Hugging Face's logo

Mediocre-Judge
/
bengali_qa_model_AGGRO_roberta-base

Question Answering
Transformers
Safetensors
roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
Mediocre-Judge commited on
Commit
d0c58db
路
verified 路
1 Parent(s): 11ff7f7

End of training

Browse files
Files changed (1) hide show
  1. README.md +54 -15
README.md CHANGED
@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 4.2878
20
- - Exact Match: 24.7143
21
- - F1 Score: 63.3563
22
 
23
  ## Model description
24
 
@@ -37,7 +37,7 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - learning_rate: 2e-05
41
  - train_batch_size: 4
42
  - eval_batch_size: 4
43
  - seed: 3407
@@ -52,17 +52,56 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Exact Match | F1 Score |
54
  |:-------------:|:------:|:----:|:---------------:|:-----------:|:--------:|
55
- | 5.9561 | 0.0053 | 1 | 5.9848 | 0.0 | 46.8295 |
56
- | 5.9358 | 0.0107 | 2 | 5.9287 | 0.0 | 46.9461 |
57
- | 5.9153 | 0.0160 | 3 | 5.8180 | 0.0 | 48.6568 |
58
- | 5.8287 | 0.0214 | 4 | 5.6566 | 0.0752 | 52.8081 |
59
- | 5.6816 | 0.0267 | 5 | 5.4437 | 5.5639 | 58.2723 |
60
- | 5.4698 | 0.0321 | 6 | 5.1322 | 20.0 | 62.7335 |
61
- | 5.2037 | 0.0374 | 7 | 4.7557 | 29.7744 | 65.3137 |
62
- | 4.8684 | 0.0428 | 8 | 4.2898 | 26.8421 | 65.3919 |
63
- | 4.4376 | 0.0481 | 9 | 3.7808 | 19.6241 | 63.6051 |
64
- | 3.9423 | 0.0535 | 10 | 3.3519 | 14.2105 | 62.4085 |
65
- | 3.7279 | 0.0588 | 11 | 3.0749 | 12.7820 | 61.8234 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
 
