update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- cnn_dailymail
|
| 7 |
+
metrics:
|
| 8 |
+
- rouge
|
| 9 |
+
model-index:
|
| 10 |
+
- name: mt5-small-finetuned-cnn-dailymail
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Sequence-to-sequence Language Modeling
|
| 14 |
+
type: text2text-generation
|
| 15 |
+
dataset:
|
| 16 |
+
name: cnn_dailymail
|
| 17 |
+
type: cnn_dailymail
|
| 18 |
+
config: 3.0.0
|
| 19 |
+
split: train
|
| 20 |
+
args: 3.0.0
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Rouge1
|
| 23 |
+
type: rouge
|
| 24 |
+
value: 32.8352
|
| 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 |
+
# mt5-small-finetuned-cnn-dailymail
|
| 31 |
+
|
| 32 |
+
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the cnn_dailymail dataset.
|
| 33 |
+
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 1.7294
|
| 35 |
+
- Rouge1: 32.8352
|
| 36 |
+
- Rouge2: 17.0633
|
| 37 |
+
- Rougel: 29.0888
|
| 38 |
+
- Rougelsum: 30.8226
|
| 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: 5.6e-05
|
| 58 |
+
- train_batch_size: 32
|
| 59 |
+
- eval_batch_size: 32
|
| 60 |
+
- seed: 42
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- num_epochs: 8
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
| 68 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
|
| 69 |
+
| No log | 1.0 | 8973 | 1.9272 | 31.6634 | 16.1653 | 28.1624 | 29.7819 |
|
| 70 |
+
| No log | 2.0 | 17946 | 1.8282 | 32.1032 | 16.4388 | 28.4914 | 30.1856 |
|
| 71 |
+
| No log | 3.0 | 26919 | 1.7967 | 32.5721 | 16.8392 | 28.8483 | 30.5764 |
|
| 72 |
+
| 2.1615 | 4.0 | 35892 | 1.7640 | 32.6788 | 16.94 | 28.994 | 30.6883 |
|
| 73 |
+
| 2.1615 | 5.0 | 44865 | 1.7450 | 32.8129 | 17.048 | 29.0788 | 30.8106 |
|
| 74 |
+
| 2.1615 | 6.0 | 53838 | 1.7379 | 32.7074 | 16.9641 | 28.9745 | 30.7043 |
|
| 75 |
+
| 2.1615 | 7.0 | 62811 | 1.7317 | 32.7692 | 17.0116 | 29.0395 | 30.7685 |
|
| 76 |
+
| 2.0886 | 8.0 | 71784 | 1.7294 | 32.8352 | 17.0633 | 29.0888 | 30.8226 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- Transformers 4.24.0
|
| 82 |
+
- Pytorch 1.11.0+cu102
|
| 83 |
+
- Datasets 2.7.1
|
| 84 |
+
- Tokenizers 0.13.2
|