update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- accuracy
|
7 |
+
model-index:
|
8 |
+
- name: distilbert-base-uncased__subj__train-8-4
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# distilbert-base-uncased__subj__train-8-4
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.3305
|
20 |
+
- Accuracy: 0.8565
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 2e-05
|
40 |
+
- train_batch_size: 4
|
41 |
+
- eval_batch_size: 4
|
42 |
+
- seed: 42
|
43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
+
- lr_scheduler_type: linear
|
45 |
+
- num_epochs: 50
|
46 |
+
- mixed_precision_training: Native AMP
|
47 |
+
|
48 |
+
### Training results
|
49 |
+
|
50 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
51 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
52 |
+
| 0.6991 | 1.0 | 3 | 0.6772 | 0.75 |
|
53 |
+
| 0.6707 | 2.0 | 6 | 0.6704 | 0.75 |
|
54 |
+
| 0.6402 | 3.0 | 9 | 0.6608 | 1.0 |
|
55 |
+
| 0.5789 | 4.0 | 12 | 0.6547 | 0.75 |
|
56 |
+
| 0.5211 | 5.0 | 15 | 0.6434 | 0.75 |
|
57 |
+
| 0.454 | 6.0 | 18 | 0.6102 | 1.0 |
|
58 |
+
| 0.4187 | 7.0 | 21 | 0.5701 | 1.0 |
|
59 |
+
| 0.3401 | 8.0 | 24 | 0.5289 | 1.0 |
|
60 |
+
| 0.3107 | 9.0 | 27 | 0.4737 | 1.0 |
|
61 |
+
| 0.2381 | 10.0 | 30 | 0.4255 | 1.0 |
|
62 |
+
| 0.1982 | 11.0 | 33 | 0.3685 | 1.0 |
|
63 |
+
| 0.1631 | 12.0 | 36 | 0.3200 | 1.0 |
|
64 |
+
| 0.1234 | 13.0 | 39 | 0.2798 | 1.0 |
|
65 |
+
| 0.0993 | 14.0 | 42 | 0.2455 | 1.0 |
|
66 |
+
| 0.0781 | 15.0 | 45 | 0.2135 | 1.0 |
|
67 |
+
| 0.0586 | 16.0 | 48 | 0.1891 | 1.0 |
|
68 |
+
| 0.0513 | 17.0 | 51 | 0.1671 | 1.0 |
|
69 |
+
| 0.043 | 18.0 | 54 | 0.1427 | 1.0 |
|
70 |
+
| 0.0307 | 19.0 | 57 | 0.1225 | 1.0 |
|
71 |
+
| 0.0273 | 20.0 | 60 | 0.1060 | 1.0 |
|
72 |
+
| 0.0266 | 21.0 | 63 | 0.0920 | 1.0 |
|
73 |
+
| 0.0233 | 22.0 | 66 | 0.0823 | 1.0 |
|
74 |
+
| 0.0185 | 23.0 | 69 | 0.0751 | 1.0 |
|
75 |
+
| 0.0173 | 24.0 | 72 | 0.0698 | 1.0 |
|
76 |
+
| 0.0172 | 25.0 | 75 | 0.0651 | 1.0 |
|
77 |
+
| 0.0142 | 26.0 | 78 | 0.0613 | 1.0 |
|
78 |
+
| 0.0151 | 27.0 | 81 | 0.0583 | 1.0 |
|
79 |
+
| 0.0117 | 28.0 | 84 | 0.0563 | 1.0 |
|
80 |
+
| 0.0123 | 29.0 | 87 | 0.0546 | 1.0 |
|
81 |
+
| 0.0121 | 30.0 | 90 | 0.0531 | 1.0 |
|
82 |
+
| 0.0123 | 31.0 | 93 | 0.0511 | 1.0 |
|
83 |
+
| 0.0112 | 32.0 | 96 | 0.0496 | 1.0 |
|
84 |
+
| 0.0103 | 33.0 | 99 | 0.0481 | 1.0 |
|
85 |
+
| 0.0086 | 34.0 | 102 | 0.0468 | 1.0 |
|
86 |
+
| 0.0096 | 35.0 | 105 | 0.0457 | 1.0 |
|
87 |
+
| 0.0107 | 36.0 | 108 | 0.0447 | 1.0 |
|
88 |
+
| 0.0095 | 37.0 | 111 | 0.0439 | 1.0 |
|
89 |
+
| 0.0102 | 38.0 | 114 | 0.0429 | 1.0 |
|
90 |
+
| 0.0077 | 39.0 | 117 | 0.0422 | 1.0 |
|
91 |
+
| 0.0092 | 40.0 | 120 | 0.0415 | 1.0 |
|
92 |
+
| 0.0083 | 41.0 | 123 | 0.0409 | 1.0 |
|
93 |
+
| 0.0094 | 42.0 | 126 | 0.0404 | 1.0 |
|
94 |
+
| 0.0084 | 43.0 | 129 | 0.0400 | 1.0 |
|
95 |
+
| 0.0085 | 44.0 | 132 | 0.0396 | 1.0 |
|
96 |
+
| 0.0092 | 45.0 | 135 | 0.0392 | 1.0 |
|
97 |
+
| 0.0076 | 46.0 | 138 | 0.0389 | 1.0 |
|
98 |
+
| 0.0073 | 47.0 | 141 | 0.0388 | 1.0 |
|
99 |
+
| 0.0085 | 48.0 | 144 | 0.0387 | 1.0 |
|
100 |
+
| 0.0071 | 49.0 | 147 | 0.0386 | 1.0 |
|
101 |
+
| 0.0079 | 50.0 | 150 | 0.0386 | 1.0 |
|
102 |
+
|
103 |
+
|
104 |
+
### Framework versions
|
105 |
+
|
106 |
+
- Transformers 4.15.0
|
107 |
+
- Pytorch 1.10.2+cu102
|
108 |
+
- Datasets 1.18.2
|
109 |
+
- Tokenizers 0.10.3
|