Commit
•
1a59aef
1
Parent(s):
141ddbf
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
metrics:
|
5 |
+
- rouge
|
6 |
+
model-index:
|
7 |
+
- name: bert2gpt2_med_ml_orange_summ-finetuned_med_sum_new
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# bert2gpt2_med_ml_orange_summ-finetuned_med_sum_new
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [Chemsseddine/bert2gpt2_med_ml_orange_summ](https://huggingface.co/Chemsseddine/bert2gpt2_med_ml_orange_summ) on the None dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 2.5474
|
19 |
+
- Rouge1: 31.8871
|
20 |
+
- Rouge2: 14.4411
|
21 |
+
- Rougel: 31.6716
|
22 |
+
- Rougelsum: 31.579
|
23 |
+
- Gen Len: 22.8412
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 2e-05
|
43 |
+
- train_batch_size: 1
|
44 |
+
- eval_batch_size: 1
|
45 |
+
- seed: 42
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: linear
|
48 |
+
- num_epochs: 10
|
49 |
+
- mixed_precision_training: Native AMP
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
55 |
+
| 2.5621 | 1.0 | 900 | 1.9724 | 30.3731 | 13.8412 | 29.9606 | 29.9716 | 22.6353 |
|
56 |
+
| 1.3692 | 2.0 | 1800 | 1.9634 | 29.6409 | 13.7674 | 29.5202 | 29.5207 | 22.5059 |
|
57 |
+
| 0.8308 | 3.0 | 2700 | 2.1431 | 30.9317 | 14.5594 | 30.8021 | 30.7287 | 22.6118 |
|
58 |
+
| 0.4689 | 4.0 | 3600 | 2.2970 | 30.1132 | 14.6407 | 29.9657 | 30.0182 | 23.3235 |
|
59 |
+
| 0.2875 | 5.0 | 4500 | 2.3787 | 30.9378 | 14.7108 | 30.861 | 30.9097 | 22.7529 |
|
60 |
+
| 0.1564 | 6.0 | 5400 | 2.4137 | 30.5338 | 13.9702 | 30.1252 | 30.1975 | 23.1588 |
|
61 |
+
| 0.1007 | 7.0 | 6300 | 2.4822 | 30.872 | 14.9353 | 30.835 | 30.7694 | 23.0529 |
|
62 |
+
| 0.0783 | 8.0 | 7200 | 2.4974 | 29.9825 | 14.1702 | 29.7507 | 29.7271 | 23.1882 |
|
63 |
+
| 0.0504 | 9.0 | 8100 | 2.5175 | 31.96 | 15.0705 | 31.9669 | 31.9839 | 23.0588 |
|
64 |
+
| 0.0339 | 10.0 | 9000 | 2.5474 | 31.8871 | 14.4411 | 31.6716 | 31.579 | 22.8412 |
|
65 |
+
|
66 |
+
|
67 |
+
### Framework versions
|
68 |
+
|
69 |
+
- Transformers 4.20.1
|
70 |
+
- Pytorch 1.11.0+cu113
|
71 |
+
- Datasets 2.3.2
|
72 |
+
- Tokenizers 0.12.1
|