raquelclemente
commited on
Commit
•
d2f5c65
1
Parent(s):
6146e53
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
metrics:
|
5 |
+
- rouge
|
6 |
+
model-index:
|
7 |
+
- name: meeting-sensai
|
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 |
+
# meeting-sensai
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [raquelclemente/tmp_trainer](https://huggingface.co/raquelclemente/tmp_trainer) on the None dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 6.7917
|
19 |
+
- Rouge1: 0.4205
|
20 |
+
- Rouge2: 0.1795
|
21 |
+
- Rougel: 0.3536
|
22 |
+
- Rougelsum: 0.3536
|
23 |
+
|
24 |
+
## Model description
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Intended uses & limitations
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training and evaluation data
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training procedure
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 0.0005
|
42 |
+
- train_batch_size: 2
|
43 |
+
- eval_batch_size: 2
|
44 |
+
- seed: 42
|
45 |
+
- gradient_accumulation_steps: 16
|
46 |
+
- total_train_batch_size: 32
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: linear
|
49 |
+
- lr_scheduler_warmup_steps: 5
|
50 |
+
- num_epochs: 50
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
|
56 |
+
| No log | 2.46 | 5 | 3.6391 | 0.4237 | 0.1996 | 0.3660 | 0.3660 |
|
57 |
+
| No log | 4.91 | 10 | 3.4021 | 0.4118 | 0.2142 | 0.3778 | 0.3778 |
|
58 |
+
| No log | 7.46 | 15 | 3.5879 | 0.4314 | 0.2218 | 0.3965 | 0.3965 |
|
59 |
+
| No log | 9.91 | 20 | 3.8672 | 0.4730 | 0.2043 | 0.4003 | 0.4003 |
|
60 |
+
| No log | 12.46 | 25 | 3.8721 | 0.4083 | 0.1885 | 0.3675 | 0.3675 |
|
61 |
+
| No log | 14.91 | 30 | 4.0905 | 0.4609 | 0.2105 | 0.4070 | 0.4070 |
|
62 |
+
| No log | 17.46 | 35 | 4.4786 | 0.4580 | 0.1856 | 0.3739 | 0.3739 |
|
63 |
+
| No log | 19.91 | 40 | 4.9112 | 0.4689 | 0.2573 | 0.4224 | 0.4224 |
|
64 |
+
| No log | 22.46 | 45 | 5.2165 | 0.4698 | 0.2151 | 0.4096 | 0.4096 |
|
65 |
+
| No log | 24.91 | 50 | 5.5533 | 0.4014 | 0.1837 | 0.3062 | 0.3062 |
|
66 |
+
| No log | 27.46 | 55 | 5.8611 | 0.4109 | 0.1854 | 0.3686 | 0.3686 |
|
67 |
+
| No log | 29.91 | 60 | 5.8393 | 0.4715 | 0.2370 | 0.4074 | 0.4074 |
|
68 |
+
| No log | 32.46 | 65 | 6.1874 | 0.4297 | 0.1898 | 0.3479 | 0.3479 |
|
69 |
+
| No log | 34.91 | 70 | 6.3194 | 0.4350 | 0.2207 | 0.3802 | 0.3802 |
|
70 |
+
| No log | 37.46 | 75 | 6.2460 | 0.4009 | 0.2056 | 0.3518 | 0.3518 |
|
71 |
+
| No log | 39.91 | 80 | 6.5997 | 0.3992 | 0.1869 | 0.3713 | 0.3713 |
|
72 |
+
| No log | 42.46 | 85 | 6.5516 | 0.4345 | 0.1928 | 0.3756 | 0.3756 |
|
73 |
+
| No log | 44.91 | 90 | 6.5791 | 0.4078 | 0.2036 | 0.3489 | 0.3489 |
|
74 |
+
| No log | 47.46 | 95 | 6.7799 | 0.4205 | 0.1795 | 0.3536 | 0.3536 |
|
75 |
+
| No log | 49.91 | 100 | 6.7917 | 0.4205 | 0.1795 | 0.3536 | 0.3536 |
|
76 |
+
|
77 |
+
|
78 |
+
### Framework versions
|
79 |
+
|
80 |
+
- Transformers 4.26.1
|
81 |
+
- Pytorch 1.13.0
|
82 |
+
- Datasets 2.1.0
|
83 |
+
- Tokenizers 0.13.2
|