File size: 4,927 Bytes
f3ed916
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bde85da
 
 
f3ed916
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bde85da
 
f3ed916
 
 
bde85da
f3ed916
 
 
 
 
bde85da
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f3ed916
 
 
 
 
 
 
 
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
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
---
license: mit
base_model: microsoft/deberta-v3-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: 1_microsoft_deberta_V1.0
  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. -->

# 1_microsoft_deberta_V1.0

This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co./microsoft/deberta-v3-large) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5748
- Map@3: 0.8700
- Accuracy: 0.785

## 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: 2
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 21
- total_train_batch_size: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1

### Training results

| Training Loss | Epoch | Step | Validation Loss | Map@3  | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|
| 1.6136        | 0.02  | 25   | 1.6092          | 0.4850 | 0.335    |
| 1.6116        | 0.04  | 50   | 1.6063          | 0.7250 | 0.61     |
| 1.4598        | 0.05  | 75   | 1.2186          | 0.7575 | 0.63     |
| 1.0137        | 0.07  | 100  | 0.9068          | 0.7908 | 0.665    |
| 0.9483        | 0.09  | 125  | 0.9574          | 0.8108 | 0.69     |
| 0.9619        | 0.1   | 150  | 0.8634          | 0.8183 | 0.71     |
| 0.8679        | 0.12  | 175  | 0.7644          | 0.8292 | 0.73     |
| 0.8594        | 0.14  | 200  | 0.8161          | 0.8067 | 0.7      |
| 0.8105        | 0.16  | 225  | 0.8355          | 0.82   | 0.715    |
| 0.8315        | 0.17  | 250  | 0.7381          | 0.8275 | 0.73     |
| 0.8275        | 0.19  | 275  | 0.7636          | 0.8433 | 0.745    |
| 0.8252        | 0.21  | 300  | 0.7196          | 0.8217 | 0.73     |
| 0.7801        | 0.23  | 325  | 0.6940          | 0.8367 | 0.745    |
| 0.8078        | 0.24  | 350  | 0.7185          | 0.8567 | 0.775    |
| 0.7583        | 0.26  | 375  | 0.7007          | 0.8433 | 0.75     |
| 0.7772        | 0.28  | 400  | 0.7032          | 0.8417 | 0.75     |
| 0.8204        | 0.3   | 425  | 0.7062          | 0.8500 | 0.76     |
| 0.8269        | 0.32  | 450  | 0.7082          | 0.8617 | 0.785    |
| 0.7418        | 0.33  | 475  | 0.7288          | 0.8517 | 0.78     |
| 0.7376        | 0.35  | 500  | 0.7021          | 0.8633 | 0.78     |
| 0.7519        | 0.37  | 525  | 0.6943          | 0.8642 | 0.785    |
| 0.7469        | 0.39  | 550  | 0.6807          | 0.8725 | 0.805    |
| 0.7244        | 0.4   | 575  | 0.6622          | 0.8692 | 0.79     |
| 0.7297        | 0.42  | 600  | 0.6783          | 0.8583 | 0.775    |
| 0.7259        | 0.44  | 625  | 0.6788          | 0.8550 | 0.765    |
| 0.6893        | 0.46  | 650  | 0.6571          | 0.8625 | 0.785    |
| 0.6871        | 0.47  | 675  | 0.6587          | 0.8492 | 0.76     |
| 0.7003        | 0.49  | 700  | 0.6485          | 0.8683 | 0.785    |
| 0.7094        | 0.51  | 725  | 0.6320          | 0.8675 | 0.795    |
| 0.7052        | 0.53  | 750  | 0.6554          | 0.8583 | 0.78     |
| 0.6873        | 0.54  | 775  | 0.6121          | 0.8550 | 0.775    |
| 0.6152        | 0.56  | 800  | 0.6060          | 0.8675 | 0.785    |
| 0.6741        | 0.58  | 825  | 0.6191          | 0.8808 | 0.815    |
| 0.7098        | 0.59  | 850  | 0.6213          | 0.8817 | 0.815    |
| 0.7029        | 0.61  | 875  | 0.6533          | 0.8725 | 0.79     |
| 0.6489        | 0.63  | 900  | 0.6127          | 0.8667 | 0.79     |
| 0.7289        | 0.65  | 925  | 0.6261          | 0.8750 | 0.81     |
| 0.6589        | 0.67  | 950  | 0.6019          | 0.8708 | 0.8      |
| 0.6876        | 0.68  | 975  | 0.6076          | 0.8725 | 0.805    |
| 0.6624        | 0.7   | 1000 | 0.5810          | 0.8708 | 0.79     |
| 0.6746        | 0.72  | 1025 | 0.5922          | 0.8708 | 0.79     |
| 0.6644        | 0.73  | 1050 | 0.5827          | 0.8675 | 0.785    |
| 0.668         | 0.75  | 1075 | 0.5814          | 0.8725 | 0.795    |
| 0.6115        | 0.77  | 1100 | 0.5680          | 0.8750 | 0.8      |
| 0.6799        | 0.79  | 1125 | 0.5767          | 0.8750 | 0.8      |
| 0.6466        | 0.81  | 1150 | 0.5700          | 0.8725 | 0.795    |
| 0.6765        | 0.82  | 1175 | 0.5700          | 0.8717 | 0.79     |
| 0.6936        | 0.84  | 1200 | 0.5758          | 0.8683 | 0.785    |
| 0.6239        | 0.86  | 1225 | 0.5748          | 0.8700 | 0.785    |


### Framework versions

- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.9.0
- Tokenizers 0.13.3