Hugging Face's logo

g-assismoraes
/
mdeberta-semeval25_fold1

Text Classification
Transformers
TensorBoard
Safetensors
deberta-v2
Generated from Trainer
Model card Files
xet
Metrics Community
g-assismoraes commited on
Commit
2e2c8f6
·
verified ·
1 Parent(s): d03262d

End of training

Browse files
Files changed (2) hide show
  1. README.md +23 -23
  2. runs/Oct26_21-07-52_icuff-Z790-UD/events.out.tfevents.1729987672.icuff-Z790-UD.65127.0 +2 -2
README.md CHANGED
@@ -16,19 +16,19 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 8.3902
20
- - Precision Samples: 0.0536
21
- - Recall Samples: 0.8615
22
- - F1 Samples: 0.0984
23
- - Precision Macro: 0.3774
24
- - Recall Macro: 0.6843
25
- - F1 Macro: 0.2260
26
- - Precision Micro: 0.0540
27
- - Recall Micro: 0.8395
28
- - F1 Micro: 0.1015
29
- - Precision Weighted: 0.1474
30
- - Recall Weighted: 0.8395
31
- - F1 Weighted: 0.1312
32
 
33
  ## Model description
34
 
@@ -59,16 +59,16 @@ The following hyperparameters were used during training:
59
 
60
  | Training Loss | Epoch | Step | Validation Loss | Precision Samples | Recall Samples | F1 Samples | Precision Macro | Recall Macro | F1 Macro | Precision Micro | Recall Micro | F1 Micro | Precision Weighted | Recall Weighted | F1 Weighted |
61
  |:-------------:|:-----:|:----:|:---------------:|:-----------------:|:--------------:|:----------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:--------:|:------------------:|:---------------:|:-----------:|
62
- | 10.7532 | 1.0 | 19 | 9.6249 | 0.0406 | 0.7489 | 0.0753 | 0.3858 | 0.5840 | 0.1131 | 0.0405 | 0.7068 | 0.0765 | 0.2573 | 0.7068 | 0.1074 |
63
- | 10.3058 | 2.0 | 38 | 9.2286 | 0.0498 | 0.7498 | 0.0908 | 0.6093 | 0.5594 | 0.2239 | 0.0498 | 0.7006 | 0.0929 | 0.3206 | 0.7006 | 0.1066 |
64
- | 9.667 | 3.0 | 57 | 9.0537 | 0.0478 | 0.7913 | 0.0879 | 0.5741 | 0.6 | 0.2344 | 0.0477 | 0.7438 | 0.0896 | 0.2715 | 0.7438 | 0.1055 |
65
- | 9.9892 | 4.0 | 76 | 8.9374 | 0.0500 | 0.7776 | 0.0914 | 0.5854 | 0.5873 | 0.2347 | 0.0501 | 0.7377 | 0.0938 | 0.2874 | 0.7377 | 0.1063 |
66
- | 9.6435 | 5.0 | 95 | 8.8197 | 0.0519 | 0.8190 | 0.0951 | 0.5468 | 0.6158 | 0.2440 | 0.0523 | 0.7870 | 0.0980 | 0.2414 | 0.7870 | 0.1217 |
67
- | 9.6045 | 6.0 | 114 | 8.6672 | 0.0509 | 0.8212 | 0.0934 | 0.5138 | 0.6095 | 0.2437 | 0.0513 | 0.7901 | 0.0963 | 0.2046 | 0.7901 | 0.1207 |
68
- | 9.1893 | 7.0 | 133 | 8.5546 | 0.0522 | 0.8259 | 0.0956 | 0.4816 | 0.6216 | 0.2468 | 0.0525 | 0.7994 | 0.0985 | 0.1872 | 0.7994 | 0.1239 |
69
- | 9.3328 | 8.0 | 152 | 8.4512 | 0.0530 | 0.8496 | 0.0973 | 0.4293 | 0.6593 | 0.2420 | 0.0533 | 0.8272 | 0.1002 | 0.1548 | 0.8272 | 0.1279 |
70
- | 9.2334 | 9.0 | 171 | 8.3911 | 0.0532 | 0.8464 | 0.0975 | 0.4298 | 0.6736 | 0.2542 | 0.0535 | 0.8272 | 0.1005 | 0.1551 | 0.8272 | 0.1315 |
71
- | 8.9167 | 10.0 | 190 | 8.3902 | 0.0536 | 0.8615 | 0.0984 | 0.3774 | 0.6843 | 0.2260 | 0.0540 | 0.8395 | 0.1015 | 0.1474 | 0.8395 | 0.1312 |
72
 
