Hugging Face's logo

judy93536
/
pegasus-xsum

Text2Text Generation
Transformers
PyTorch
pegasus
Generated from Trainer
Inference Endpoints
Model card Files Community
judy93536 commited on
Commit
5782631
•
1 Parent(s): 51bce5c

End of training

Browse files
Files changed (3) hide show
  1. README.md +13 -28
  2. generation_config.json +1 -1
  3. pytorch_model.bin +1 -1
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [google/pegasus-xsum](https://huggingface.co/google/pegasus-xsum) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.1676
18
 
19
  ## Model description
20
 
@@ -34,43 +34,28 @@ More information needed
34
 
35
  The following hyperparameters were used during training:
36
  - learning_rate: 0.0001
37
- - train_batch_size: 4
38
- - eval_batch_size: 4
39
  - seed: 42
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
  - lr_scheduler_warmup_steps: 500
43
- - num_epochs: 8
44
 
45
  ### Training results
46
 
47
- | Training Loss | Epoch | Step | Validation Loss |
48
- |:-------------:|:-----:|:-----:|:---------------:|
49
- | 2.2677 | 0.4 | 1000 | 0.1598 |
50
- | 0.0619 | 0.8 | 2000 | 0.1363 |
51
- | 0.0478 | 1.2 | 3000 | 0.1305 |
52
- | 0.0421 | 1.6 | 4000 | 0.1298 |
53
- | 0.0403 | 2.0 | 5000 | 0.1256 |
54
- | 0.0328 | 2.4 | 6000 | 0.1309 |
55
- | 0.0314 | 2.8 | 7000 | 0.1296 |
56
- | 0.0294 | 3.2 | 8000 | 0.1381 |
57
- | 0.0259 | 3.6 | 9000 | 0.1383 |
58
- | 0.0251 | 4.0 | 10000 | 0.1380 |
59
- | 0.0213 | 4.4 | 11000 | 0.1450 |
60
- | 0.0206 | 4.8 | 12000 | 0.1427 |
61
- | 0.0201 | 5.2 | 13000 | 0.1520 |
62
- | 0.0189 | 5.6 | 14000 | 0.1505 |
63
- | 0.0164 | 6.0 | 15000 | 0.1515 |
64
- | 0.0153 | 6.4 | 16000 | 0.1594 |
65
- | 0.0153 | 6.8 | 17000 | 0.1570 |
66
- | 0.0149 | 7.2 | 18000 | 0.1652 |
67
- | 0.0136 | 7.6 | 19000 | 0.1662 |
68
- | 0.013 | 8.0 | 20000 | 0.1676 |
69
 
70
 
71
  ### Framework versions
72
 
73
- - Transformers 4.34.0
74
- - Pytorch 2.0.1+cu118
75
  - Datasets 2.14.5
76
  - Tokenizers 0.14.1
 
14
 
15
  This model is a fine-tuned version of [google/pegasus-xsum](https://huggingface.co/google/pegasus-xsum) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.1821
18
 
19
  ## Model description
20
 
 
34
 
35
  The following hyperparameters were used during training:
36
  - learning_rate: 0.0001
37
+ - train_batch_size: 8
38
+ - eval_batch_size: 8
39
  - seed: 42
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
  - lr_scheduler_warmup_steps: 500
43
+ - num_epochs: 16
44
 
45
  ### Training results
46
 
47
+ | Training Loss | Epoch | Step | Validation Loss |
48
+ |:-------------:|:-----:|:----:|:---------------:|
49
+ | 1.8765 | 3.19 | 1000 | 0.1773 |
50
+ | 0.0278 | 6.39 | 2000 | 0.1685 |
51
+ | 0.0175 | 9.58 | 3000 | 0.1647 |
52
+ | 0.0127 | 12.78 | 4000 | 0.1771 |
53
+ | 0.0103 | 15.97 | 5000 | 0.1821 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
54
 
55
 
56
  ### Framework versions
57
 
58
+ - Transformers 4.34.1
59
+ - Pytorch 2.1.0+cu118
60
  - Datasets 2.14.5
61
  - Tokenizers 0.14.1
generation_config.json CHANGED
@@ -8,5 +8,5 @@
8
  "max_length": 64,
9
  "num_beams": 8,
10
  "pad_token_id": 0,
11
- "transformers_version": "4.34.0"
12
  }
 
8
  "max_length": 64,
9
  "num_beams": 8,
10
  "pad_token_id": 0,
11
+ "transformers_version": "4.34.1"
12
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:104bf2ab9876d25edc510318c9447d48cd9ad5cb4c07e66b87a88ce9113e8505
3
  size 2279610794
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:761d2b18d58a47e963b10ba6b10772c8dccd4650420382e3510c494f8f5e6244
3
  size 2279610794