afbudiman commited on
Commit
f1e71ca
1 Parent(s): 907f749

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +86 -0
README.md ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - indonlu
7
+ metrics:
8
+ - accuracy
9
+ - f1
10
+ model-index:
11
+ - name: indobert-distilled-optimized-for-classification
12
+ results:
13
+ - task:
14
+ name: Text Classification
15
+ type: text-classification
16
+ dataset:
17
+ name: indonlu
18
+ type: indonlu
19
+ args: smsa
20
+ metrics:
21
+ - name: Accuracy
22
+ type: accuracy
23
+ value: 0.9023809523809524
24
+ - name: F1
25
+ type: f1
26
+ value: 0.9020516403647337
27
+ ---
28
+
29
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
30
+ should probably proofread and complete it, then remove this comment. -->
31
+
32
+ # indobert-distilled-optimized-for-classification
33
+
34
+ This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the indonlu dataset.
35
+ It achieves the following results on the evaluation set:
36
+ - Loss: 0.5991
37
+ - Accuracy: 0.9024
38
+ - F1: 0.9021
39
+
40
+ ## Model description
41
+
42
+ More information needed
43
+
44
+ ## Intended uses & limitations
45
+
46
+ More information needed
47
+
48
+ ## Training and evaluation data
49
+
50
+ More information needed
51
+
52
+ ## Training procedure
53
+
54
+ ### Training hyperparameters
55
+
56
+ The following hyperparameters were used during training:
57
+ - learning_rate: 5.262995179171344e-05
58
+ - train_batch_size: 16
59
+ - eval_batch_size: 16
60
+ - seed: 33
61
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
+ - lr_scheduler_type: linear
63
+ - num_epochs: 10
64
+
65
+ ### Training results
66
+
67
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
68
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
69
+ | 1.2938 | 1.0 | 688 | 0.8433 | 0.8484 | 0.8513 |
70
+ | 0.711 | 2.0 | 1376 | 0.6408 | 0.8881 | 0.8878 |
71
+ | 0.4416 | 3.0 | 2064 | 0.7964 | 0.8794 | 0.8793 |
72
+ | 0.2907 | 4.0 | 2752 | 0.7559 | 0.8897 | 0.8900 |
73
+ | 0.2065 | 5.0 | 3440 | 0.6892 | 0.8968 | 0.8974 |
74
+ | 0.1574 | 6.0 | 4128 | 0.6881 | 0.8913 | 0.8906 |
75
+ | 0.1131 | 7.0 | 4816 | 0.6224 | 0.8984 | 0.8982 |
76
+ | 0.0865 | 8.0 | 5504 | 0.6312 | 0.8976 | 0.8970 |
77
+ | 0.0678 | 9.0 | 6192 | 0.6187 | 0.8992 | 0.8989 |
78
+ | 0.0526 | 10.0 | 6880 | 0.5991 | 0.9024 | 0.9021 |
79
+
80
+
81
+ ### Framework versions
82
+
83
+ - Transformers 4.18.0
84
+ - Pytorch 1.10.0+cu111
85
+ - Datasets 2.1.0
86
+ - Tokenizers 0.12.1