Commit
•
8685653
1
Parent(s):
bbc7f59
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- rouge
|
7 |
+
model-index:
|
8 |
+
- name: bart-large-cnn-finetuned-pubmed-finetuned-roundup-e16
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# bart-large-cnn-finetuned-pubmed-finetuned-roundup-e16
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [theojolliffe/bart-large-cnn-finetuned-pubmed](https://huggingface.co/theojolliffe/bart-large-cnn-finetuned-pubmed) on an unknown dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 2.6815
|
20 |
+
- Rouge1: 48.7608
|
21 |
+
- Rouge2: 29.554
|
22 |
+
- Rougel: 30.5554
|
23 |
+
- Rougelsum: 46.4001
|
24 |
+
- Gen Len: 142.0
|
25 |
+
|
26 |
+
## Model description
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Intended uses & limitations
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training and evaluation data
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training procedure
|
39 |
+
|
40 |
+
### Training hyperparameters
|
41 |
+
|
42 |
+
The following hyperparameters were used during training:
|
43 |
+
- learning_rate: 2e-05
|
44 |
+
- train_batch_size: 2
|
45 |
+
- eval_batch_size: 2
|
46 |
+
- seed: 42
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: linear
|
49 |
+
- num_epochs: 16
|
50 |
+
- mixed_precision_training: Native AMP
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
56 |
+
| No log | 1.0 | 25 | 1.4287 | 46.5701 | 28.6267 | 34.7827 | 45.0622 | 142.0 |
|
57 |
+
| No log | 2.0 | 50 | 1.4419 | 46.6171 | 27.4276 | 31.0085 | 43.1797 | 142.0 |
|
58 |
+
| No log | 3.0 | 75 | 1.5418 | 50.1144 | 29.3433 | 32.0144 | 46.9217 | 142.0 |
|
59 |
+
| No log | 4.0 | 100 | 1.7125 | 49.1395 | 28.611 | 30.9759 | 46.8346 | 142.0 |
|
60 |
+
| No log | 5.0 | 125 | 1.8978 | 43.9629 | 24.1224 | 26.0032 | 41.2272 | 142.0 |
|
61 |
+
| No log | 6.0 | 150 | 2.0990 | 49.0579 | 29.5182 | 31.5829 | 46.0207 | 142.0 |
|
62 |
+
| No log | 7.0 | 175 | 2.2380 | 48.8754 | 27.7691 | 28.8597 | 45.3281 | 142.0 |
|
63 |
+
| No log | 8.0 | 200 | 2.2922 | 48.311 | 29.2517 | 33.8241 | 46.6099 | 142.0 |
|
64 |
+
| No log | 9.0 | 225 | 2.3820 | 45.4663 | 23.9904 | 27.5497 | 41.9446 | 142.0 |
|
65 |
+
| No log | 10.0 | 250 | 2.4856 | 48.2224 | 27.7455 | 28.159 | 45.4726 | 142.0 |
|
66 |
+
| No log | 11.0 | 275 | 2.4731 | 46.1799 | 22.1941 | 26.8254 | 43.9986 | 142.0 |
|
67 |
+
| No log | 12.0 | 300 | 2.5278 | 47.8623 | 27.6514 | 26.6377 | 42.9255 | 142.0 |
|
68 |
+
| No log | 13.0 | 325 | 2.6229 | 45.573 | 25.4966 | 27.7158 | 42.2306 | 142.0 |
|
69 |
+
| No log | 14.0 | 350 | 2.6032 | 48.1972 | 27.0387 | 28.336 | 45.0293 | 142.0 |
|
70 |
+
| No log | 15.0 | 375 | 2.6600 | 47.7301 | 27.3567 | 29.3389 | 44.3516 | 142.0 |
|
71 |
+
| No log | 16.0 | 400 | 2.6815 | 48.7608 | 29.554 | 30.5554 | 46.4001 | 142.0 |
|
72 |
+
|
73 |
+
|
74 |
+
### Framework versions
|
75 |
+
|
76 |
+
- Transformers 4.18.0
|
77 |
+
- Pytorch 1.11.0+cu113
|
78 |
+
- Datasets 2.1.0
|
79 |
+
- Tokenizers 0.12.1
|