Hugging Face's logo

k4black
/
textattack-roberta-base-MNLI-e-snli-classification-nli-base

Text Classification
Transformers
PyTorch
roberta
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
2
k4black commited on
Commit
5040fba
·
1 Parent(s): aaf3c61

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +103 -0
README.md ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - generated_from_trainer
4
+ datasets:
5
+ - esnli
6
+ metrics:
7
+ - f1
8
+ - accuracy
9
+ model-index:
10
+ - name: textattack-roberta-base-MNLI-e-snli-classification-nli-base
11
+ results:
12
+ - task:
13
+ name: Text Classification
14
+ type: text-classification
15
+ dataset:
16
+ name: esnli
17
+ type: esnli
18
+ config: plain_text
19
+ split: validation
20
+ args: plain_text
21
+ metrics:
22
+ - name: F1
23
+ type: f1
24
+ value: 0.9106202958294739
25
+ - name: Accuracy
26
+ type: accuracy
27
+ value: 0.9110953058321479
28
+ ---
29
+
30
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
31
+ should probably proofread and complete it, then remove this comment. -->
32
+
33
+ # textattack-roberta-base-MNLI-e-snli-classification-nli-base
34
+
35
+ This model is a fine-tuned version of [textattack/roberta-base-MNLI](https://huggingface.co/textattack/roberta-base-MNLI) on the esnli dataset.
36
+ It achieves the following results on the evaluation set:
37
+ - Loss: 0.2488
38
+ - F1: 0.9106
39
+ - Accuracy: 0.9111
40
+
41
+ ## Model description
42
+
43
+ More information needed
44
+
45
+ ## Intended uses & limitations
46
+
47
+ More information needed
48
+
49
+ ## Training and evaluation data
50
+
51
+ More information needed
52
+
53
+ ## Training procedure
54
+
55
+ ### Training hyperparameters
56
+
57
+ The following hyperparameters were used during training:
58
+ - learning_rate: 1e-05
59
+ - train_batch_size: 64
60
+ - eval_batch_size: 64
61
+ - seed: 42
62
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
+ - lr_scheduler_type: linear
64
+ - lr_scheduler_warmup_ratio: 0.05
65
+ - num_epochs: 3
66
+ - mixed_precision_training: Native AMP
67
+
68
+ ### Training results
69
+
70
+ | Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
71
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|
72
+ | 1.5376 | 0.05 | 400 | 0.4010 | 0.8556 | 0.8556 |
73
+ | 0.4352 | 0.09 | 800 | 0.3349 | 0.8795 | 0.8800 |
74
+ | 0.4 | 0.14 | 1200 | 0.3180 | 0.8851 | 0.8854 |
75
+ | 0.3801 | 0.19 | 1600 | 0.2975 | 0.8918 | 0.8921 |
76
+ | 0.3599 | 0.23 | 2000 | 0.2949 | 0.8951 | 0.8955 |
77
+ | 0.3612 | 0.28 | 2400 | 0.2802 | 0.8987 | 0.8987 |
78
+ | 0.3519 | 0.33 | 2800 | 0.2763 | 0.8977 | 0.8980 |
79
+ | 0.349 | 0.37 | 3200 | 0.2766 | 0.9020 | 0.9023 |
80
+ | 0.3432 | 0.42 | 3600 | 0.2748 | 0.9000 | 0.9001 |
81
+ | 0.3435 | 0.47 | 4000 | 0.2702 | 0.9051 | 0.9051 |
82
+ | 0.3352 | 0.51 | 4400 | 0.2728 | 0.9034 | 0.9039 |
83
+ | 0.3277 | 0.56 | 4800 | 0.2634 | 0.9039 | 0.9043 |
84
+ | 0.3307 | 0.61 | 5200 | 0.2623 | 0.9050 | 0.9057 |
85
+ | 0.3247 | 0.65 | 5600 | 0.2685 | 0.9059 | 0.9063 |
86
+ | 0.3175 | 0.7 | 6000 | 0.2589 | 0.9081 | 0.9084 |
87
+ | 0.3144 | 0.75 | 6400 | 0.2586 | 0.9088 | 0.9093 |
88
+ | 0.3102 | 0.79 | 6800 | 0.2547 | 0.9088 | 0.9090 |
89
+ | 0.3223 | 0.84 | 7200 | 0.2526 | 0.9093 | 0.9096 |
90
+ | 0.3166 | 0.89 | 7600 | 0.2490 | 0.9115 | 0.9118 |
91
+ | 0.3124 | 0.93 | 8000 | 0.2503 | 0.9106 | 0.9107 |
92
+ | 0.3053 | 0.98 | 8400 | 0.2452 | 0.9099 | 0.9101 |
93
+ | 0.2908 | 1.03 | 8800 | 0.2575 | 0.9113 | 0.9119 |
94
+ | 0.2853 | 1.07 | 9200 | 0.2464 | 0.9114 | 0.9118 |
95
+ | 0.2796 | 1.12 | 9600 | 0.2488 | 0.9106 | 0.9111 |
96
+
97
+
98
+ ### Framework versions
99
+
100
+ - Transformers 4.27.1
101
+ - Pytorch 1.12.1+cu113
102
+ - Datasets 2.10.1
103
+ - Tokenizers 0.13.2