Commit
•
daddfeb
1
Parent(s):
26fb891
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- summarization
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- samsum
|
| 7 |
+
metrics:
|
| 8 |
+
- rouge
|
| 9 |
+
model-index:
|
| 10 |
+
- name: ssr-base-finetuned-samsum-en
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Sequence-to-sequence Language Modeling
|
| 14 |
+
type: text2text-generation
|
| 15 |
+
dataset:
|
| 16 |
+
name: samsum
|
| 17 |
+
type: samsum
|
| 18 |
+
args: samsum
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Rouge1
|
| 21 |
+
type: rouge
|
| 22 |
+
value: 46.7505
|
| 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 |
+
# ssr-base-finetuned-samsum-en
|
| 29 |
+
|
| 30 |
+
This model is a fine-tuned version of [microsoft/ssr-base](https://huggingface.co/microsoft/ssr-base) on the samsum dataset.
|
| 31 |
+
It achieves the following results on the evaluation set:
|
| 32 |
+
- Loss: 1.6231
|
| 33 |
+
- Rouge1: 46.7505
|
| 34 |
+
- Rouge2: 22.3968
|
| 35 |
+
- Rougel: 37.1784
|
| 36 |
+
- Rougelsum: 42.891
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 5.6e-05
|
| 56 |
+
- train_batch_size: 10
|
| 57 |
+
- eval_batch_size: 10
|
| 58 |
+
- seed: 42
|
| 59 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
+
- lr_scheduler_type: linear
|
| 61 |
+
- num_epochs: 10
|
| 62 |
+
|
| 63 |
+
### Training results
|
| 64 |
+
|
| 65 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
| 66 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
|
| 67 |
+
| 1.9682 | 1.0 | 300 | 1.6432 | 44.2182 | 20.8486 | 35.0914 | 40.9852 |
|
| 68 |
+
| 1.6475 | 2.0 | 600 | 1.5946 | 45.3919 | 21.6955 | 36.2411 | 41.8532 |
|
| 69 |
+
| 1.5121 | 3.0 | 900 | 1.5737 | 46.1769 | 22.4178 | 36.9762 | 42.6614 |
|
| 70 |
+
| 1.4112 | 4.0 | 1200 | 1.5774 | 46.6047 | 22.8227 | 37.2457 | 43.1935 |
|
| 71 |
+
| 1.323 | 5.0 | 1500 | 1.5825 | 46.6162 | 22.485 | 37.2846 | 42.9834 |
|
| 72 |
+
| 1.2613 | 6.0 | 1800 | 1.5883 | 46.4253 | 22.1199 | 37.0491 | 42.5189 |
|
| 73 |
+
| 1.2077 | 7.0 | 2100 | 1.5965 | 46.485 | 22.3636 | 37.2677 | 42.7499 |
|
| 74 |
+
| 1.1697 | 8.0 | 2400 | 1.6174 | 46.8654 | 22.6291 | 37.4201 | 43.0875 |
|
| 75 |
+
| 1.1367 | 9.0 | 2700 | 1.6188 | 46.707 | 22.305 | 37.156 | 42.9087 |
|
| 76 |
+
| 1.118 | 10.0 | 3000 | 1.6231 | 46.7505 | 22.3968 | 37.1784 | 42.891 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- Transformers 4.19.2
|
| 82 |
+
- Pytorch 1.11.0+cu113
|
| 83 |
+
- Datasets 2.2.2
|
| 84 |
+
- Tokenizers 0.12.1
|