73
 
74
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 8.3259
20
+ - Precision Samples: 0.0548
21
+ - Recall Samples: 0.9029
22
+ - F1 Samples: 0.1006
23
+ - Precision Macro: 0.3664
24
+ - Recall Macro: 0.7660
25
+ - F1 Macro: 0.2045
26
+ - Precision Micro: 0.0544
27
+ - Recall Micro: 0.8735
28
+ - F1 Micro: 0.1024
29
+ - Precision Weighted: 0.1465
30
+ - Recall Weighted: 0.8735
31
+ - F1 Weighted: 0.1391
32
 
33
  ## Model description
34
 
 
59
 
60
  | Training Loss | Epoch | Step | Validation Loss | Precision Samples | Recall Samples | F1 Samples | Precision Macro | Recall Macro | F1 Macro | Precision Micro | Recall Micro | F1 Micro | Precision Weighted | Recall Weighted | F1 Weighted |
61
  |:-------------:|:-----:|:----:|:---------------:|:-----------------:|:--------------:|:----------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:--------:|:------------------:|:---------------:|:-----------:|
62
+ | 10.7598 | 1.0 | 19 | 9.6321 | 0.0361 | 0.8335 | 0.0679 | 0.3216 | 0.7097 | 0.1401 | 0.0363 | 0.7994 | 0.0694 | 0.1978 | 0.7994 | 0.1114 |
63
+ | 10.3024 | 2.0 | 38 | 9.2562 | 0.0422 | 0.8489 | 0.0787 | 0.4876 | 0.6843 | 0.2168 | 0.0422 | 0.8210 | 0.0803 | 0.2184 | 0.8210 | 0.1101 |
64
+ | 9.696 | 3.0 | 57 | 9.0731 | 0.0446 | 0.8616 | 0.0828 | 0.5117 | 0.695 | 0.2535 | 0.0445 | 0.8241 | 0.0844 | 0.2189 | 0.8241 | 0.1111 |
65
+ | 9.9846 | 4.0 | 76 | 8.8691 | 0.0462 | 0.8614 | 0.0856 | 0.4678 | 0.6926 | 0.2319 | 0.0458 | 0.8241 | 0.0868 | 0.1959 | 0.8241 | 0.1139 |
66
+ | 9.5643 | 5.0 | 95 | 8.6872 | 0.0492 | 0.8677 | 0.0908 | 0.4610 | 0.7167 | 0.2393 | 0.0487 | 0.8395 | 0.0921 | 0.1975 | 0.8395 | 0.1216 |
67
+ | 9.522 | 6.0 | 114 | 8.5495 | 0.0499 | 0.8787 | 0.0922 | 0.4629 | 0.7317 | 0.2542 | 0.0497 | 0.8457 | 0.0939 | 0.1827 | 0.8457 | 0.1234 |
68
+ | 9.1263 | 7.0 | 133 | 8.4783 | 0.0518 | 0.8849 | 0.0954 | 0.4306 | 0.7354 | 0.2449 | 0.0513 | 0.8488 | 0.0968 | 0.1533 | 0.8488 | 0.1307 |
69
+ | 9.2682 | 8.0 | 152 | 8.3858 | 0.0536 | 0.8928 | 0.0985 | 0.4086 | 0.7496 | 0.2234 | 0.0532 | 0.8642 | 0.1001 | 0.1551 | 0.8642 | 0.1342 |
70
+ | 9.1804 | 9.0 | 171 | 8.3447 | 0.0536 | 0.8928 | 0.0985 | 0.3996 | 0.7503 | 0.2145 | 0.0531 | 0.8642 | 0.1001 | 0.1581 | 0.8642 | 0.1365 |
71
+ | 8.8361 | 10.0 | 190 | 8.3259 | 0.0548 | 0.9029 | 0.1006 | 0.3664 | 0.7660 | 0.2045 | 0.0544 | 0.8735 | 0.1024 | 0.1465 | 0.8735 | 0.1391 |
72
 
73
 
74
  ### Framework versions
runs/Oct26_21-07-52_icuff-Z790-UD/events.out.tfevents.1729987672.icuff-Z790-UD.65127.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f61835e101c75072154abf20a640e21135955f15dea06cc76a32404502895688
3
- size 21406
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7279f3326a1c443c3ab0a84b68befd82162668dc3e837044e1a37989e20cb359
3
+ size 22714