Hugging Face's logo

NaeemCSECUET18
/
banglabert-MLTC-BB1

Text Classification
Transformers
Safetensors
electra
Generated from Trainer
Inference Endpoints
Model card Files Community
NaeemCSECUET18 commited on
Commit
12e0bb0
•
1 Parent(s): 1d6d3aa

End of training

Browse files
Files changed (1) hide show
  1. README.md +20 -19
README.md CHANGED
@@ -17,13 +17,14 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [csebuetnlp/banglabert](https://huggingface.co/csebuetnlp/banglabert) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.4011
21
- - F1: 0.8602
22
- - Roc Auc: 0.8579
23
- - Accuracy: 0.5758
24
- - Hamming Loss: 0.1420
25
- - Jaccard Score: 0.7547
26
- - Zero One Loss: 0.4242
 
27
 
28
  ## Model description
29
 
@@ -52,18 +53,18 @@ The following hyperparameters were used during training:
52
 
53
  ### Training results
54
 
55
- | Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy | Hamming Loss | Jaccard Score | Zero One Loss |
56
- |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|:------------:|:-------------:|:-------------:|
57
- | 0.2462 | 1.0 | 49 | 0.3759 | 0.8582 | 0.8534 | 0.5758 | 0.1465 | 0.7516 | 0.4242 |
58
- | 0.2099 | 2.0 | 98 | 0.3534 | 0.8656 | 0.8650 | 0.5964 | 0.1350 | 0.7630 | 0.4036 |
59
- | 0.2067 | 3.0 | 147 | 0.3660 | 0.8613 | 0.8599 | 0.5861 | 0.1401 | 0.7564 | 0.4139 |
60
- | 0.168 | 4.0 | 196 | 0.3672 | 0.8582 | 0.8567 | 0.5835 | 0.1433 | 0.7517 | 0.4165 |
61
- | 0.1425 | 5.0 | 245 | 0.3745 | 0.8555 | 0.8547 | 0.5656 | 0.1452 | 0.7475 | 0.4344 |
62
- | 0.1545 | 6.0 | 294 | 0.3894 | 0.8544 | 0.8522 | 0.5578 | 0.1478 | 0.7459 | 0.4422 |
63
- | 0.1115 | 7.0 | 343 | 0.3995 | 0.8579 | 0.8560 | 0.5681 | 0.1440 | 0.7511 | 0.4319 |
64
- | 0.1158 | 8.0 | 392 | 0.4054 | 0.8580 | 0.8554 | 0.5681 | 0.1446 | 0.7514 | 0.4319 |
65
- | 0.1055 | 9.0 | 441 | 0.3996 | 0.8575 | 0.8560 | 0.5681 | 0.1440 | 0.7506 | 0.4319 |
66
- | 0.105 | 10.0 | 490 | 0.4011 | 0.8602 | 0.8579 | 0.5758 | 0.1420 | 0.7547 | 0.4242 |
67
 
68
 
69
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [csebuetnlp/banglabert](https://huggingface.co/csebuetnlp/banglabert) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.4352
21
+ - F1: 0.8638
22
+ - F1 Weighted: 0.8620
23
+ - Roc Auc: 0.8618
24
+ - Accuracy: 0.5835
25
+ - Hamming Loss: 0.1382
26
+ - Jaccard Score: 0.7603
27
+ - Zero One Loss: 0.4165
28
 
29
  ## Model description
30
 
 
53
 
54
  ### Training results
55
 
56
+ | Training Loss | Epoch | Step | Validation Loss | F1 | F1 Weighted | Roc Auc | Accuracy | Hamming Loss | Jaccard Score | Zero One Loss |
57
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:-----------:|:-------:|:--------:|:------------:|:-------------:|:-------------:|
58
+ | 0.1781 | 1.0 | 49 | 0.4033 | 0.8538 | 0.8508 | 0.8476 | 0.5527 | 0.1523 | 0.7449 | 0.4473 |
59
+ | 0.1593 | 2.0 | 98 | 0.3744 | 0.8575 | 0.8548 | 0.8560 | 0.5758 | 0.1440 | 0.7506 | 0.4242 |
60
+ | 0.1733 | 3.0 | 147 | 0.3996 | 0.8564 | 0.8532 | 0.8554 | 0.5707 | 0.1446 | 0.7489 | 0.4293 |
61
+ | 0.1254 | 4.0 | 196 | 0.3902 | 0.8539 | 0.8515 | 0.8528 | 0.5604 | 0.1472 | 0.7450 | 0.4396 |
62
+ | 0.1067 | 5.0 | 245 | 0.4059 | 0.8568 | 0.8540 | 0.8547 | 0.5630 | 0.1452 | 0.7494 | 0.4370 |
63
+ | 0.1157 | 6.0 | 294 | 0.4180 | 0.8631 | 0.8608 | 0.8605 | 0.5835 | 0.1395 | 0.7592 | 0.4165 |
64
+ | 0.0807 | 7.0 | 343 | 0.4307 | 0.8579 | 0.8552 | 0.8560 | 0.5656 | 0.1440 | 0.7511 | 0.4344 |
65
+ | 0.0741 | 8.0 | 392 | 0.4361 | 0.8631 | 0.8608 | 0.8612 | 0.5758 | 0.1388 | 0.7592 | 0.4242 |
66
+ | 0.0716 | 9.0 | 441 | 0.4295 | 0.8615 | 0.8601 | 0.8599 | 0.5835 | 0.1401 | 0.7567 | 0.4165 |
67
+ | 0.0707 | 10.0 | 490 | 0.4352 | 0.8638 | 0.8620 | 0.8618 | 0.5835 | 0.1382 | 0.7603 | 0.4165 |
68
 
69
 
70
  ### Framework versions