Hugging Face's logo

JasperV13
/
DarijaBERT-finetuned-Arabic-SQuAD

Question Answering
Transformers
PyTorch
TensorBoard
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
2
JasperV13 commited on
Commit
288432f
•
1 Parent(s): 4f7732b

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -102
README.md CHANGED
@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  This model is a fine-tuned version of [SI2M-Lab/DarijaBERT](https://huggingface.co/SI2M-Lab/DarijaBERT) on the None dataset.
15
  It achieves the following results on the evaluation set:
16
- - Loss: 7.3855
17
 
18
  ## Model description
19
 
@@ -38,112 +38,16 @@ The following hyperparameters were used during training:
38
  - seed: 42
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: linear
41
- - num_epochs: 100
42
 
43
  ### Training results
44
 
45
  | Training Loss | Epoch | Step | Validation Loss |
46
  |:-------------:|:-----:|:----:|:---------------:|
47
- | No log | 1.0 | 4 | 3.6892 |
48
- | No log | 2.0 | 8 | 4.0008 |
49
- | No log | 3.0 | 12 | 4.2185 |
50
- | No log | 4.0 | 16 | 4.4023 |
51
- | No log | 5.0 | 20 | 4.6039 |
52
- | No log | 6.0 | 24 | 4.9472 |
53
- | No log | 7.0 | 28 | 5.1248 |
54
- | No log | 8.0 | 32 | 5.1359 |
55
- | No log | 9.0 | 36 | 5.2324 |
56
- | No log | 10.0 | 40 | 5.3148 |
57
- | No log | 11.0 | 44 | 5.3485 |
58
- | No log | 12.0 | 48 | 5.4560 |
59
- | No log | 13.0 | 52 | 5.5046 |
60
- | No log | 14.0 | 56 | 5.6621 |
61
- | No log | 15.0 | 60 | 5.7310 |
62
- | No log | 16.0 | 64 | 5.7329 |
63
- | No log | 17.0 | 68 | 5.8181 |
64
- | No log | 18.0 | 72 | 5.8559 |
65
- | No log | 19.0 | 76 | 5.7975 |
66
- | No log | 20.0 | 80 | 5.9184 |
67
- | No log | 21.0 | 84 | 6.0450 |
68
- | No log | 22.0 | 88 | 5.9877 |
69
- | No log | 23.0 | 92 | 5.9378 |
70
- | No log | 24.0 | 96 | 5.8960 |
71
- | No log | 25.0 | 100 | 6.0414 |
72
- | No log | 26.0 | 104 | 6.1865 |
73
- | No log | 27.0 | 108 | 6.0761 |
74
- | No log | 28.0 | 112 | 6.0355 |
75
- | No log | 29.0 | 116 | 6.2873 |
76
- | No log | 30.0 | 120 | 6.4127 |
77
- | No log | 31.0 | 124 | 6.0708 |
78
- | No log | 32.0 | 128 | 6.0292 |
79
- | No log | 33.0 | 132 | 6.3732 |
80
- | No log | 34.0 | 136 | 6.5437 |
81
- | No log | 35.0 | 140 | 6.5159 |
82
- | No log | 36.0 | 144 | 6.6191 |
83
- | No log | 37.0 | 148 | 6.6116 |
84
- | No log | 38.0 | 152 | 6.5964 |
85
- | No log | 39.0 | 156 | 6.8168 |
86
- | No log | 40.0 | 160 | 6.9130 |
87
- | No log | 41.0 | 164 | 6.9621 |
88
- | No log | 42.0 | 168 | 6.7498 |
89
- | No log | 43.0 | 172 | 6.7138 |
90
- | No log | 44.0 | 176 | 7.0168 |
91
- | No log | 45.0 | 180 | 7.1698 |
92
- | No log | 46.0 | 184 | 6.8416 |
93
- | No log | 47.0 | 188 | 6.5821 |
94
- | No log | 48.0 | 192 | 7.1047 |
95
