abdoutony207
commited on
Commit
•
075155b
1
Parent(s):
8fd1023
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- opus100
|
7 |
+
metrics:
|
8 |
+
- bleu
|
9 |
+
model-index:
|
10 |
+
- name: m2m100_418M-evaluated-en-to-ar-2000instancesopus-leaningRate2e-05-batchSize16-20epoch-1
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Sequence-to-sequence Language Modeling
|
14 |
+
type: text2text-generation
|
15 |
+
dataset:
|
16 |
+
name: opus100
|
17 |
+
type: opus100
|
18 |
+
args: ar-en
|
19 |
+
metrics:
|
20 |
+
- name: Bleu
|
21 |
+
type: bleu
|
22 |
+
value: 13.1835
|
23 |
+
---
|
24 |
+
|
25 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
26 |
+
should probably proofread and complete it, then remove this comment. -->
|
27 |
+
|
28 |
+
# m2m100_418M-evaluated-en-to-ar-2000instancesopus-leaningRate2e-05-batchSize16-20epoch-1
|
29 |
+
|
30 |
+
This model is a fine-tuned version of [facebook/m2m100_418M](https://huggingface.co/facebook/m2m100_418M) on the opus100 dataset.
|
31 |
+
It achieves the following results on the evaluation set:
|
32 |
+
- Loss: 0.3640
|
33 |
+
- Bleu: 13.1835
|
34 |
+
- Meteor: 0.1189
|
35 |
+
- Gen Len: 17.72
|
36 |
+
|
37 |
+
## Model description
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Intended uses & limitations
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Training and evaluation data
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training procedure
|
50 |
+
|
51 |
+
### Training hyperparameters
|
52 |
+
|
53 |
+
The following hyperparameters were used during training:
|
54 |
+
- learning_rate: 2e-05
|
55 |
+
- train_batch_size: 16
|
56 |
+
- eval_batch_size: 16
|
57 |
+
- seed: 42
|
58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
+
- lr_scheduler_type: linear
|
60 |
+
- num_epochs: 20
|
61 |
+
- mixed_precision_training: Native AMP
|
62 |
+
|
63 |
+
### Training results
|
64 |
+
|
65 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Meteor | Gen Len |
|
66 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|
|
67 |
+
| 6.1776 | 1.0 | 100 | 3.8904 | 10.5866 | 0.0995 | 16.64 |
|
68 |
+
| 2.4531 | 2.0 | 200 | 1.0928 | 12.3452 | 0.1108 | 17.0575 |
|
69 |
+
| 0.512 | 3.0 | 300 | 0.3625 | 10.5224 | 0.0982 | 17.2575 |
|
70 |
+
| 0.1924 | 4.0 | 400 | 0.3342 | 12.4242 | 0.1098 | 16.6325 |
|
71 |
+
| 0.1227 | 5.0 | 500 | 0.3403 | 13.0526 | 0.1185 | 17.3475 |
|
72 |
+
| 0.0889 | 6.0 | 600 | 0.3481 | 13.1323 | 0.1133 | 17.815 |
|
73 |
+
| 0.0651 | 7.0 | 700 | 0.3601 | 12.6684 | 0.1133 | 17.3525 |
|
74 |
+
| 0.0533 | 8.0 | 800 | 0.3640 | 13.1835 | 0.1189 | 17.72 |
|
75 |
+
|
76 |
+
|
77 |
+
### Framework versions
|
78 |
+
|
79 |
+
- Transformers 4.18.0
|
80 |
+
- Pytorch 1.11.0
|
81 |
+
- Datasets 2.1.0
|
82 |
+
- Tokenizers 0.12.1
|