Hugging Face's logo

sergiomvazq
/
DeBERTaV3_model_V4

Text Classification
Transformers
TensorBoard
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
sergiomvazq commited on
Commit
84c71e5
•
1 Parent(s): d757d18

End of training

Browse files
Files changed (5) hide show
  1. README.md +18 -19
  2. model.safetensors +1 -1
  3. runs/Jun11_07-28-26_0417-111206-hkgnini8-10-45-16-12/events.out.tfevents.1718090906.0417-111206-hkgnini8-10-45-16-12.6999.0 +3 -0
  4. runs/Jun11_07-28-26_0417-111206-hkgnini8-10-45-16-12/events.out.tfevents.1718092632.0417-111206-hkgnini8-10-45-16-12.6999.1 +3 -0
  5. training_args.bin +1 -1
README.md CHANGED
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.1871
24
- - Accuracy: 0.9316
25
- - F1: 0.6863
26
- - Precision: 0.8046
27
- - Recall: 0.5983
28
 
29
  ## Model description
30
 
@@ -49,25 +49,24 @@ The following hyperparameters were used during training:
49
  - seed: 42
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
- - num_epochs: 13
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
57
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
58
- | No log | 1.0 | 100 | 0.3528 | 0.875 | 0.0 | 0.0 | 0.0 |
59
- | No log | 2.0 | 200 | 0.3333 | 0.8761 | 0.0333 | 0.6667 | 0.0171 |
60
- | No log | 3.0 | 300 | 0.2886 | 0.8942 | 0.3529 | 0.75 | 0.2308 |
61
- | No log | 4.0 | 400 | 0.2616 | 0.9103 | 0.5059 | 0.8113 | 0.3675 |
62
- | 0.3381 | 5.0 | 500 | 0.2445 | 0.9113 | 0.5146 | 0.8148 | 0.3761 |
63
- | 0.3381 | 6.0 | 600 | 0.2250 | 0.9135 | 0.5207 | 0.8462 | 0.3761 |
64
- | 0.3381 | 7.0 | 700 | 0.2155 | 0.9167 | 0.5568 | 0.8305 | 0.4188 |
65
- | 0.3381 | 8.0 | 800 | 0.2011 | 0.9263 | 0.6145 | 0.8871 | 0.4701 |
66
- | 0.3381 | 9.0 | 900 | 0.2037 | 0.9284 | 0.6564 | 0.8205 | 0.5470 |
67
- | 0.1823 | 10.0 | 1000 | 0.1912 | 0.9295 | 0.6562 | 0.84 | 0.5385 |
68
- | 0.1823 | 11.0 | 1100 | 0.1907 | 0.9284 | 0.6599 | 0.8125 | 0.5556 |
69
- | 0.1823 | 12.0 | 1200 | 0.1879 | 0.9295 | 0.6765 | 0.7931 | 0.5897 |
70
- | 0.1823 | 13.0 | 1300 | 0.1871 | 0.9316 | 0.6863 | 0.8046 | 0.5983 |
71
 
72
 
73
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.1603
24
+ - Accuracy: 0.9375
25
+ - F1: 0.7346
26
+ - Precision: 0.7829
27
+ - Recall: 0.6919
28
 
29
  ## Model description
30
 
 
49
  - seed: 42
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
+ - num_epochs: 12
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
57
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
58
+ | No log | 1.0 | 138 | 0.3584 | 0.875 | 0.0 | 0.0 | 0.0 |
59
+ | No log | 2.0 | 276 | 0.3378 | 0.875 | 0.0 | 0.0 | 0.0 |
60
+ | No log | 3.0 | 414 | 0.2835 | 0.8881 | 0.1979 | 0.95 | 0.1105 |
61
+ | 0.3594 | 4.0 | 552 | 0.2480 | 0.9048 | 0.4018 | 0.9362 | 0.2558 |
62
+ | 0.3594 | 5.0 | 690 | 0.2053 | 0.9346 | 0.6617 | 0.9362 | 0.5116 |
63
+ | 0.3594 | 6.0 | 828 | 0.1863 | 0.9419 | 0.7279 | 0.8770 | 0.6221 |
64
+ | 0.3594 | 7.0 | 966 | 0.1759 | 0.9433 | 0.7484 | 0.8406 | 0.6744 |
65
+ | 0.1954 | 8.0 | 1104 | 0.1657 | 0.9419 | 0.75 | 0.8108 | 0.6977 |
66
+ | 0.1954 | 9.0 | 1242 | 0.1625 | 0.9411 | 0.7508 | 0.7974 | 0.7093 |
67
+ | 0.1954 | 10.0 | 1380 | 0.1627 | 0.9382 | 0.7416 | 0.7771 | 0.7093 |
68
+ | 0.1124 | 11.0 | 1518 | 0.1608 | 0.9360 | 0.7317 | 0.7692 | 0.6977 |
69
+ | 0.1124 | 12.0 | 1656 | 0.1603 | 0.9375 | 0.7346 | 0.7829 | 0.6919 |
 
70
 
71
 
72
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3bcb84e7edd24e9128c2d2115d2ea02f7d87ae6d558b827d00b0637b86d70816
3
  size 567617008
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f41a8c09a7a65fce3325899ed0d9a4a2c747d049ec7078cb0e7c0fb2b1448c2
3
  size 567617008
runs/Jun11_07-28-26_0417-111206-hkgnini8-10-45-16-12/events.out.tfevents.1718090906.0417-111206-hkgnini8-10-45-16-12.6999.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6cee944a0b6f8b7d052b4f12ab6be78e40b7f2f8cb98e2b96624a8981c7d834
3
+ size 11980
runs/Jun11_07-28-26_0417-111206-hkgnini8-10-45-16-12/events.out.tfevents.1718092632.0417-111206-hkgnini8-10-45-16-12.6999.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df55d6c93f8fc87179c9a47f5a4bf868e5d412cbba4c29b15a194253d13f4204
3
+ size 512
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1437200755c52365b813d98fb1f0d015382ad232e2188e7ebfadddc0f2570e40
3
  size 5112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b55bde135ae7df2123453241fc9710773a2e8e03b8a272f1a7ea3a121d05f2a1
3
  size 5112