|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- pszemraj/govreport-summarization-8192 |
|
model-index: |
|
- name: led-base-16384-finetuned-govreport |
|
results: [] |
|
language: |
|
- en |
|
pipeline_tag: summarization |
|
--- |
|
|
|
# led-base-16384-finetuned-govreport |
|
|
|
This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on the [pszemraj/govreport-summarization-8192](https://huggingface.co/datasets/pszemraj/govreport-summarization-8192) dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.2887 |
|
|
|
The amount of processing time and memory required to assess the ROUGE metrics on the validation and test sets were not supported by Kaggle at this moment in time. |
|
|
|
## Model description |
|
|
|
As described in [Longformer: The Long-Document Transformer](https://arxiv.org/pdf/2004.05150.pdf) by Iz Beltagy, Matthew E. Peters, Arman Cohan, [Allenai's Longformer Encoder-Decoder (LED)](https://github.com/allenai/longformer#longformer) was initialized from [*bart-base*](https://huggingface.co/facebook/bart-base) since both models share the exact same architecture. To be able to process 16K tokens, *bart-base*'s position embedding matrix was simply copied 16 times. |
|
|
|
This model is especially interesting for long-range summarization and question answering. |
|
|
|
## Intended uses & limitations |
|
|
|
[pszemraj/govreport-summarization-8192](https://huggingface.co/datasets/pszemraj/govreport-summarization-8192) is a pre-processed version of the dataset [ccdv/govreport-summarization](https://huggingface.co/datasets/ccdv/govreport-summarization), which is a dataset for summarization of long documents adapted from this [repository](https://github.com/luyang-huang96/LongDocSum) and this [paper](https://arxiv.org/pdf/2104.02112.pdf). |
|
|
|
The Allenai's LED model was fine-tuned to this dataset, allowing the summarization of documents up to 16384 tokens. |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 2 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.1492 | 0.24 | 250 | 1.4233 | |
|
| 1.0077 | 0.49 | 500 | 1.3813 | |
|
| 1.0069 | 0.73 | 750 | 1.3499 | |
|
| 0.9639 | 0.98 | 1000 | 1.3216 | |
|
| 0.7996 | 1.22 | 1250 | 1.3172 | |
|
| 0.9395 | 1.46 | 1500 | 1.3003 | |
|
| 0.913 | 1.71 | 1750 | 1.2919 | |
|
| 0.8843 | 1.95 | 2000 | 1.2887 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.30.2 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.13.3 |
