update model card README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,8 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
|
|
|
|
|
|
| 4 |
model-index:
|
| 5 |
- name: BERT2BERT_pretrained_LC-QuAD_2.0
|
| 6 |
results: []
|
|
@@ -12,6 +14,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
# BERT2BERT_pretrained_LC-QuAD_2.0
|
| 13 |
|
| 14 |
This model was trained from scratch on an unknown dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
## Model description
|
| 17 |
|
|
@@ -32,22 +40,34 @@ More information needed
|
|
| 32 |
The following hyperparameters were used during training:
|
| 33 |
- learning_rate: 5e-05
|
| 34 |
- train_batch_size: 8
|
| 35 |
-
- eval_batch_size:
|
| 36 |
- seed: 42
|
| 37 |
-
- gradient_accumulation_steps: 100
|
| 38 |
-
- total_train_batch_size: 800
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
- lr_scheduler_type: linear
|
| 41 |
-
- lr_scheduler_warmup_steps:
|
| 42 |
-
- num_epochs:
|
| 43 |
|
| 44 |
### Training results
|
| 45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
|
| 47 |
|
| 48 |
### Framework versions
|
| 49 |
|
| 50 |
-
- Transformers 4.
|
| 51 |
- Pytorch 2.0.1+cu118
|
| 52 |
- Datasets 2.12.0
|
| 53 |
- Tokenizers 0.13.3
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- bleu
|
| 6 |
model-index:
|
| 7 |
- name: BERT2BERT_pretrained_LC-QuAD_2.0
|
| 8 |
results: []
|
|
|
|
| 14 |
# BERT2BERT_pretrained_LC-QuAD_2.0
|
| 15 |
|
| 16 |
This model was trained from scratch on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.9002
|
| 19 |
+
- Bleu: 79.5906
|
| 20 |
+
- Em: 0.08
|
| 21 |
+
- Rm: 0.9565
|
| 22 |
+
- Gen Len: 48.48
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 40 |
The following hyperparameters were used during training:
|
| 41 |
- learning_rate: 5e-05
|
| 42 |
- train_batch_size: 8
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
- seed: 42
|
|
|
|
|
|
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 2500
|
| 48 |
+
- num_epochs: 10.0
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
+
| Training Loss | Epoch | Step | Bleu | Em | Gen Len | Validation Loss | Rm |
|
| 53 |
+
|:-------------:|:-----:|:-----:|:-------:|:----:|:-------:|:---------------:|:------:|
|
| 54 |
+
| 7.2445 | 0.83 | 2000 | 1.523 | 0.0 | 204.48 | 7.5619 | nan |
|
| 55 |
+
| 3.4458 | 1.66 | 4000 | 16.0372 | 0.0 | 76.36 | 3.4202 | nan |
|
| 56 |
+
| 2.2571 | 2.49 | 6000 | 34.5826 | 0.0 | 48.98 | 2.1971 | 0.8125 |
|
| 57 |
+
| 1.5696 | 3.32 | 8000 | 53.8822 | 0.0 | 48.7 | 1.6802 | 1.0 |
|
| 58 |
+
| 1.1359 | 4.15 | 10000 | 64.5591 | 0.02 | 45.2 | 1.3464 | 0.973 |
|
| 59 |
+
| 0.9994 | 4.98 | 12000 | 68.0869 | 0.02 | 47.76 | 1.0576 | 0.8889 |
|
| 60 |
+
| 0.7275 | 5.81 | 14000 | 74.1032 | 0.02 | 46.52 | 0.9522 | 0.9556 |
|
| 61 |
+
| 0.5868 | 6.64 | 16000 | 0.9307 | 71.6124| 0.02 | 0.9556 | 47.52 |
|
| 62 |
+
| 0.4499 | 7.47 | 18000 | 0.8866 | 77.237| 0.06 | 0.9574 | 46.0 |
|
| 63 |
+
| 0.3515 | 8.3 | 20000 | 0.9070 | 77.5798| 0.08 | 0.9574 | 47.5 |
|
| 64 |
+
| 0.292 | 9.13 | 22000 | 0.8905 | 78.649| 0.06 | 0.9574 | 47.96 |
|
| 65 |
+
| 0.2658 | 9.96 | 24000 | 0.9002 | 79.5906| 0.08 | 0.9565 | 48.48 |
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
| 69 |
|
| 70 |
+
- Transformers 4.30.0.dev0
|
| 71 |
- Pytorch 2.0.1+cu118
|
| 72 |
- Datasets 2.12.0
|
| 73 |
- Tokenizers 0.13.3
|