dhruvil237 commited on
Commit
d4d631b
·
1 Parent(s): 2eeed2f

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +107 -0
README.md ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - clinc_oos
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: userutterance_classification_ver1
11
+ results:
12
+ - task:
13
+ name: Text Classification
14
+ type: text-classification
15
+ dataset:
16
+ name: clinc_oos
17
+ type: clinc_oos
18
+ config: imbalanced
19
+ split: validation
20
+ args: imbalanced
21
+ metrics:
22
+ - name: Accuracy
23
+ type: accuracy
24
+ value: 0.9538709677419355
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # userutterance_classification_ver1
31
+
32
+ This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on the clinc_oos dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 0.2898
35
+ - Accuracy: 0.9539
36
+
37
+ ## Model description
38
+
39
+ More information needed
40
+
41
+ ## Intended uses & limitations
42
+
43
+ More information needed
44
+
45
+ ## Training and evaluation data
46
+
47
+ More information needed
48
+
49
+ ## Training procedure
50
+
51
+ ### Training hyperparameters
52
+
53
+ The following hyperparameters were used during training:
54
+ - learning_rate: 4e-05
55
+ - train_batch_size: 8
56
+ - eval_batch_size: 8
57
+ - seed: 42
58
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
+ - lr_scheduler_type: linear
60
+ - lr_scheduler_warmup_steps: 130
61
+ - num_epochs: 5
62
+
63
+ ### Training results
64
+
65
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
66
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
67
+ | 4.8334 | 0.15 | 200 | 4.7254 | 0.0748 |
68
+ | 3.4798 | 0.3 | 400 | 3.4244 | 0.2971 |
69
+ | 2.319 | 0.45 | 600 | 2.4423 | 0.5184 |
70
+ | 1.5683 | 0.6 | 800 | 1.7401 | 0.6310 |
71
+ | 0.9625 | 0.75 | 1000 | 1.2750 | 0.7265 |
72
+ | 0.6922 | 0.9 | 1200 | 0.9717 | 0.7761 |
73
+ | 0.5019 | 1.05 | 1400 | 0.8036 | 0.8284 |
74
+ | 0.3538 | 1.2 | 1600 | 0.6690 | 0.8471 |
75
+ | 0.2413 | 1.35 | 1800 | 0.5585 | 0.8713 |
76
+ | 0.2623 | 1.5 | 2000 | 0.4840 | 0.8874 |
77
+ | 0.2103 | 1.66 | 2200 | 0.4261 | 0.9126 |
78
+ | 0.1456 | 1.81 | 2400 | 0.3872 | 0.9152 |
79
+ | 0.1276 | 1.96 | 2600 | 0.3329 | 0.9290 |
80
+ | 0.09 | 2.11 | 2800 | 0.2925 | 0.9432 |
81
+ | 0.0534 | 2.26 | 3000 | 0.2996 | 0.9361 |
82
+ | 0.0588 | 2.41 | 3200 | 0.2951 | 0.9403 |
83
+ | 0.044 | 2.56 | 3400 | 0.3324 | 0.9403 |
84
+ | 0.0535 | 2.71 | 3600 | 0.3155 | 0.9432 |
85
+ | 0.0537 | 2.86 | 3800 | 0.3206 | 0.9419 |
86
+ | 0.1325 | 3.01 | 4000 | 0.2945 | 0.9465 |
87
+ | 0.0611 | 3.16 | 4200 | 0.2903 | 0.9442 |
88
+ | 0.0077 | 3.31 | 4400 | 0.3052 | 0.9477 |
89
+ | 0.0187 | 3.46 | 4600 | 0.2774 | 0.95 |
90
+ | 0.0125 | 3.61 | 4800 | 0.2851 | 0.9513 |
91
+ | 0.0157 | 3.76 | 5000 | 0.2883 | 0.9523 |
92
+ | 0.0414 | 3.91 | 5200 | 0.3163 | 0.9497 |
93
+ | 0.0025 | 4.06 | 5400 | 0.2998 | 0.9494 |
94
+ | 0.0019 | 4.21 | 5600 | 0.2925 | 0.9513 |
95
+ | 0.0013 | 4.36 | 5800 | 0.2872 | 0.9526 |
96
+ | 0.0014 | 4.51 | 6000 | 0.2906 | 0.9532 |
97
+ | 0.0015 | 4.67 | 6200 | 0.2862 | 0.9529 |
98
+ | 0.0281 | 4.82 | 6400 | 0.2863 | 0.9535 |
99
+ | 0.0287 | 4.97 | 6600 | 0.2898 | 0.9539 |
100
+
101
+
102
+ ### Framework versions
103
+
104
+ - Transformers 4.26.1
105
+ - Pytorch 1.13.1+cu116
106
+ - Datasets 2.10.1
107
+ - Tokenizers 0.13.2