256
39
  - eval_batch_size: 8
40
  - seed: 42
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -46,38 +46,38 @@ The following hyperparameters were used during training:
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:------:|:-----:|:---------------:|
49
- | 0.3448 | 0.0921 | 500 | 1.0839 |
50
- | 0.2708 | 0.1842 | 1000 | 0.8948 |
51
- | 0.177 | 0.2763 | 1500 | 0.8396 |
52
- | 0.1831 | 0.3685 | 2000 | 0.7721 |
53
- | 0.2038 | 0.4606 | 2500 | 0.7446 |
54
- | 0.1309 | 0.5527 | 3000 | 0.7235 |
55
- | 0.1431 | 0.6448 | 3500 | 0.6691 |
56
- | 0.1411 | 0.7369 | 4000 | 0.6437 |
57
- | 0.0849 | 0.8290 | 4500 | 0.4805 |
58
- | 0.1026 | 0.9211 | 5000 | 0.4800 |
59
- | 0.1201 | 1.0133 | 5500 | 0.4810 |
60
- | 0.0757 | 1.1054 | 6000 | 0.4692 |
61
- | 0.0696 | 1.1975 | 6500 | 0.4764 |
62
- | 0.0911 | 1.2896 | 7000 | 0.4601 |
63
- | 0.0806 | 1.3817 | 7500 | 0.4590 |
64
- | 0.088 | 1.4738 | 8000 | 0.4654 |
65
- | 0.0878 | 1.5660 | 8500 | 0.4769 |
66
- | 0.0369 | 1.6581 | 9000 | 0.4684 |
67
- | 0.1034 | 1.7502 | 9500 | 0.4716 |
68
- | 0.0852 | 1.8423 | 10000 | 0.4720 |
69
- | 0.0493 | 1.9344 | 10500 | 0.4714 |
70
- | 0.0603 | 2.0265 | 11000 | 0.4661 |
71
- | 0.0547 | 2.1186 | 11500 | 0.4669 |
72
- | 0.0793 | 2.2108 | 12000 | 0.4664 |
73
- | 0.0415 | 2.3024 | 12500 | 0.2888 |
74
- | 0.0565 | 2.3945 | 13000 | 0.2910 |
75
- | 0.0629 | 2.4866 | 13500 | 0.2889 |
76
- | 0.0584 | 2.5787 | 14000 | 0.2874 |
77
- | 0.0582 | 2.6708 | 14500 | 0.2863 |
78
- | 0.052 | 2.7629 | 15000 | 0.2846 |
79
- | 0.0402 | 2.8550 | 15500 | 0.2835 |
80
- | 0.0518 | 2.9471 | 16000 | 0.2831 |
81
 
82
 
83
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [openai/clip-vit-large-patch14-336](https://huggingface.co/openai/clip-vit-large-patch14-336) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.5312
19
 
20
  ## Model description
21
 
 
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 5e-07
38
+ - train_batch_size: 128
39
  - eval_batch_size: 8
40
  - seed: 42
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:------:|:-----:|:---------------:|
49
+ | 0.2708 | 0.0921 | 500 | 1.0631 |
50
+ | 0.251 | 0.1842 | 1000 | 0.9428 |
51
+ | 0.235 | 0.2763 | 1500 | 0.8717 |
52
+ | 0.1529 | 0.3685 | 2000 | 0.8318 |
53
+ | 0.1781 | 0.4606 | 2500 | 0.7549 |
54
+ | 0.1681 | 0.5527 | 3000 | 0.7218 |
55
+ | 0.1064 | 0.6448 | 3500 | 0.7048 |
56
+ | 0.1357 | 0.7369 | 4000 | 0.6962 |
57
+ | 0.1098 | 0.8290 | 4500 | 0.6778 |
58
+ | 0.1142 | 0.9211 | 5000 | 0.6657 |
59
+ | 0.1113 | 1.0133 | 5500 | 0.6431 |
60
+ | 0.0572 | 1.1054 | 6000 | 0.6367 |
61
+ | 0.0746 | 1.1975 | 6500 | 0.6261 |
62
+ | 0.0494 | 1.2896 | 7000 | 0.6245 |
63
+ | 0.0788 | 1.3817 | 7500 | 0.6120 |
64
+ | 0.0808 | 1.4738 | 8000 | 0.6011 |
65
+ | 0.0536 | 1.5660 | 8500 | 0.5893 |
66
+ | 0.0869 | 1.6581 | 9000 | 0.5916 |
67
+ | 0.0752 | 1.7502 | 9500 | 0.5707 |
68
+ | 0.0577 | 1.8423 | 10000 | 0.5678 |
69
+ | 0.0891 | 1.9344 | 10500 | 0.5631 |
70
+ | 0.0559 | 2.0265 | 11000 | 0.5548 |
71
+ | 0.0385 | 2.1186 | 11500 | 0.5536 |
72
+ | 0.0185 | 2.2108 | 12000 | 0.5519 |
73
+ | 0.0642 | 2.3029 | 12500 | 0.5505 |
74
+ | 0.0456 | 2.3950 | 13000 | 0.5444 |
75
+ | 0.0476 | 2.4871 | 13500 | 0.5395 |
76
+ | 0.027 | 2.5792 | 14000 | 0.5361 |
77
+ | 0.042 | 2.6713 | 14500 | 0.5356 |
78
+ | 0.0469 | 2.7634 | 15000 | 0.5343 |
79
+ | 0.0438 | 2.8556 | 15500 | 0.5332 |
80
+ | 0.0481 | 2.9477 | 16000 | 0.5312 |
81
 
82
 
83
  ### Framework versions
runs/Aug22_08-32-24_winter-park/events.out.tfevents.1724337165.winter-park.2640878.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b7e20eec1fee8f62905442045f444f825b20485514cc6be0ceb54127ffdc3caf
3
- size 50457
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d75ff6290e963e25e4cbe3e61c2be9343fc5d0a918c3553674825fecc18dfdea
3
+ size 50811