Hugging Face's logo

arturkulmizev
/
deberta-base.as.finetuned

Fill-Mask
Transformers
Safetensors
deberta
Generated from Trainer
Inference Endpoints
Model card Files Community
arturkulmizev commited on
Commit
149d263
•
1 Parent(s): 627d646

End of training

Browse files
Files changed (2) hide show
  1. README.md +23 -66
  2. model.safetensors +1 -1
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 [](https://huggingface.co/) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
- - Loss: 5.6126
17
 
18
  ## Model description
19
 
@@ -38,77 +38,34 @@ 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: 30
42
  - mixed_precision_training: Native AMP
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:-----:|:---------------:|
48
- | 6.6155 | 0.47 | 1000 | 6.5530 |
49
- | 6.4654 | 0.93 | 2000 | 6.4358 |
50
- | 6.3684 | 1.4 | 3000 | 6.3436 |
51
- | 6.2843 | 1.87 | 4000 | 6.2675 |
52
- | 6.2171 | 2.33 | 5000 | 6.2000 |
53
- | 6.1687 | 2.8 | 6000 | 6.1522 |
54
- | 6.122 | 3.27 | 7000 | 6.1094 |
55
- | 6.0828 | 3.73 | 8000 | 6.0599 |
56
- | 6.0346 | 4.2 | 9000 | 6.0132 |
57
- | 5.9935 | 4.67 | 10000 | 5.9940 |
58
- | 5.9553 | 5.13 | 11000 | 5.9589 |
59
- | 5.9347 | 5.6 | 12000 | 5.9315 |
60
- | 5.9105 | 6.07 | 13000 | 5.8996 |
61
- | 5.8851 | 6.53 | 14000 | 5.8859 |
62
- | 5.8632 | 7.0 | 15000 | 5.8626 |
63
- | 5.847 | 7.47 | 16000 | 5.8495 |
64
- | 5.8393 | 7.93 | 17000 | 5.8276 |
65
- | 5.8041 | 8.4 | 18000 | 5.8096 |
66
- | 5.7922 | 8.87 | 19000 | 5.7968 |
67
- | 5.7758 | 9.33 | 20000 | 5.7869 |
68
- | 5.7656 | 9.8 | 21000 | 5.7795 |
69
- | 5.7533 | 10.27 | 22000 | 5.7652 |
70
- | 5.7484 | 10.73 | 23000 | 5.7511 |
71
- | 5.7352 | 11.2 | 24000 | 5.7417 |
72
- | 5.7215 | 11.67 | 25000 | 5.7327 |
73
- | 5.7126 | 12.13 | 26000 | 5.7352 |
74
- | 5.7048 | 12.6 | 27000 | 5.7233 |
75
- | 5.6887 | 13.07 | 28000 | 5.7106 |
76
- | 5.6892 | 13.53 | 29000 | 5.7003 |
77
- | 5.6819 | 14.0 | 30000 | 5.6993 |
78
- | 5.6665 | 14.47 | 31000 | 5.7001 |
79
- | 5.6704 | 14.93 | 32000 | 5.6936 |
80
- | 5.6608 | 15.4 | 33000 | 5.6828 |
81
- | 5.6575 | 15.87 | 34000 | 5.6795 |
82
- | 5.6552 | 16.33 | 35000 | 5.6686 |
83
- | 5.647 | 16.8 | 36000 | 5.6676 |
84
- | 5.639 | 17.27 | 37000 | 5.6659 |
85
- | 5.6346 | 17.73 | 38000 | 5.6644 |
86
- | 5.633 | 18.2 | 39000 | 5.6648 |
87
- | 5.6276 | 18.67 | 40000 | 5.6569 |
88
- | 5.6252 | 19.13 | 41000 | 5.6488 |
89
- | 5.6254 | 19.6 | 42000 | 5.6468 |
90
- | 5.6198 | 20.07 | 43000 | 5.6485 |
91
- | 5.6104 | 20.53 | 44000 | 5.6394 |
92
- | 5.6117 | 21.0 | 45000 | 5.6417 |
93
- | 5.6048 | 21.47 | 46000 | 5.6362 |
94
- | 5.5973 | 21.93 | 47000 | 5.6378 |
95
- | 5.5923 | 22.4 | 48000 | 5.6312 |
96
- | 5.6047 | 22.87 | 49000 | 5.6308 |
97
- | 5.5955 | 23.33 | 50000 | 5.6341 |
98
- | 5.5947 | 23.8 | 51000 | 5.6257 |
99
- | 5.5979 | 24.27 | 52000 | 5.6330 |
100
- | 5.5924 | 24.73 | 53000 | 5.6200 |
101
- | 5.5996 | 25.2 | 54000 | 5.6175 |
102
- | 5.5823 | 25.66 | 55000 | 5.6245 |
103
- | 5.5884 | 26.13 | 56000 | 5.6207 |
104
- | 5.5828 | 26.6 | 57000 | 5.6184 |
105
- | 5.5844 | 27.06 | 58000 | 5.6255 |
106
- | 5.5793 | 27.53 | 59000 | 5.6269 |
107
- | 5.5809 | 28.0 | 60000 | 5.6185 |
108
- | 5.5786 | 28.46 | 61000 | 5.6185 |
109
- | 5.5765 | 28.93 | 62000 | 5.6204 |
110
- | 5.5814 | 29.4 | 63000 | 5.6257 |
111
- | 5.5837 | 29.86 | 64000 | 5.6163 |
112
 
113
 
114
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Loss: 5.9191
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: 10
42
  - mixed_precision_training: Native AMP
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:-----:|:---------------:|
48
+ | 6.6297 | 0.46 | 1000 | 6.5539 |
49
+ | 6.4688 | 0.92 | 2000 | 6.4491 |
50
+ | 6.3844 | 1.38 | 3000 | 6.3672 |
51
+ | 6.3167 | 1.84 | 4000 | 6.2918 |
52
+ | 6.252 | 2.3 | 5000 | 6.2378 |
53
+ | 6.2027 | 2.76 | 6000 | 6.1876 |
54
+ | 6.1586 | 3.22 | 7000 | 6.1400 |
55
+ | 6.1194 | 3.68 | 8000 | 6.1085 |
56
+ | 6.0882 | 4.14 | 9000 | 6.0751 |
57
+ | 6.0543 | 4.6 | 10000 | 6.0519 |
58
+ | 6.023 | 5.06 | 11000 | 6.0291 |
59
+ | 6.0118 | 5.52 | 12000 | 6.0053 |
60
+ | 5.9866 | 5.98 | 13000 | 5.9925 |
61
+ | 5.9737 | 6.44 | 14000 | 5.9772 |
62
+ | 5.9609 | 6.9 | 15000 | 5.9589 |
63
+ | 5.9537 | 7.36 | 16000 | 5.9472 |
64
+ | 5.9366 | 7.82 | 17000 | 5.9419 |
65
+ | 5.9282 | 8.28 | 18000 | 5.9283 |
66
+ | 5.926 | 8.74 | 19000 | 5.9246 |
67
+ | 5.9194 | 9.2 | 20000 | 5.9283 |
68
+ | 5.9172 | 9.66 | 21000 | 5.9273 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
69
 
70
 
71
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:11ba52338d1d6a7513db57b7d39fbecb15103977bed389468a17f02b34b81992
3
  size 21014480
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9bf75b8842a940a8c507d6b683f97b3627e58407ad448122260213f54e7a1e3
3
  size 21014480