67
 
68
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.7412
20
+ - Exact Match: 80.5714
21
+ - F1 Score: 86.1024
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
  - train_batch_size: 4
42
  - eval_batch_size: 4
43
  - seed: 3407
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Exact Match | F1 Score |
54
  |:-------------:|:------:|:----:|:---------------:|:-----------:|:--------:|
55
+ | 5.9404 | 0.0053 | 1 | 5.9382 | 0.0 | 0.5953 |
56
+ | 5.9215 | 0.0107 | 2 | 5.9105 | 0.0 | 0.6034 |
57
+ | 5.9271 | 0.0160 | 3 | 5.8553 | 0.0 | 0.8934 |
58
+ | 5.8875 | 0.0214 | 4 | 5.7721 | 0.0 | 1.7912 |
59
+ | 5.805 | 0.0267 | 5 | 5.6573 | 0.6767 | 8.4782 |
60
+ | 5.6871 | 0.0321 | 6 | 5.5054 | 23.4586 | 42.2185 |
61
+ | 5.5539 | 0.0374 | 7 | 5.3433 | 43.7594 | 59.1302 |
62
+ | 5.4103 | 0.0428 | 8 | 5.1687 | 48.6466 | 62.0339 |
63
+ | 5.2151 | 0.0481 | 9 | 4.9787 | 49.1729 | 62.6107 |
64
+ | 5.0079 | 0.0535 | 10 | 4.7613 | 48.1955 | 62.6351 |
65
+ | 4.9345 | 0.0588 | 11 | 4.5065 | 47.9699 | 63.1193 |
66
+ | 4.6229 | 0.0641 | 12 | 4.2115 | 47.5188 | 63.2726 |
67
+ | 4.4631 | 0.0695 | 13 | 3.9029 | 52.2556 | 65.5815 |
68
+ | 4.1463 | 0.0748 | 14 | 3.6499 | 58.9474 | 68.4275 |
69
+ | 3.8714 | 0.0802 | 15 | 3.4862 | 54.7368 | 68.5473 |
70
+ | 3.5838 | 0.0855 | 16 | 3.3115 | 49.8496 | 68.1843 |
71
+ | 3.5844 | 0.0909 | 17 | 3.0756 | 50.4511 | 68.9449 |
72
+ | 3.3173 | 0.0962 | 18 | 2.7914 | 52.0301 | 69.3946 |
73
+ | 3.0913 | 0.1016 | 19 | 2.5020 | 54.2857 | 69.9967 |
74
+ | 2.9282 | 0.1069 | 20 | 2.2443 | 55.2632 | 69.9488 |
75
+ | 2.711 | 0.1123 | 21 | 2.0197 | 54.2857 | 69.7818 |
76
+ | 2.5538 | 0.1176 | 22 | 1.8205 | 54.5113 | 70.1570 |
77
+ | 2.1309 | 0.1230 | 23 | 1.6450 | 55.8647 | 71.1004 |
78
+ | 2.1398 | 0.1283 | 24 | 1.4972 | 57.3684 | 71.6500 |
79
+ | 2.0536 | 0.1336 | 25 | 1.3786 | 59.3233 | 72.3145 |
80
+ | 1.8463 | 0.1390 | 26 | 1.2891 | 61.2030 | 73.0502 |
81
+ | 1.7279 | 0.1443 | 27 | 1.2158 | 62.7820 | 73.8093 |
82
+ | 1.6699 | 0.1497 | 28 | 1.1546 | 64.4361 | 74.8352 |
83
+ | 1.516 | 0.1550 | 29 | 1.1015 | 66.6917 | 75.9196 |
84
+ | 1.4137 | 0.1604 | 30 | 1.0558 | 68.3459 | 77.1076 |
85
+ | 1.3907 | 0.1657 | 31 | 1.0158 | 69.0977 | 77.7101 |
86
+ | 1.1152 | 0.1711 | 32 | 0.9823 | 69.7744 | 78.3412 |
87
+ | 1.0033 | 0.1764 | 33 | 0.9529 | 70.4511 | 78.8287 |
88
+ | 1.0707 | 0.1818 | 34 | 0.9263 | 71.5038 | 79.7847 |
89
+ | 1.0864 | 0.1871 | 35 | 0.9025 | 72.3308 | 80.4039 |
90
+ | 1.1194 | 0.1924 | 36 | 0.8810 | 72.9323 | 80.9981 |
91
+ | 0.8841 | 0.1978 | 37 | 0.8609 | 73.6842 | 81.5823 |
92
+ | 1.1392 | 0.2031 | 38 | 0.8420 | 74.6617 | 82.4632 |
93
+ | 1.0528 | 0.2085 | 39 | 0.8252 | 75.1880 | 82.8512 |
94
+ | 1.0969 | 0.2138 | 40 | 0.8095 | 75.6391 | 83.3666 |
95
+ | 1.0979 | 0.2192 | 41 | 0.7943 | 76.0902 | 83.7383 |
96
+ | 0.809 | 0.2245 | 42 | 0.7808 | 76.3158 | 83.7564 |
97
+ | 0.901 | 0.2299 | 43 | 0.7688 | 77.4436 | 84.4723 |
98
+ | 0.8812 | 0.2352 | 44 | 0.7588 | 78.3459 | 85.1076 |
99
+ | 0.8275 | 0.2406 | 45 | 0.7510 | 78.7970 | 85.4737 |
100
+ | 1.0578 | 0.2459 | 46 | 0.7451 | 79.0226 | 85.5693 |
101
+ | 0.8329 | 0.2513 | 47 | 0.7412 | 79.2481 | 85.8848 |
102
+ | 0.8682 | 0.2566 | 48 | 0.7390 | 79.3233 | 85.9460 |
103
+ | 1.0013 | 0.2619 | 49 | 0.7380 | 79.3985 | 86.0087 |
104
+ | 1.0948 | 0.2673 | 50 | 0.7377 | 79.4737 | 86.0459 |
105
 
106
 
107
  ### Framework versions