- | No log | 49.0 | 196 | 7.2546 |
96
- | No log | 50.0 | 200 | 7.2263 |
97
- | No log | 51.0 | 204 | 6.9599 |
98
- | No log | 52.0 | 208 | 7.0655 |
99
- | No log | 53.0 | 212 | 7.0357 |
100
- | No log | 54.0 | 216 | 6.8836 |
101
- | No log | 55.0 | 220 | 6.8562 |
102
- | No log | 56.0 | 224 | 7.0020 |
103
- | No log | 57.0 | 228 | 7.0476 |
104
- | No log | 58.0 | 232 | 7.0757 |
105
- | No log | 59.0 | 236 | 7.0527 |
106
- | No log | 60.0 | 240 | 7.0973 |
107
- | No log | 61.0 | 244 | 7.1660 |
108
- | No log | 62.0 | 248 | 7.1815 |
109
- | No log | 63.0 | 252 | 7.1370 |
110
- | No log | 64.0 | 256 | 7.0314 |
111
- | No log | 65.0 | 260 | 7.0141 |
112
- | No log | 66.0 | 264 | 7.1435 |
113
- | No log | 67.0 | 268 | 7.2527 |
114
- | No log | 68.0 | 272 | 7.2785 |
115
- | No log | 69.0 | 276 | 7.2528 |
116
- | No log | 70.0 | 280 | 7.1664 |
117
- | No log | 71.0 | 284 | 7.2089 |
118
- | No log | 72.0 | 288 | 7.2244 |
119
- | No log | 73.0 | 292 | 7.1812 |
120
- | No log | 74.0 | 296 | 7.0677 |
121
- | No log | 75.0 | 300 | 7.0805 |
122
- | No log | 76.0 | 304 | 7.0525 |
123
- | No log | 77.0 | 308 | 6.9768 |
124
- | No log | 78.0 | 312 | 7.0477 |
125
- | No log | 79.0 | 316 | 7.0929 |
126
- | No log | 80.0 | 320 | 7.1310 |
127
- | No log | 81.0 | 324 | 7.1982 |
128
- | No log | 82.0 | 328 | 7.2412 |
129
- | No log | 83.0 | 332 | 7.2509 |
130
- | No log | 84.0 | 336 | 7.2792 |
131
- | No log | 85.0 | 340 | 7.3272 |
132
- | No log | 86.0 | 344 | 7.3738 |
133
- | No log | 87.0 | 348 | 7.4164 |
134
- | No log | 88.0 | 352 | 7.4374 |
135
- | No log | 89.0 | 356 | 7.4510 |
136
- | No log | 90.0 | 360 | 7.4228 |
137
- | No log | 91.0 | 364 | 7.3908 |
138
- | No log | 92.0 | 368 | 7.3702 |
139
- | No log | 93.0 | 372 | 7.3700 |
140
- | No log | 94.0 | 376 | 7.3739 |
141
- | No log | 95.0 | 380 | 7.3786 |
142
- | No log | 96.0 | 384 | 7.3795 |
143
- | No log | 97.0 | 388 | 7.3746 |
144
- | No log | 98.0 | 392 | 7.3792 |
145
- | No log | 99.0 | 396 | 7.3839 |
146
- | No log | 100.0 | 400 | 7.3855 |
147
 
148
 
149
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [SI2M-Lab/DarijaBERT](https://huggingface.co/SI2M-Lab/DarijaBERT) on the None dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Loss: 4.0576
17
 
18
  ## Model description
19
 
 
38
  - seed: 42
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: linear
41
+ - num_epochs: 4
42
 
43
  ### Training results
44
 
45
  | Training Loss | Epoch | Step | Validation Loss |
46
  |:-------------:|:-----:|:----:|:---------------:|
47
+ | No log | 1.0 | 200 | 3.8683 |
48
+ | No log | 2.0 | 400 | 3.8842 |
49
+ | 3.7505 | 3.0 | 600 | 3.9500 |
50
+ | 3.7505 | 4.0 | 800 | 4.0576 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
51
 
52
 
53
  ### Framework versions