Hugging Face's logo

kaifanli
/
bart-base-japanese-tobyoki-pairwise

Text2Text Generation
Transformers
PyTorch
TensorBoard
Safetensors
mbart
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
kaifanli commited on
Commit
e7b293e
•
1 Parent(s): 390a995

Model save

Browse files
Files changed (3) hide show
  1. README.md +14 -19
  2. model.safetensors +1 -1
  3. runs/Feb21_17-15-41_kogecha/events.out.tfevents.1708503352.kogecha +2 -2
README.md CHANGED
@@ -17,12 +17,12 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [ku-nlp/bart-base-japanese](https://huggingface.co/ku-nlp/bart-base-japanese) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 2.3030
21
- - Rouge1: 20.3362
22
- - Rouge2: 3.6623
23
- - Rougel: 10.9234
24
- - Rougelsum: 16.1471
25
- - Gen Len: 289.1
26
 
27
  ## Model description
28
 
@@ -42,8 +42,8 @@ More information needed
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 5e-05
45
- - train_batch_size: 1
46
- - eval_batch_size: 1
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
@@ -51,17 +51,12 @@ The following hyperparameters were used during training:
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
55
- |:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
56
- | No log | 1.0 | 10 | 2.7065 | 17.4135 | 1.8293 | 10.3284 | 13.7135 | 300.0 |
57
- | No log | 2.0 | 20 | 2.4983 | 15.5531 | 1.4496 | 9.0138 | 12.8628 | 300.0 |
58
- | No log | 3.0 | 30 | 2.4181 | 17.7832 | 2.7959 | 9.969 | 13.5642 | 300.0 |
59
- | No log | 4.0 | 40 | 2.3930 | 20.0103 | 2.9976 | 11.0143 | 15.4415 | 300.0 |
60
- | No log | 5.0 | 50 | 2.3328 | 20.5428 | 3.6354 | 11.3166 | 16.0252 | 300.0 |
61
- | No log | 6.0 | 60 | 2.3246 | 24.0821 | 4.2604 | 12.4489 | 19.152 | 294.2 |
62
- | No log | 7.0 | 70 | 2.3077 | 20.4387 | 3.8043 | 11.4571 | 16.7923 | 300.0 |
63
- | No log | 8.0 | 80 | 2.3060 | 22.0434 | 4.3648 | 12.386 | 18.3828 | 300.0 |
64
- | No log | 9.0 | 90 | 2.3030 | 20.3362 | 3.6623 | 10.9234 | 16.1471 | 289.1 |
65
 
66
 
67
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [ku-nlp/bart-base-japanese](https://huggingface.co/ku-nlp/bart-base-japanese) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.0672
21
+ - Rouge1: 21.0544
22
+ - Rouge2: 10.5652
23
+ - Rougel: 16.8852
24
+ - Rougelsum: 18.96
25
+ - Gen Len: 63.0687
26
 
27
  ## Model description
28
 
 
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 5e-05
45
+ - train_batch_size: 8
46
+ - eval_batch_size: 8
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
 
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
55
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:--------:|
56
+ | 1.5614 | 1.0 | 550 | 0.5451 | 21.2559 | 5.6179 | 13.9926 | 18.8544 | 116.9277 |
57
+ | 0.6179 | 2.0 | 1100 | 0.1827 | 19.8231 | 7.6726 | 14.2842 | 17.1362 | 76.1356 |
58
+ | 0.3266 | 3.0 | 1650 | 0.0943 | 18.0723 | 7.6543 | 14.0923 | 15.9801 | 57.7577 |
59
+ | 0.2039 | 4.0 | 2200 | 0.0672 | 21.0544 | 10.5652 | 16.8852 | 18.96 | 63.0687 |
 
 
 
 
 
60
 
61
 
62
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4ba9aade624dad071fca482c4659a070b9f2a7369fb75b9f226d96be338f2cb5
3
  size 501748312
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60f1cf73003e01c263cd0f1467478ae6a7404aeaa6ea121af69f87d2748b025
3
  size 501748312
runs/Feb21_17-15-41_kogecha/events.out.tfevents.1708503352.kogecha CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aaf38c26a7d6a2e1971a8a10b25b90bd7c8b1d6c26316ad984b1f08ae7983ec9
3
- size 8196
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b77ab44c88da6f74684fea51c9db654674100c7a97ccc04fcba0a634a3f994c
3
+ size 9075