Hugging Face's logo

ksabeh
/
dbert-finetuned

Fill-Mask
Transformers
PyTorch
distilbert
Generated from Trainer
Model card Files
xet
Community
1
ksabeh commited on
Commit
aa0bf53
·
1 Parent(s): 731596e

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -69
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 2.6212
18
 
19
  ## Model description
20
 
@@ -39,79 +39,22 @@ The following hyperparameters were used during training:
39
  - seed: 42
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
- - num_epochs: 5
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:-----:|:---------------:|
48
- | 4.2642 | 0.06 | 500 | 3.8662 |
49
- | 3.862 | 0.13 | 1000 | 3.6551 |
50
- | 3.7033 | 0.19 | 1500 | 3.5304 |
51
- | 3.594 | 0.25 | 2000 | 3.4363 |
52
- | 3.5172 | 0.31 | 2500 | 3.3680 |
53
- | 3.4492 | 0.38 | 3000 | 3.3045 |
54
- | 3.3982 | 0.44 | 3500 | 3.2585 |
55
- | 3.3494 | 0.5 | 4000 | 3.2151 |
56
- | 3.3035 | 0.57 | 4500 | 3.1740 |
57
- | 3.2706 | 0.63 | 5000 | 3.1383 |
58
- | 3.2441 | 0.69 | 5500 | 3.1118 |
59
- | 3.2028 | 0.76 | 6000 | 3.0817 |
60
- | 3.1827 | 0.82 | 6500 | 3.0541 |
61
- | 3.1629 | 0.88 | 7000 | 3.0330 |
62
- | 3.1345 | 0.94 | 7500 | 3.0085 |
63
- | 3.1091 | 1.01 | 8000 | 2.9891 |
64
- | 3.0844 | 1.07 | 8500 | 2.9714 |
65
- | 3.0672 | 1.13 | 9000 | 2.9549 |
66
- | 3.0445 | 1.2 | 9500 | 2.9333 |
67
- | 3.0311 | 1.26 | 10000 | 2.9234 |
68
- | 3.019 | 1.32 | 10500 | 2.9044 |
69
- | 3.0036 | 1.38 | 11000 | 2.8889 |
70
- | 2.9957 | 1.45 | 11500 | 2.8787 |
71
- | 2.9752 | 1.51 | 12000 | 2.8641 |
72
- | 2.965 | 1.57 | 12500 | 2.8543 |
73
- | 2.95 | 1.64 | 13000 | 2.8408 |
74
- | 2.9406 | 1.7 | 13500 | 2.8314 |
75
- | 2.93 | 1.76 | 14000 | 2.8202 |
76
- | 2.9278 | 1.82 | 14500 | 2.8096 |
77
- | 2.914 | 1.89 | 15000 | 2.8008 |
78
- | 2.9052 | 1.95 | 15500 | 2.7915 |
79
- | 2.8946 | 2.01 | 16000 | 2.7853 |
80
- | 2.8806 | 2.08 | 16500 | 2.7737 |
81
- | 2.8665 | 2.14 | 17000 | 2.7657 |
82
- | 2.8601 | 2.2 | 17500 | 2.7600 |
83
- | 2.8633 | 2.27 | 18000 | 2.7531 |
84
- | 2.8515 | 2.33 | 18500 | 2.7447 |
85
- | 2.8452 | 2.39 | 19000 | 2.7372 |
86
- | 2.8345 | 2.45 | 19500 | 2.7298 |
87
- | 2.8404 | 2.52 | 20000 | 2.7222 |
88
- | 2.8223 | 2.58 | 20500 | 2.7144 |
89
- | 2.8177 | 2.64 | 21000 | 2.7108 |
90
- | 2.8099 | 2.71 | 21500 | 2.7066 |
91
- | 2.8008 | 2.77 | 22000 | 2.6971 |
92
- | 2.7979 | 2.83 | 22500 | 2.6943 |
93
- | 2.7993 | 2.89 | 23000 | 2.6908 |
94
- | 2.7864 | 2.96 | 23500 | 2.6879 |
95
- | 2.7891 | 3.02 | 24000 | 2.6797 |
96
- | 2.7766 | 3.08 | 24500 | 2.6751 |
97
- | 2.7757 | 3.15 | 25000 | 2.6724 |
98
- | 2.7674 | 3.21 | 25500 | 2.6664 |
99
- | 2.7639 | 3.27 | 26000 | 2.6621 |
100
- | 2.7664 | 3.34 | 26500 | 2.6589 |
101
- | 2.7509 | 3.4 | 27000 | 2.6530 |
102
- | 2.7519 | 3.46 | 27500 | 2.6503 |
103
- | 2.7529 | 3.52 | 28000 | 2.6494 |
104
- | 2.7484 | 3.59 | 28500 | 2.6480 |
105
- | 2.7454 | 3.65 | 29000 | 2.6394 |
106
- | 2.7426 | 3.71 | 29500 | 2.6407 |
107
- | 2.7425 | 3.78 | 30000 | 2.6384 |
108
- | 2.7302 | 3.84 | 30500 | 2.6323 |
109
- | 2.7419 | 3.9 | 31000 | 2.6308 |
110
- | 2.7365 | 3.96 | 31500 | 2.6264 |
111
- | 2.7246 | 4.03 | 32000 | 2.6263 |
112
- | 2.7219 | 4.09 | 32500 | 2.6210 |
113
- | 2.7223 | 4.15 | 33000 | 2.6216 |
114
- | 2.718 | 4.22 | 33500 | 2.6212 |
115
 
116
 
117
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 2.3699
18
 
19
  ## Model description
20
 
 
39
  - seed: 42
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
+ - num_epochs: 10
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:-----:|:---------------:|
48
+ | 3.1176 | 1.0 | 7946 | 2.9781 |
49
+ | 2.873 | 2.0 | 15892 | 2.7518 |
50
+ | 2.7337 | 3.0 | 23838 | 2.6254 |
51
+ | 2.6536 | 4.0 | 31784 | 2.5434 |
52
+ | 2.5838 | 5.0 | 39730 | 2.4846 |
53
+ | 2.5376 | 6.0 | 47676 | 2.4394 |
54
+ | 2.513 | 7.0 | 55622 | 2.4142 |
55
+ | 2.4814 | 8.0 | 63568 | 2.3870 |
56
+ | 2.4737 | 9.0 | 71514 | 2.3759 |
57
+ | 2.467 | 10.0 | 79460 | 2.3699 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
 
59
 
60
  ### Framework versions