Hugging Face's logo

powerpuf-bot
/
mdeberta-v3-th-wiki-qa_hyp-params

Question Answering
Transformers
TensorBoard
Safetensors
deberta-v2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Chananchida commited on
Commit
18a35eb
路
1 Parent(s): f4f60e9

Model save

Browse files
Files changed (1) hide show
  1. README.md +106 -0
README.md ADDED
@@ -0,0 +1,106 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: timpal0l/mdeberta-v3-base-squad2
4
+ tags:
5
+ - generated_from_trainer
6
+ model-index:
7
+ - name: model1
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ # model1
15
+
16
+ This model is a fine-tuned version of [timpal0l/mdeberta-v3-base-squad2](https://huggingface.co/timpal0l/mdeberta-v3-base-squad2) on the None dataset.
17
+ It achieves the following results on the evaluation set:
18
+ - Loss: 2.4437
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 1.2922909480977358e-06
38
+ - train_batch_size: 4
39
+ - eval_batch_size: 4
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
42
+ - lr_scheduler_type: linear
43
+ - training_steps: 500
44
+
45
+ ### Training results
46
+
47
+ | Training Loss | Epoch | Step | Validation Loss |
48
+ |:-------------:|:-----:|:----:|:---------------:|
49
+ | 4.2146 | 0.0 | 10 | 4.1627 |
50
+ | 4.1667 | 0.0 | 20 | 3.9059 |
51
+ | 4.0555 | 0.01 | 30 | 3.7982 |
52
+ | 3.9331 | 0.01 | 40 | 3.7342 |
53
+ | 3.8012 | 0.01 | 50 | 3.6719 |
54
+ | 3.7713 | 0.01 | 60 | 3.6077 |
55
+ | 3.8391 | 0.02 | 70 | 3.5548 |
56
+ | 3.8842 | 0.02 | 80 | 3.5134 |
57
+ | 3.6894 | 0.02 | 90 | 3.4823 |
58
+ | 3.5359 | 0.02 | 100 | 3.4466 |
59
+ | 3.6247 | 0.03 | 110 | 3.4096 |
60
+ | 3.6347 | 0.03 | 120 | 3.3807 |
61
+ | 3.5752 | 0.03 | 130 | 3.3459 |
62
+ | 3.467 | 0.03 | 140 | 3.2778 |
63
+ | 3.6188 | 0.04 | 150 | 3.2198 |
64
+ | 3.444 | 0.04 | 160 | 3.1880 |
65
+ | 3.4635 | 0.04 | 170 | 3.1494 |
66
+ | 3.3998 | 0.04 | 180 | 3.1107 |
67
+ | 3.1465 | 0.04 | 190 | 3.0675 |
68
+ | 3.4321 | 0.05 | 200 | 3.0380 |
69
+ | 3.3174 | 0.05 | 210 | 3.0122 |
70
+ | 3.6018 | 0.05 | 220 | 2.9566 |
71
+ | 3.4178 | 0.05 | 230 | 2.9099 |
72
+ | 3.2037 | 0.06 | 240 | 2.8755 |
73
+ | 3.3974 | 0.06 | 250 | 2.8493 |
74
+ | 3.109 | 0.06 | 260 | 2.8209 |
75
+ | 3.1127 | 0.06 | 270 | 2.7751 |
76
+ | 3.2408 | 0.07 | 280 | 2.7458 |
77
+ | 3.274 | 0.07 | 290 | 2.7211 |
78
+ | 3.0695 | 0.07 | 300 | 2.6946 |
79
+ | 2.9757 | 0.07 | 310 | 2.6713 |
80
+ | 3.0846 | 0.08 | 320 | 2.6415 |
81
+ | 3.0576 | 0.08 | 330 | 2.6209 |
82
+ | 2.8623 | 0.08 | 340 | 2.6041 |
83
+ | 3.165 | 0.08 | 350 | 2.5913 |
84
+ | 2.8874 | 0.08 | 360 | 2.5797 |
85
+ | 3.046 | 0.09 | 370 | 2.5627 |
86
+ | 2.8727 | 0.09 | 380 | 2.5391 |
87
+ | 2.7942 | 0.09 | 390 | 2.5188 |
88
+ | 2.7494 | 0.09 | 400 | 2.5031 |
89
+ | 2.8419 | 0.1 | 410 | 2.4905 |
90
+ | 2.8411 | 0.1 | 420 | 2.4792 |
91
+ | 2.9188 | 0.1 | 430 | 2.4696 |
92
+ | 2.9239 | 0.1 | 440 | 2.4622 |
93
+ | 3.0064 | 0.11 | 450 | 2.4551 |
94
+ | 2.9781 | 0.11 | 460 | 2.4504 |
95
+ | 2.8582 | 0.11 | 470 | 2.4483 |
96
+ | 2.8701 | 0.11 | 480 | 2.4456 |
97
+ | 2.7012 | 0.12 | 490 | 2.4442 |
98
+ | 2.827 | 0.12 | 500 | 2.4437 |
99
+
100
+
101
+ ### Framework versions
102
+
103
+ - Transformers 4.35.2
104
+ - Pytorch 2.1.0+cu121
105
+ - Datasets 2.15.0
106
+ - Tokenizers 0.15.0