update model card README.md
Browse files
README.md
CHANGED
@@ -4,24 +4,9 @@ tags:
|
|
4 |
- generated_from_trainer
|
5 |
datasets:
|
6 |
- opus100
|
7 |
-
metrics:
|
8 |
-
- bleu
|
9 |
model-index:
|
10 |
- name: en2zh_opus_100
|
11 |
-
results:
|
12 |
-
- task:
|
13 |
-
name: Sequence-to-sequence Language Modeling
|
14 |
-
type: text2text-generation
|
15 |
-
dataset:
|
16 |
-
name: opus100
|
17 |
-
type: opus100
|
18 |
-
config: en-zh
|
19 |
-
split: test
|
20 |
-
args: en-zh
|
21 |
-
metrics:
|
22 |
-
- name: Bleu
|
23 |
-
type: bleu
|
24 |
-
value: 12.2521
|
25 |
---
|
26 |
|
27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -30,10 +15,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
30 |
# en2zh_opus_100
|
31 |
|
32 |
This model is a fine-tuned version of [HAJIWEE/en2zh_opus_100](https://huggingface.co/HAJIWEE/en2zh_opus_100) on the opus100 dataset.
|
33 |
-
It achieves the following results on the evaluation set:
|
34 |
-
- Loss: 2.2328
|
35 |
-
- Bleu: 12.2521
|
36 |
-
- Gen Len: 15.2895
|
37 |
|
38 |
## Model description
|
39 |
|
@@ -60,17 +41,9 @@ The following hyperparameters were used during training:
|
|
60 |
- lr_scheduler_type: linear
|
61 |
- num_epochs: 2
|
62 |
|
63 |
-
### Training results
|
64 |
-
|
65 |
-
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
66 |
-
|:-------------:|:-----:|:------:|:---------------:|:-------:|:-------:|
|
67 |
-
| 2.9054 | 1.0 | 166667 | 2.3065 | 12.141 | 15.299 |
|
68 |
-
| 2.764 | 2.0 | 333334 | 2.2328 | 12.2521 | 15.2895 |
|
69 |
-
|
70 |
-
|
71 |
### Framework versions
|
72 |
|
73 |
- Transformers 4.27.4
|
74 |
-
- Pytorch 2.0.0
|
75 |
- Datasets 2.11.0
|
76 |
-
- Tokenizers 0.13.
|
|
|
4 |
- generated_from_trainer
|
5 |
datasets:
|
6 |
- opus100
|
|
|
|
|
7 |
model-index:
|
8 |
- name: en2zh_opus_100
|
9 |
+
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
---
|
11 |
|
12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
15 |
# en2zh_opus_100
|
16 |
|
17 |
This model is a fine-tuned version of [HAJIWEE/en2zh_opus_100](https://huggingface.co/HAJIWEE/en2zh_opus_100) on the opus100 dataset.
|
|
|
|
|
|
|
|
|
18 |
|
19 |
## Model description
|
20 |
|
|
|
41 |
- lr_scheduler_type: linear
|
42 |
- num_epochs: 2
|
43 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
### Framework versions
|
45 |
|
46 |
- Transformers 4.27.4
|
47 |
+
- Pytorch 2.0.0+cu117
|
48 |
- Datasets 2.11.0
|
49 |
+
- Tokenizers 0.13.3
|