|
--- |
|
license: apache-2.0 |
|
base_model: google/long-t5-tglobal-base |
|
tags: |
|
- generated_from_trainer |
|
- synthsumm |
|
metrics: |
|
- rouge |
|
datasets: |
|
- pszemraj/synthsumm |
|
language: |
|
- en |
|
pipeline_tag: summarization |
|
inference: |
|
parameters: |
|
max_length: 64 |
|
min_length: 8 |
|
no_repeat_ngram_size: 3 |
|
early_stopping: true |
|
repetition_penalty: 3.5 |
|
encoder_no_repeat_ngram_size: 4 |
|
num_beams: 3 |
|
--- |
|
|
|
# long-t5-tglobal-base-synthsumm_direct |
|
|
|
Fine-tuned on a synthetic dataset of curated long-context text and `GPT-3.5-turbo-1106` summaries spanning multiple domains + "random" long-context examples from pretraining datasets |
|
|
|
|
|
- Note: this model has **not** been fine-tuned on any other summarization datasets, just the `synthsumm` data |
|
|
|
Try it: [gradio demo](https://huggingface.co/spaces/pszemraj/document-summarization) | free [HF inference api](https://gist.github.com/pszemraj/08f527380ed00ef2f2169e220341c489) via `requests`| [.md with example outputs](evals-outputs/GAUNTLET.md) (gauntlet) |
|
|
|
## Usage |
|
|
|
It's recommended to use this model with [beam search decoding](https://huggingface.co/docs/transformers/generation_strategies#beamsearch-decoding). If interested, you can also use the `textsum` [util repo](https://github.com/pszemraj/textsum) to have most of this abstracted out for you: |
|
|
|
```bash |
|
pip install -U textsum |
|
``` |
|
|
|
```python |
|
from textsum.summarize import Summarizer |
|
|
|
model_name = "pszemraj/long-t5-tglobal-base-synthsumm_direct" |
|
summarizer = Summarizer(model_name) # GPU auto-detected |
|
text = "put the text you don't want to read here" |
|
summary = summarizer.summarize_string(text) |
|
print(summary) |
|
``` |
|
|
|
## Details |
|
|
|
This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.4378 |
|
- Rouge1: 48.0918 |
|
- Rouge2: 21.2531 |
|
- Rougel: 34.4307 |
|
- Rougelsum: 43.0271 |
|
- Gen Len: 84.5231 |
|
|
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0003 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 1 |
|
- seed: 26605 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: inverse_sqrt |
|
- lr_scheduler_warmup_ratio: 0.03 |
|
- num_epochs: 2.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:--------:| |
|
| 1.9183 | 0.38 | 125 | 1.5762 | 38.7221 | 15.0873 | 28.3123 | 34.9655 | 129.2154 | |
|
| 1.8815 | 0.77 | 250 | 1.5230 | 44.3531 | 17.9384 | 31.7417 | 39.5563 | 87.3538 | |
|
| 1.7264 | 1.15 | 375 | 1.4735 | 45.7781 | 20.102 | 33.329 | 41.4737 | 101.9231 | |
|
| 1.8545 | 1.54 | 500 | 1.4505 | 47.0134 | 20.6159 | 33.6118 | 41.6579 | 88.2308 | |
|
| 1.7444 | 1.92 | 625 | 1.4378 | 48.0918 | 21.2531 | 34.4307 | 43.0271 | 84.5231 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.0 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
