update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- wikihow
|
| 7 |
+
metrics:
|
| 8 |
+
- rouge
|
| 9 |
+
model-index:
|
| 10 |
+
- name: t5-small-finetuned-wikihow_3epoch_b8_lr3e-5
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Sequence-to-sequence Language Modeling
|
| 14 |
+
type: text2text-generation
|
| 15 |
+
dataset:
|
| 16 |
+
name: wikihow
|
| 17 |
+
type: wikihow
|
| 18 |
+
args: all
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Rouge1
|
| 21 |
+
type: rouge
|
| 22 |
+
value: 25.9411
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 26 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 27 |
+
|
| 28 |
+
# t5-small-finetuned-wikihow_3epoch_b8_lr3e-5
|
| 29 |
+
|
| 30 |
+
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the wikihow dataset.
|
| 31 |
+
It achieves the following results on the evaluation set:
|
| 32 |
+
- Loss: 2.4836
|
| 33 |
+
- Rouge1: 25.9411
|
| 34 |
+
- Rouge2: 9.226
|
| 35 |
+
- Rougel: 21.9087
|
| 36 |
+
- Rougelsum: 25.2863
|
| 37 |
+
- Gen Len: 18.4076
|
| 38 |
+
|
| 39 |
+
## Model description
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Intended uses & limitations
|
| 44 |
+
|
| 45 |
+
More information needed
|
| 46 |
+
|
| 47 |
+
## Training and evaluation data
|
| 48 |
+
|
| 49 |
+
More information needed
|
| 50 |
+
|
| 51 |
+
## Training procedure
|
| 52 |
+
|
| 53 |
+
### Training hyperparameters
|
| 54 |
+
|
| 55 |
+
The following hyperparameters were used during training:
|
| 56 |
+
- learning_rate: 3e-05
|
| 57 |
+
- train_batch_size: 8
|
| 58 |
+
- eval_batch_size: 8
|
| 59 |
+
- seed: 42
|
| 60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 61 |
+
- lr_scheduler_type: linear
|
| 62 |
+
- num_epochs: 3
|
| 63 |
+
- mixed_precision_training: Native AMP
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
| 68 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
|
| 69 |
+
| 2.912 | 0.25 | 5000 | 2.6285 | 23.6659 | 7.8535 | 19.9837 | 22.9884 | 18.3867 |
|
| 70 |
+
| 2.8115 | 0.51 | 10000 | 2.5820 | 24.7979 | 8.4888 | 20.8719 | 24.1321 | 18.3292 |
|
| 71 |
+
| 2.767 | 0.76 | 15000 | 2.5555 | 25.0857 | 8.6437 | 21.149 | 24.4256 | 18.2981 |
|
| 72 |
+
| 2.742 | 1.02 | 20000 | 2.5330 | 25.3431 | 8.8393 | 21.425 | 24.7032 | 18.3749 |
|
| 73 |
+
| 2.7092 | 1.27 | 25000 | 2.5203 | 25.5338 | 8.9281 | 21.5378 | 24.9045 | 18.3399 |
|
| 74 |
+
| 2.6989 | 1.53 | 30000 | 2.5065 | 25.4792 | 8.9745 | 21.4941 | 24.8458 | 18.4565 |
|
| 75 |
+
| 2.6894 | 1.78 | 35000 | 2.5018 | 25.6815 | 9.1218 | 21.6958 | 25.0557 | 18.406 |
|
| 76 |
+
| 2.6897 | 2.03 | 40000 | 2.4944 | 25.8241 | 9.2127 | 21.8205 | 25.1801 | 18.4228 |
|
| 77 |
+
| 2.6664 | 2.29 | 45000 | 2.4891 | 25.8241 | 9.1662 | 21.7807 | 25.1615 | 18.4258 |
|
| 78 |
+
| 2.6677 | 2.54 | 50000 | 2.4855 | 25.7435 | 9.145 | 21.765 | 25.0858 | 18.4329 |
|
| 79 |
+
| 2.6631 | 2.8 | 55000 | 2.4836 | 25.9411 | 9.226 | 21.9087 | 25.2863 | 18.4076 |
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
### Framework versions
|
| 83 |
+
|
| 84 |
+
- Transformers 4.18.0
|
| 85 |
+
- Pytorch 1.10.0+cu111
|
| 86 |
+
- Datasets 2.0.0
|
| 87 |
+
- Tokenizers 0.11.6
|