Hugging Face's logo

Shawon16
/
working

Safetensors
videomae
Generated from Trainer
Model card Files
xet
Community
Shawon16 commited on
Commit
ff92a06
·
verified ·
1 Parent(s): 444f838

Upload trained VideoMAE with metrics and all figures

Browse files
Files changed (9) hide show
  1. README.md +18 -10
  2. confusion_matrix_test_50.png +0 -0
  3. confusion_matrix_train_50.png +0 -0
  4. confusion_matrix_valid_50.png +0 -0
  5. longtail_f1_vs_freq.png +0 -0
  6. model.safetensors +1 -1
  7. per_class_report.csv +4 -4
  8. sample.gif +2 -2
  9. training_args.bin +1 -1
README.md CHANGED
@@ -17,14 +17,14 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 2.3200
21
- - Accuracy: 0.125
22
- - Top 1 Accuracy: 0.125
23
- - Top 5 Accuracy: 0.5
24
  - Top 10 Accuracy: 1.0
25
- - Macro Precision: 0.0125
26
- - Macro Recall: 0.1
27
- - Macro F1: 0.0222
28
 
29
  ## Model description
30
 
@@ -52,14 +52,22 @@ The following hyperparameters were used during training:
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
  - lr_scheduler_warmup_ratio: 0.1
55
- - training_steps: 17
56
 
57
  ### Training results
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Top 1 Accuracy | Top 5 Accuracy | Top 10 Accuracy | Macro Precision | Macro Recall | Macro F1 |
60
  |:-------------:|:------:|:----:|:---------------:|:--------:|:--------------:|:--------------:|:---------------:|:---------------:|:------------:|:--------:|
61
- | No log | 0.4706 | 8 | 2.3092 | 0.1562 | 0.1562 | 0.5312 | 1.0 | 0.1133 | 0.1333 | 0.0735 |
62
- | No log | 1.4853 | 17 | 2.3200 | 0.125 | 0.125 | 0.5 | 1.0 | 0.0125 | 0.1 | 0.0222 |
 
 
 
 
 
 
 
 
63
 
64
 
65
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 2.3393
21
+ - Accuracy: 0.1562
22
+ - Top 1 Accuracy: 0.1562
23
+ - Top 5 Accuracy: 0.5312
24
  - Top 10 Accuracy: 1.0
25
+ - Macro Precision: 0.1368
26
+ - Macro Recall: 0.1417
27
+ - Macro F1: 0.0954
28
 
29
  ## Model description
30
 
 
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
  - lr_scheduler_warmup_ratio: 0.1
55
+ - training_steps: 85
56
 
57
  ### Training results
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Top 1 Accuracy | Top 5 Accuracy | Top 10 Accuracy | Macro Precision | Macro Recall | Macro F1 |
60
  |:-------------:|:------:|:----:|:---------------:|:--------:|:--------------:|:--------------:|:---------------:|:---------------:|:------------:|:--------:|
61
+ | No log | 0.0941 | 8 | 2.3526 | 0.125 | 0.125 | 0.5312 | 1.0 | 0.1272 | 0.1083 | 0.0796 |
62
+ | No log | 1.0971 | 17 | 2.2900 | 0.1562 | 0.1562 | 0.5938 | 1.0 | 0.0958 | 0.1333 | 0.0881 |
63
+ | No log | 2.1 | 26 | 2.3085 | 0.1562 | 0.1562 | 0.5938 | 1.0 | 0.132 | 0.1333 | 0.0929 |
64
+ | No log | 3.1029 | 35 | 2.2905 | 0.125 | 0.125 | 0.5625 | 1.0 | 0.0292 | 0.1083 | 0.0437 |
65
+ | No log | 4.0941 | 43 | 2.2742 | 0.1562 | 0.1562 | 0.5625 | 1.0 | 0.1464 | 0.1333 | 0.0956 |
66
+ | No log | 5.0971 | 52 | 2.3259 | 0.125 | 0.125 | 0.5312 | 1.0 | 0.1318 | 0.1083 | 0.0840 |
67
+ | No log | 6.1 | 61 | 2.3297 | 0.125 | 0.125 | 0.5 | 1.0 | 0.0317 | 0.1083 | 0.0472 |
68
+ | No log | 7.1029 | 70 | 2.3367 | 0.125 | 0.125 | 0.5 | 1.0 | 0.0279 | 0.1167 | 0.045 |
69
+ | No log | 8.0941 | 78 | 2.3388 | 0.1562 | 0.1562 | 0.5625 | 1.0 | 0.1361 | 0.1417 | 0.0945 |
70
+ | No log | 9.0735 | 85 | 2.3393 | 0.1562 | 0.1562 | 0.5312 | 1.0 | 0.1368 | 0.1417 | 0.0954 |
71
 
72
 
73
  ### Framework versions
confusion_matrix_test_50.png CHANGED
confusion_matrix_train_50.png CHANGED
confusion_matrix_valid_50.png CHANGED
longtail_f1_vs_freq.png CHANGED
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0285333c3d645cb733ed4589c7fac629825479201254ec332803d9e452b40821
3
  size 344961984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb1317df7dcbed4bb21e19d978191763a3ab88347069e8ebf59a67faa0425be4
3
  size 344961984
per_class_report.csv CHANGED
@@ -5,10 +5,10 @@ all,0.0,0.0,0.0,3.0
5
  apple,0.0,0.0,0.0,2.0
6
  basketball,0.0,0.0,0.0,2.0
7
  bed,0.0,0.0,0.0,2.0
8
- before,0.14814814814814814,1.0,0.25806451612903225,4.0
9
  bird,0.0,0.0,0.0,2.0
10
  birthday,0.0,0.0,0.0,3.0
11
  black,0.0,0.0,0.0,3.0
12
- accuracy,0.14814814814814814,0.14814814814814814,0.14814814814814814,0.14814814814814814
13
- macro avg,0.014814814814814814,0.1,0.025806451612903226,27.0
14
- weighted avg,0.02194787379972565,0.14814814814814814,0.038231780167264036,27.0
 
5
  apple,0.0,0.0,0.0,2.0
6
  basketball,0.0,0.0,0.0,2.0
7
  bed,0.0,0.0,0.0,2.0
8
+ before,0.08695652173913043,0.5,0.14814814814814817,4.0
9
  bird,0.0,0.0,0.0,2.0
10
  birthday,0.0,0.0,0.0,3.0
11
  black,0.0,0.0,0.0,3.0
12
+ accuracy,0.07407407407407407,0.07407407407407407,0.07407407407407407,0.07407407407407407
13
+ macro avg,0.008695652173913044,0.05,0.014814814814814817,27.0
14
+ weighted avg,0.01288244766505636,0.07407407407407407,0.021947873799725653,27.0
sample.gif CHANGED

Git LFS Details

  • SHA256: 718f5f69260f42de7cb9e4d1543929d4467c3f7d8d8604882bf6492f84f09267
  • Pointer size: 131 Bytes
  • Size of remote file: 328 kB

Git LFS Details

  • SHA256: fe19ad6e6ecde3a63dbabbe92e65191dc9c7904111f67f424a474f26156f8f84
  • Pointer size: 131 Bytes
  • Size of remote file: 273 kB
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3582f73824cac9f2727ba1778e06122783f38ec15e606c13da7720ab2c33922a
3
  size 4719
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68caca144e10f72dc31835e9b67b443ba71ce25e2e775603b22c6125066b3522
3
  size 4719