Update README.md
Browse files
README.md
CHANGED
@@ -1,29 +1,53 @@
|
|
1 |
---
|
|
|
|
|
|
|
2 |
tags:
|
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 |
## Training procedure
|
29 |
|
@@ -31,11 +55,11 @@ More information needed
|
|
31 |
|
32 |
The following hyperparameters were used during training:
|
33 |
- learning_rate: 0.001
|
34 |
-
- train_batch_size:
|
35 |
- eval_batch_size: 64
|
36 |
- seed: 42
|
37 |
- gradient_accumulation_steps: 2
|
38 |
-
- total_train_batch_size:
|
39 |
- optimizer: Adafactor
|
40 |
- lr_scheduler_type: linear
|
41 |
- num_epochs: 10.0
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: apache-2.0
|
5 |
tags:
|
6 |
+
- t5-small
|
7 |
+
- text2text-generation
|
8 |
+
- dialog state tracking
|
9 |
+
- conversational system
|
10 |
+
- task-oriented dialog
|
11 |
+
datasets:
|
12 |
+
- ConvLab/sgd
|
13 |
+
metrics:
|
14 |
+
- Joint Goal Accuracy
|
15 |
+
- Slot F1
|
16 |
|
17 |
+
model-index:
|
18 |
+
- name: t5-small-dst-sgd
|
19 |
+
results:
|
20 |
+
- task:
|
21 |
+
type: text2text-generation
|
22 |
+
name: dialog state tracking
|
23 |
+
dataset:
|
24 |
+
type: ConvLab/sgd
|
25 |
+
name: SGD
|
26 |
+
split: test
|
27 |
+
revision: 6e8c79b888b21cc658cf9c0ce128d263241cf70f
|
28 |
+
metrics:
|
29 |
+
- type: Joint Goal Accuracy
|
30 |
+
value: 52.6
|
31 |
+
name: JGA
|
32 |
+
- type: Slot F1
|
33 |
+
value: 91.9
|
34 |
+
name: Slot F1
|
35 |
|
36 |
+
widget:
|
37 |
+
- text: "user: Hi, could you get me a restaurant booking on the 8th please?\nsystem: Any preference on the restaurant, location and time?\nuser: Could you get me a reservation at P.f. Chang's in Corte Madera at afternoon 12?"
|
38 |
+
- text: "user: I need to book a dinner reservation for a date. Help me reserve a table at a restaurant.\nsystem: What time and location do you have in mind?\nuser: Something around 8 in the night should be fine. Oh, and look in the San Jose area."
|
39 |
|
40 |
+
inference:
|
41 |
+
parameters:
|
42 |
+
max_length: 100
|
43 |
|
44 |
+
---
|
45 |
|
46 |
+
# t5-small-dst-sgd
|
47 |
|
48 |
+
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on [Schema-Guided Dialog](https://huggingface.co/datasets/ConvLab/sgd).
|
49 |
|
50 |
+
Refer to [ConvLab-3](https://github.com/ConvLab/ConvLab-3) for model description and usage.
|
51 |
|
52 |
## Training procedure
|
53 |
|
|
|
55 |
|
56 |
The following hyperparameters were used during training:
|
57 |
- learning_rate: 0.001
|
58 |
+
- train_batch_size: 128
|
59 |
- eval_batch_size: 64
|
60 |
- seed: 42
|
61 |
- gradient_accumulation_steps: 2
|
62 |
+
- total_train_batch_size: 256
|
63 |
- optimizer: Adafactor
|
64 |
- lr_scheduler_type: linear
|
65 |
- num_epochs: 10.0
|