File size: 5,251 Bytes
b337288
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: apache-2.0
base_model: distilbert-base-cased
tags:
- generated_from_trainer
model-index:
- name: distilbert
  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. -->

# distilbert

This model is a fine-tuned version of [distilbert-base-cased](https://huggingface.co/distilbert-base-cased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0005
- `o` precision: 0.9946
- `o` recall: 0.9960
- `o` f1: 0.9953
- `i` precision: 0.9994
- `i` recall: 0.9993
- `i` f1: 0.9994
- Weighted avg f1: 0.9989

## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 426
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2

### Training results

| Training Loss | Epoch | Step  | Validation Loss | `o` precision | `o` recall | `o` f1 | `i` precision | `i` recall | `i` f1 | Weighted avg f1 |
|:-------------:|:-----:|:-----:|:---------------:|:-------------:|:----------:|:------:|:-------------:|:----------:|:------:|:---------------:|
| 0.0187        | 0.08  | 500   | 0.0033          | 0.9363        | 0.9990     | 0.9666 | 0.9999        | 0.9906     | 0.9952 | 0.9918          |
| 0.002         | 0.16  | 1000  | 0.0014          | 0.9762        | 0.9947     | 0.9854 | 0.9993        | 0.9967     | 0.9980 | 0.9964          |
| 0.0016        | 0.24  | 1500  | 0.0012          | 0.9813        | 0.9918     | 0.9865 | 0.9989        | 0.9974     | 0.9981 | 0.9967          |
| 0.0015        | 0.32  | 2000  | 0.0012          | 0.9801        | 0.9960     | 0.9880 | 0.9994        | 0.9972     | 0.9983 | 0.9971          |
| 0.0013        | 0.4   | 2500  | 0.0010          | 0.9834        | 0.9960     | 0.9896 | 0.9994        | 0.9977     | 0.9986 | 0.9975          |
| 0.001         | 0.48  | 3000  | 0.0008          | 0.9881        | 0.9959     | 0.9920 | 0.9994        | 0.9983     | 0.9989 | 0.9980          |
| 0.0009        | 0.56  | 3500  | 0.0009          | 0.9854        | 0.9955     | 0.9904 | 0.9994        | 0.9980     | 0.9987 | 0.9977          |
| 0.0009        | 0.64  | 4000  | 0.0008          | 0.9883        | 0.9946     | 0.9914 | 0.9993        | 0.9984     | 0.9988 | 0.9979          |
| 0.0009        | 0.72  | 4500  | 0.0009          | 0.9935        | 0.9884     | 0.9910 | 0.9984        | 0.9991     | 0.9988 | 0.9978          |
| 0.0008        | 0.8   | 5000  | 0.0008          | 0.9913        | 0.9926     | 0.9920 | 0.9990        | 0.9988     | 0.9989 | 0.9981          |
| 0.0008        | 0.88  | 5500  | 0.0007          | 0.9874        | 0.9976     | 0.9925 | 0.9997        | 0.9982     | 0.9990 | 0.9982          |
| 0.0008        | 0.96  | 6000  | 0.0008          | 0.9924        | 0.9923     | 0.9923 | 0.9989        | 0.9990     | 0.9989 | 0.9982          |
| 0.0005        | 1.04  | 6500  | 0.0007          | 0.9924        | 0.9948     | 0.9936 | 0.9993        | 0.9990     | 0.9991 | 0.9985          |
| 0.0005        | 1.12  | 7000  | 0.0007          | 0.9885        | 0.9973     | 0.9929 | 0.9996        | 0.9984     | 0.9990 | 0.9983          |
| 0.0005        | 1.2   | 7500  | 0.0007          | 0.9890        | 0.9970     | 0.9930 | 0.9996        | 0.9985     | 0.9990 | 0.9983          |
| 0.0006        | 1.28  | 8000  | 0.0006          | 0.9927        | 0.9965     | 0.9946 | 0.9995        | 0.9990     | 0.9993 | 0.9987          |
| 0.0004        | 1.36  | 8500  | 0.0005          | 0.9934        | 0.9962     | 0.9948 | 0.9995        | 0.9991     | 0.9993 | 0.9987          |
| 0.0004        | 1.44  | 9000  | 0.0006          | 0.9941        | 0.9953     | 0.9947 | 0.9994        | 0.9992     | 0.9993 | 0.9987          |
| 0.0004        | 1.52  | 9500  | 0.0005          | 0.9940        | 0.9951     | 0.9946 | 0.9993        | 0.9992     | 0.9993 | 0.9987          |
| 0.0004        | 1.6   | 10000 | 0.0005          | 0.9942        | 0.9958     | 0.9950 | 0.9994        | 0.9992     | 0.9993 | 0.9988          |
| 0.0003        | 1.68  | 10500 | 0.0006          | 0.9940        | 0.9951     | 0.9945 | 0.9993        | 0.9992     | 0.9992 | 0.9987          |
| 0.0005        | 1.76  | 11000 | 0.0005          | 0.9953        | 0.9947     | 0.9950 | 0.9993        | 0.9994     | 0.9993 | 0.9988          |
| 0.0004        | 1.84  | 11500 | 0.0005          | 0.9944        | 0.9958     | 0.9951 | 0.9994        | 0.9992     | 0.9993 | 0.9988          |
| 0.0004        | 1.92  | 12000 | 0.0005          | 0.9943        | 0.9962     | 0.9953 | 0.9995        | 0.9992     | 0.9993 | 0.9989          |
| 0.0004        | 2.0   | 12500 | 0.0005          | 0.9946        | 0.9960     | 0.9953 | 0.9994        | 0.9993     | 0.9994 | 0.9989          |


### Framework versions

- Transformers 4.34.1
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1