File size: 1,898 Bytes
43f5918
525085d
 
 
43f5918
525085d
 
 
 
 
 
 
 
 
 
8bdc4ef
 
 
 
43f5918
8bdc4ef
 
 
 
 
 
 
 
 
525085d
 
 
 
 
 
 
 
8bdc4ef
525085d
 
 
8bdc4ef
 
 
 
 
 
525085d
43f5918
525085d
43f5918
525085d
43f5918
525085d
43f5918
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
525085d
 
 
 
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
---
language:
- en
license: apache-2.0
tags:
- t5-small
- text2text-generation
- dialog state tracking
- conversational system
- task-oriented dialog
datasets:
- ConvLab/multiwoz21
metrics:
- Joint Goal Accuracy
- Slot F1
widget:
- text: 'user: I would like a taxi from Saint John''s college to Pizza Hut Fen Ditton.

    system: What time do you want to leave and what time do you want to arrive by?

    user: I want to leave after 17:15.'
- text: "user: I want to find a moderately priced restaurant. \nsystem: I have many\
    \ options available for you! Is there a certain area or cuisine that interests\
    \ you?\nuser: Yes I would like the restaurant to be located in the center of the\
    \ town."
inference:
  parameters:
    max_length: 100
base_model: t5-small
model-index:
- name: t5-small-dst-multiwoz21
  results:
  - task:
      type: text2text-generation
      name: dialog state tracking
    dataset:
      name: MultiWOZ 2.1
      type: ConvLab/multiwoz21
      split: test
      revision: 5f55375edbfe0270c20bcf770751ad982c0e6614
    metrics:
    - type: Joint Goal Accuracy
      value: 52.6
      name: JGA
    - type: Slot F1
      value: 91.9
      name: Slot F1
---

# t5-small-dst-multiwoz21

This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on [MultiWOZ 2.1](https://huggingface.co/datasets/ConvLab/multiwoz21).

Refer to [ConvLab-3](https://github.com/ConvLab/ConvLab-3) for model description and usage.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adafactor
- lr_scheduler_type: linear
- num_epochs: 10.0

### Framework versions

- Transformers 4.20.1
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1