Commit
•
a81a557
1
Parent(s):
0bbaf9d
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- math_qa
|
7 |
+
metrics:
|
8 |
+
- rouge
|
9 |
+
model-index:
|
10 |
+
- name: t5-small-mathT5-finetune_qatoexp
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Sequence-to-sequence Language Modeling
|
14 |
+
type: text2text-generation
|
15 |
+
dataset:
|
16 |
+
name: math_qa
|
17 |
+
type: math_qa
|
18 |
+
config: default
|
19 |
+
split: train
|
20 |
+
args: default
|
21 |
+
metrics:
|
22 |
+
- name: Rouge1
|
23 |
+
type: rouge
|
24 |
+
value: 21.9174
|
25 |
+
---
|
26 |
+
|
27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
28 |
+
should probably proofread and complete it, then remove this comment. -->
|
29 |
+
|
30 |
+
# t5-small-mathT5-finetune_qatoexp
|
31 |
+
|
32 |
+
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the math_qa dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 1.8677
|
35 |
+
- Rouge1: 21.9174
|
36 |
+
- Rouge2: 8.4401
|
37 |
+
- Rougel: 19.1645
|
38 |
+
- Rougelsum: 19.8239
|
39 |
+
- Gen Len: 18.9765
|
40 |
+
|
41 |
+
## Model description
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Intended uses & limitations
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training and evaluation data
|
50 |
+
|
51 |
+
More information needed
|
52 |
+
|
53 |
+
## Training procedure
|
54 |
+
|
55 |
+
### Training hyperparameters
|
56 |
+
|
57 |
+
The following hyperparameters were used during training:
|
58 |
+
- learning_rate: 2e-05
|
59 |
+
- train_batch_size: 10
|
60 |
+
- eval_batch_size: 10
|
61 |
+
- seed: 42
|
62 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
63 |
+
- lr_scheduler_type: linear
|
64 |
+
- num_epochs: 10
|
65 |
+
- mixed_precision_training: Native AMP
|
66 |
+
|
67 |
+
### Training results
|
68 |
+
|
69 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
70 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
|
71 |
+
| 2.4496 | 1.0 | 2984 | 2.2096 | 19.6477 | 6.508 | 16.9295 | 17.5212 | 18.9064 |
|
72 |
+
| 2.2893 | 2.0 | 5968 | 2.0837 | 20.4879 | 7.2528 | 17.7778 | 18.4085 | 18.968 |
|
73 |
+
| 2.1869 | 3.0 | 8952 | 2.0125 | 20.8462 | 7.6105 | 18.1516 | 18.8343 | 18.9837 |
|
74 |
+
| 2.1456 | 4.0 | 11936 | 1.9633 | 20.7623 | 7.7113 | 18.1274 | 18.783 | 18.9886 |
|
75 |
+
| 2.1171 | 5.0 | 14920 | 1.9321 | 21.0648 | 7.8897 | 18.4162 | 19.0551 | 18.9844 |
|
76 |
+
| 2.0854 | 6.0 | 17904 | 1.9061 | 21.4445 | 8.0883 | 18.8038 | 19.4176 | 18.9812 |
|
77 |
+
| 2.0592 | 7.0 | 20888 | 1.8902 | 21.5714 | 8.2751 | 18.8864 | 19.537 | 18.9772 |
|
78 |
+
| 2.0609 | 8.0 | 23872 | 1.8770 | 21.7737 | 8.3297 | 19.022 | 19.6897 | 18.9763 |
|
79 |
+
| 2.0285 | 9.0 | 26856 | 1.8701 | 21.964 | 8.4358 | 19.1701 | 19.845 | 18.9747 |
|
80 |
+
| 2.0165 | 10.0 | 29840 | 1.8677 | 21.9174 | 8.4401 | 19.1645 | 19.8239 | 18.9765 |
|
81 |
+
|
82 |
+
|
83 |
+
### Framework versions
|
84 |
+
|
85 |
+
- Transformers 4.22.2
|
86 |
+
- Pytorch 1.12.1+cu113
|
87 |
+
- Datasets 2.5.2
|
88 |
+
- Tokenizers 0.12.1
|