Commit
·
4245f0a
1
Parent(s):
38446bc
update model card README.md
Browse files
README.md
CHANGED
|
@@ -3,10 +3,23 @@ license: apache-2.0
|
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
-
-
|
|
|
|
|
|
|
| 7 |
model-index:
|
| 8 |
- name: opus-mt-tr-en-finetuned-tr-to-en
|
| 9 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -14,7 +27,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# opus-mt-tr-en-finetuned-tr-to-en
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-tr-en](https://huggingface.co/Helsinki-NLP/opus-mt-tr-en) on the
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -33,18 +50,27 @@ More information needed
|
|
| 33 |
### Training hyperparameters
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
-
- learning_rate:
|
| 37 |
-
- train_batch_size:
|
| 38 |
-
- eval_batch_size:
|
| 39 |
- seed: 42
|
| 40 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 41 |
- lr_scheduler_type: linear
|
| 42 |
-
- num_epochs:
|
| 43 |
- mixed_precision_training: Native AMP
|
| 44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
### Framework versions
|
| 46 |
|
| 47 |
- Transformers 4.19.2
|
| 48 |
-
- Pytorch 1.
|
| 49 |
-
- Datasets 2.2.
|
| 50 |
- Tokenizers 0.12.1
|
|
|
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
+
- opus_infopankki
|
| 7 |
+
metrics:
|
| 8 |
+
- bleu
|
| 9 |
model-index:
|
| 10 |
- name: opus-mt-tr-en-finetuned-tr-to-en
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Sequence-to-sequence Language Modeling
|
| 14 |
+
type: text2text-generation
|
| 15 |
+
dataset:
|
| 16 |
+
name: opus_infopankki
|
| 17 |
+
type: opus_infopankki
|
| 18 |
+
args: en-tr
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Bleu
|
| 21 |
+
type: bleu
|
| 22 |
+
value: 37.5209
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# opus-mt-tr-en-finetuned-tr-to-en
|
| 29 |
|
| 30 |
+
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-tr-en](https://huggingface.co/Helsinki-NLP/opus-mt-tr-en) on the opus_infopankki dataset.
|
| 31 |
+
It achieves the following results on the evaluation set:
|
| 32 |
+
- Loss: 1.3456
|
| 33 |
+
- Bleu: 37.5209
|
| 34 |
+
- Gen Len: 13.5457
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
|
|
|
| 50 |
### Training hyperparameters
|
| 51 |
|
| 52 |
The following hyperparameters were used during training:
|
| 53 |
+
- learning_rate: 2e-06
|
| 54 |
+
- train_batch_size: 32
|
| 55 |
+
- eval_batch_size: 32
|
| 56 |
- seed: 42
|
| 57 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 58 |
- lr_scheduler_type: linear
|
| 59 |
+
- num_epochs: 3
|
| 60 |
- mixed_precision_training: Native AMP
|
| 61 |
|
| 62 |
+
### Training results
|
| 63 |
+
|
| 64 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
| 65 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
|
| 66 |
+
| No log | 1.0 | 138 | 1.4283 | 35.8087 | 13.5806 |
|
| 67 |
+
| No log | 2.0 | 276 | 1.3649 | 36.8833 | 13.5446 |
|
| 68 |
+
| No log | 3.0 | 414 | 1.3456 | 37.5209 | 13.5457 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
### Framework versions
|
| 72 |
|
| 73 |
- Transformers 4.19.2
|
| 74 |
+
- Pytorch 1.7.1+cu110
|
| 75 |
+
- Datasets 2.2.2
|
| 76 |
- Tokenizers 0.12.1
|