Commit
•
227b859
1
Parent(s):
5dbb5a0
update model card README.md
Browse files
README.md
CHANGED
|
@@ -2,8 +2,6 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
-
metrics:
|
| 6 |
-
- bleu
|
| 7 |
model-index:
|
| 8 |
- name: tokipona_to_eng_model_v0.4
|
| 9 |
results: []
|
|
@@ -16,9 +14,14 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
-
|
| 20 |
-
-
|
| 21 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -37,36 +40,15 @@ More information needed
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
-
- learning_rate:
|
| 41 |
- train_batch_size: 16
|
| 42 |
- eval_batch_size: 16
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
-
- num_epochs:
|
| 47 |
- mixed_precision_training: Native AMP
|
| 48 |
|
| 49 |
-
### Training results
|
| 50 |
-
|
| 51 |
-
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
| 52 |
-
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
|
| 53 |
-
| No log | 1.0 | 61 | 3.9664 | 0.2896 | 14.4631 |
|
| 54 |
-
| No log | 2.0 | 122 | 3.9429 | 0.4145 | 14.1527 |
|
| 55 |
-
| No log | 3.0 | 183 | 3.9243 | 0.5075 | 14.3103 |
|
| 56 |
-
| No log | 4.0 | 244 | 3.9089 | 0.4954 | 14.3498 |
|
| 57 |
-
| No log | 5.0 | 305 | 3.8915 | 0.4979 | 13.8966 |
|
| 58 |
-
| No log | 6.0 | 366 | 3.8795 | 0.6291 | 13.2611 |
|
| 59 |
-
| No log | 7.0 | 427 | 3.8690 | 0.6926 | 13.4138 |
|
| 60 |
-
| No log | 8.0 | 488 | 3.8568 | 0.6898 | 13.0493 |
|
| 61 |
-
| 4.0338 | 9.0 | 549 | 3.8464 | 0.7046 | 12.9212 |
|
| 62 |
-
| 4.0338 | 10.0 | 610 | 3.8395 | 0.6875 | 12.734 |
|
| 63 |
-
| 4.0338 | 11.0 | 671 | 3.8343 | 0.6789 | 12.9557 |
|
| 64 |
-
| 4.0338 | 12.0 | 732 | 3.8306 | 0.6854 | 12.8571 |
|
| 65 |
-
| 4.0338 | 13.0 | 793 | 3.8279 | 0.6854 | 12.8571 |
|
| 66 |
-
| 4.0338 | 14.0 | 854 | 3.8265 | 0.6855 | 12.7044 |
|
| 67 |
-
| 4.0338 | 15.0 | 915 | 3.8256 | 0.6866 | 12.7094 |
|
| 68 |
-
|
| 69 |
-
|
| 70 |
### Framework versions
|
| 71 |
|
| 72 |
- Transformers 4.28.1
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
|
|
|
|
|
|
| 5 |
model-index:
|
| 6 |
- name: tokipona_to_eng_model_v0.4
|
| 7 |
results: []
|
|
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
+
- eval_loss: 3.7209
|
| 18 |
+
- eval_bleu: 1.4637
|
| 19 |
+
- eval_gen_len: 13.0049
|
| 20 |
+
- eval_runtime: 2.6154
|
| 21 |
+
- eval_samples_per_second: 77.617
|
| 22 |
+
- eval_steps_per_second: 4.971
|
| 23 |
+
- epoch: 5.0
|
| 24 |
+
- step: 305
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 40 |
### Training hyperparameters
|
| 41 |
|
| 42 |
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 9e-05
|
| 44 |
- train_batch_size: 16
|
| 45 |
- eval_batch_size: 16
|
| 46 |
- seed: 42
|
| 47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
- lr_scheduler_type: linear
|
| 49 |
+
- num_epochs: 10
|
| 50 |
- mixed_precision_training: Native AMP
|
| 51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
### Framework versions
|
| 53 |
|
| 54 |
- Transformers 4.28.1
|