version 1
Browse files- README.md +27 -6
- generation_config.json +4 -0
README.md
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
---
|
| 2 |
-
license:
|
| 3 |
-
base_model: facebook/bart-large
|
| 4 |
tags:
|
| 5 |
- text2text-generation
|
| 6 |
- generated_from_trainer
|
|
|
|
|
|
|
| 7 |
model-index:
|
| 8 |
- name: model
|
| 9 |
results: []
|
|
@@ -14,7 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# model
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [facebook/bart-large
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -34,16 +39,32 @@ More information needed
|
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
- learning_rate: 1e-05
|
| 37 |
-
- train_batch_size:
|
| 38 |
-
- eval_batch_size:
|
| 39 |
- seed: 42
|
| 40 |
- gradient_accumulation_steps: 4
|
| 41 |
-
- total_train_batch_size:
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
- num_epochs: 10
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
### Framework versions
|
| 48 |
|
| 49 |
- Transformers 4.39.3
|
|
|
|
| 1 |
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: facebook/bart-large
|
| 4 |
tags:
|
| 5 |
- text2text-generation
|
| 6 |
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- sacrebleu
|
| 9 |
model-index:
|
| 10 |
- name: model
|
| 11 |
results: []
|
|
|
|
| 16 |
|
| 17 |
# model
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [facebook/bart-large](https://huggingface.co/facebook/bart-large) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.6118
|
| 22 |
+
- Sacrebleu: 67.0674
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 39 |
|
| 40 |
The following hyperparameters were used during training:
|
| 41 |
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 16
|
| 43 |
+
- eval_batch_size: 16
|
| 44 |
- seed: 42
|
| 45 |
- gradient_accumulation_steps: 4
|
| 46 |
+
- total_train_batch_size: 64
|
| 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 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|
|
| 56 |
+
| No log | 1.0 | 218 | 0.5748 | 65.7729 |
|
| 57 |
+
| No log | 2.0 | 437 | 0.5609 | 66.4458 |
|
| 58 |
+
| No log | 3.0 | 656 | 0.5786 | 66.8800 |
|
| 59 |
+
| No log | 4.0 | 875 | 0.5485 | 66.8774 |
|
| 60 |
+
| No log | 5.0 | 1093 | 0.5506 | 66.8932 |
|
| 61 |
+
| No log | 6.0 | 1312 | 0.5890 | 66.9609 |
|
| 62 |
+
| No log | 7.0 | 1531 | 0.5674 | 66.8977 |
|
| 63 |
+
| No log | 8.0 | 1750 | 0.5903 | 66.9902 |
|
| 64 |
+
| No log | 9.0 | 1968 | 0.6196 | 67.0418 |
|
| 65 |
+
| No log | 9.97 | 2180 | 0.6118 | 67.0674 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
### Framework versions
|
| 69 |
|
| 70 |
- Transformers 4.39.3
|
generation_config.json
CHANGED
|
@@ -1,8 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"bos_token_id": 0,
|
| 3 |
"decoder_start_token_id": 2,
|
|
|
|
| 4 |
"eos_token_id": 2,
|
|
|
|
| 5 |
"forced_eos_token_id": 2,
|
|
|
|
|
|
|
| 6 |
"pad_token_id": 1,
|
| 7 |
"transformers_version": "4.39.3"
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"bos_token_id": 0,
|
| 3 |
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 0,
|
| 7 |
"forced_eos_token_id": 2,
|
| 8 |
+
"no_repeat_ngram_size": 3,
|
| 9 |
+
"num_beams": 4,
|
| 10 |
"pad_token_id": 1,
|
| 11 |
"transformers_version": "4.39.3"
|
| 12 |
}
|