|
|
--- |
|
|
library_name: peft |
|
|
license: apache-2.0 |
|
|
base_model: allenai/led-base-16384 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- rouge |
|
|
- bleu |
|
|
- precision |
|
|
- recall |
|
|
- f1 |
|
|
model-index: |
|
|
- name: Lora_LED_sum_approach |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
|
|
# Lora_LED_sum_approach |
|
|
|
|
|
This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 3.5646 |
|
|
- Rouge1: 0.4521 |
|
|
- Rouge2: 0.2422 |
|
|
- Rougel: 0.3904 |
|
|
- Rougelsum: 0.3905 |
|
|
- Gen Len: 29.4 |
|
|
- Bleu: 0.1533 |
|
|
- Precisions: 0.2152 |
|
|
- Brevity Penalty: 0.8831 |
|
|
- Length Ratio: 0.8894 |
|
|
- Translation Length: 1086.0 |
|
|
- Reference Length: 1221.0 |
|
|
- Precision: 0.9043 |
|
|
- Recall: 0.9002 |
|
|
- F1: 0.9021 |
|
|
- Hashcode: roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
|
|
|
|
|
## Model description |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Intended uses & limitations |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Training and evaluation data |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Training procedure |
|
|
|
|
|
### Training hyperparameters |
|
|
|
|
|
The following hyperparameters were used during training: |
|
|
- learning_rate: 0.001 |
|
|
- train_batch_size: 1 |
|
|
- eval_batch_size: 1 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 16 |
|
|
- total_train_batch_size: 16 |
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 10 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | Bleu | Precisions | Brevity Penalty | Length Ratio | Translation Length | Reference Length | Precision | Recall | F1 | Hashcode | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|:------:|:----------:|:---------------:|:------------:|:------------------:|:----------------:|:---------:|:------:|:------:|:---------------------------------------------------------:| |
|
|
| 8.0757 | 1.0 | 7 | 7.6798 | 0.3128 | 0.1085 | 0.253 | 0.2533 | 32.0 | 0.0733 | 0.1062 | 1.0 | 1.0663 | 1302.0 | 1221.0 | 0.8685 | 0.8728 | 0.8706 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 6.5609 | 2.0 | 14 | 5.7642 | 0.4165 | 0.2088 | 0.3627 | 0.3626 | 30.64 | 0.1358 | 0.1742 | 1.0 | 1.036 | 1265.0 | 1221.0 | 0.8922 | 0.8861 | 0.889 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 4.9145 | 3.0 | 21 | 4.4340 | 0.4234 | 0.2265 | 0.3669 | 0.3685 | 25.84 | 0.1246 | 0.2092 | 0.765 | 0.7887 | 963.0 | 1221.0 | 0.9057 | 0.894 | 0.8996 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 4.0682 | 4.0 | 28 | 3.9241 | 0.4454 | 0.2452 | 0.3952 | 0.3971 | 27.26 | 0.1446 | 0.2209 | 0.8115 | 0.8272 | 1010.0 | 1221.0 | 0.9059 | 0.8983 | 0.9019 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 3.6834 | 5.0 | 35 | 3.7361 | 0.4521 | 0.237 | 0.3828 | 0.3837 | 27.58 | 0.1433 | 0.2137 | 0.8327 | 0.8452 | 1032.0 | 1221.0 | 0.9031 | 0.8973 | 0.9 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 3.5042 | 6.0 | 42 | 3.6285 | 0.4567 | 0.247 | 0.3901 | 0.3908 | 27.86 | 0.1451 | 0.2184 | 0.8336 | 0.846 | 1033.0 | 1221.0 | 0.9067 | 0.9003 | 0.9033 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 3.4173 | 7.0 | 49 | 3.5881 | 0.4458 | 0.2389 | 0.3839 | 0.3852 | 27.16 | 0.1439 | 0.2226 | 0.7929 | 0.8116 | 991.0 | 1221.0 | 0.9056 | 0.8973 | 0.9013 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 3.3572 | 8.0 | 56 | 3.5698 | 0.4514 | 0.2331 | 0.3836 | 0.3862 | 29.12 | 0.147 | 0.2081 | 0.884 | 0.8903 | 1087.0 | 1221.0 | 0.9026 | 0.8994 | 0.9009 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 3.3165 | 9.0 | 63 | 3.5700 | 0.4592 | 0.2422 | 0.3954 | 0.3957 | 29.28 | 0.1502 | 0.2113 | 0.8922 | 0.8976 | 1096.0 | 1221.0 | 0.9056 | 0.9012 | 0.9033 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
| 3.3094 | 10.0 | 70 | 3.5646 | 0.4521 | 0.2422 | 0.3904 | 0.3905 | 29.4 | 0.1533 | 0.2152 | 0.8831 | 0.8894 | 1086.0 | 1221.0 | 0.9043 | 0.9002 | 0.9021 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.15.2 |
|
|
- Transformers 4.53.1 |
|
|
- Pytorch 2.7.0+cu126 |
|
|
- Datasets 3.6.0 |
|
|
- Tokenizers 0.21.1 |
