File size: 4,737 Bytes
d60d587
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
---
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: dit-tiny_tobacco3482_kd_CEKD_t1.5_a0.5
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# dit-tiny_tobacco3482_kd_CEKD_t1.5_a0.5

This model is a fine-tuned version of [microsoft/dit-base](https://huggingface.co/microsoft/dit-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.9246
- Accuracy: 0.18
- Brier Loss: 0.8755
- Nll: 6.7967
- F1 Micro: 0.18
- F1 Macro: 0.0306
- Ece: 0.2497
- Aurc: 0.8499

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 25

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Brier Loss | Nll     | F1 Micro | F1 Macro | Ece    | Aurc   |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:----------:|:-------:|:--------:|:--------:|:------:|:------:|
| No log        | 0.96  | 3    | 3.1239          | 0.145    | 0.8999     | 10.1580 | 0.145    | 0.0253   | 0.2222 | 0.8467 |
| No log        | 1.96  | 6    | 3.0895          | 0.145    | 0.8946     | 10.5934 | 0.145    | 0.0253   | 0.2303 | 0.8470 |
| No log        | 2.96  | 9    | 3.0385          | 0.165    | 0.8866     | 8.6307  | 0.165    | 0.0502   | 0.2200 | 0.8458 |
| No log        | 3.96  | 12   | 2.9972          | 0.21     | 0.8806     | 6.5449  | 0.2100   | 0.0615   | 0.2512 | 0.8364 |
| No log        | 4.96  | 15   | 2.9719          | 0.155    | 0.8776     | 6.7565  | 0.155    | 0.0271   | 0.2414 | 0.8884 |
| No log        | 5.96  | 18   | 2.9579          | 0.215    | 0.8768     | 7.0870  | 0.2150   | 0.0643   | 0.2713 | 0.8778 |
| No log        | 6.96  | 21   | 2.9485          | 0.18     | 0.8768     | 7.0291  | 0.18     | 0.0308   | 0.2482 | 0.8532 |
| No log        | 7.96  | 24   | 2.9417          | 0.18     | 0.8770     | 6.9706  | 0.18     | 0.0306   | 0.2559 | 0.8525 |
| No log        | 8.96  | 27   | 2.9360          | 0.18     | 0.8768     | 6.9349  | 0.18     | 0.0306   | 0.2498 | 0.8527 |
| No log        | 9.96  | 30   | 2.9326          | 0.18     | 0.8767     | 6.9268  | 0.18     | 0.0306   | 0.2635 | 0.8533 |
| No log        | 10.96 | 33   | 2.9303          | 0.18     | 0.8765     | 6.9226  | 0.18     | 0.0306   | 0.2637 | 0.8531 |
| No log        | 11.96 | 36   | 2.9289          | 0.18     | 0.8764     | 6.9217  | 0.18     | 0.0306   | 0.2591 | 0.8524 |
| No log        | 12.96 | 39   | 2.9279          | 0.18     | 0.8762     | 6.8547  | 0.18     | 0.0306   | 0.2505 | 0.8526 |
| No log        | 13.96 | 42   | 2.9270          | 0.18     | 0.8760     | 6.8491  | 0.18     | 0.0306   | 0.2500 | 0.8520 |
| No log        | 14.96 | 45   | 2.9263          | 0.18     | 0.8759     | 6.8471  | 0.18     | 0.0306   | 0.2463 | 0.8518 |
| No log        | 15.96 | 48   | 2.9258          | 0.18     | 0.8758     | 6.8445  | 0.18     | 0.0306   | 0.2462 | 0.8520 |
| No log        | 16.96 | 51   | 2.9255          | 0.18     | 0.8758     | 6.8452  | 0.18     | 0.0306   | 0.2587 | 0.8511 |
| No log        | 17.96 | 54   | 2.9256          | 0.18     | 0.8758     | 6.7940  | 0.18     | 0.0306   | 0.2585 | 0.8513 |
| No log        | 18.96 | 57   | 2.9256          | 0.18     | 0.8758     | 6.7930  | 0.18     | 0.0306   | 0.2625 | 0.8508 |
| No log        | 19.96 | 60   | 2.9252          | 0.18     | 0.8757     | 6.7945  | 0.18     | 0.0306   | 0.2580 | 0.8506 |
| No log        | 20.96 | 63   | 2.9250          | 0.18     | 0.8756     | 6.7999  | 0.18     | 0.0306   | 0.2539 | 0.8505 |
| No log        | 21.96 | 66   | 2.9248          | 0.18     | 0.8756     | 6.8441  | 0.18     | 0.0306   | 0.2538 | 0.8502 |
| No log        | 22.96 | 69   | 2.9247          | 0.18     | 0.8755     | 6.8439  | 0.18     | 0.0306   | 0.2497 | 0.8500 |
| No log        | 23.96 | 72   | 2.9247          | 0.18     | 0.8755     | 6.7977  | 0.18     | 0.0306   | 0.2497 | 0.8500 |
| No log        | 24.96 | 75   | 2.9246          | 0.18     | 0.8755     | 6.7967  | 0.18     | 0.0306   | 0.2497 | 0.8499 |


### Framework versions

- Transformers 4.26.1
- Pytorch 1.13.1.post200
- Datasets 2.9.0
- Tokenizers 0.13.2