Steven Liu
commited on
Commit
•
a68e896
1
Parent(s):
c61955e
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- summarization
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- billsum
|
8 |
+
metrics:
|
9 |
+
- rouge
|
10 |
+
model-index:
|
11 |
+
- name: t5-small-finetuned-billsum-ca_test
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Sequence-to-sequence Language Modeling
|
15 |
+
type: text2text-generation
|
16 |
+
dataset:
|
17 |
+
name: billsum
|
18 |
+
type: billsum
|
19 |
+
args: default
|
20 |
+
metrics:
|
21 |
+
- name: Rouge1
|
22 |
+
type: rouge
|
23 |
+
value: 52.2582
|
24 |
+
---
|
25 |
+
|
26 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
27 |
+
should probably proofread and complete it, then remove this comment. -->
|
28 |
+
|
29 |
+
# t5-small-finetuned-billsum-ca_test
|
30 |
+
|
31 |
+
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the billsum dataset.
|
32 |
+
It achieves the following results on the evaluation set:
|
33 |
+
- Loss: 1.5234
|
34 |
+
- Rouge1: 52.2582
|
35 |
+
- Rouge2: 34.8162
|
36 |
+
- Rougel: 50.5491
|
37 |
+
- Rougelsum: 50.6121
|
38 |
+
- Gen Len: 18.996
|
39 |
+
|
40 |
+
## Model description
|
41 |
+
|
42 |
+
More information needed
|
43 |
+
|
44 |
+
## Intended uses & limitations
|
45 |
+
|
46 |
+
More information needed
|
47 |
+
|
48 |
+
## Training and evaluation data
|
49 |
+
|
50 |
+
More information needed
|
51 |
+
|
52 |
+
## Training procedure
|
53 |
+
|
54 |
+
### Training hyperparameters
|
55 |
+
|
56 |
+
The following hyperparameters were used during training:
|
57 |
+
- learning_rate: 2e-05
|
58 |
+
- train_batch_size: 2
|
59 |
+
- eval_batch_size: 2
|
60 |
+
- seed: 42
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- num_epochs: 4
|
64 |
+
- mixed_precision_training: Native AMP
|
65 |
+
|
66 |
+
### Training results
|
67 |
+
|
68 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
69 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
70 |
+
| No log | 1.0 | 495 | 1.8113 | 58.4024 | 41.7432 | 56.9521 | 57.0516 | 18.9597 |
|
71 |
+
| 2.709 | 2.0 | 990 | 1.6230 | 47.7769 | 32.1777 | 46.0344 | 46.046 | 18.996 |
|
72 |
+
| 1.9323 | 3.0 | 1485 | 1.5459 | 51.2371 | 33.8242 | 49.4532 | 49.5038 | 18.996 |
|
73 |
+
| 1.7842 | 4.0 | 1980 | 1.5234 | 52.2582 | 34.8162 | 50.5491 | 50.6121 | 18.996 |
|
74 |
+
|
75 |
+
|
76 |
+
### Framework versions
|
77 |
+
|
78 |
+
- Transformers 4.12.2
|
79 |
+
- Pytorch 1.9.0+cu111
|
80 |
+
- Datasets 1.14.0
|
81 |
+
- Tokenizers 0.10.3
|