quantization

Step-by-step

This directory contains examples for finetuning and evaluating transformers on summarization tasks.

run_summarization.py is a lightweight example of how to download and preprocess a dataset from the 🤗 Datasets library or use your own files (jsonlines or csv), then fine-tune one of the architectures above on it.

Prerequisite​

1. Create Environment

pip install intel-intel-for-transformers
pip install -r requirements.txt
# if run pegasus-samsum, need to downgrade the protobuf package to 3.20.x or lower.
pip install protobuf==3.20

Run

1. Quantization

For PyTorch, Here is an example on a summarization task:

python run_summarization.py \
    --model_name_or_path stacked-summaries/flan-t5-large-stacked-samsum-1024 \
    --dataset_name samsum \
    --do_train \
    --do_eval \
    --output_dir /tmp/tst-summarization \
    --per_device_train_batch_size=8 \
    --per_device_eval_batch_size=8 \
    --overwrite_output_dir \
    --tune \
    --predict_with_generate \
    --perf_tol 0.03

T5 model t5-base t5-large must use an additional argument: --source_prefix "summarize: ".

We used CNN/DailyMail dataset in this example as t5-small was trained on it and one can get good scores even when pre-training with a very small sample.

Extreme Summarization (XSum) Dataset is another commonly used dataset for the task of summarization. To use it replace --dataset_name cnn_dailymail --dataset_config "3.0.0" with --dataset_name xsum.

And here is how you would use it on your own files, after adjusting the values for the arguments --train_file, --validation_file, --text_column and --summary_column to match your setup:

python examples/pytorch/summarization/run_summarization.py \
    --model_name_or_path stacked-summaries/flan-t5-large-stacked-samsum-1024 \
    --dataset_name samsum \
    --do_train \
    --do_eval \
    --train_file path_to_csv_or_jsonlines_file \
    --validation_file path_to_csv_or_jsonlines_file \
    --output_dir /tmp/tst-summarization \
    --overwrite_output_dir \
    --per_device_train_batch_size=8 \
    --per_device_eval_batch_size=8 \
    --tune \
    --predict_with_generate \
    --perf_tol 0.03

2. Validated Model List

Dataset	Pretrained model	PostTrainingDynamic	PostTrainingStatic	QuantizationAwareTraining
samsum	pegasus_samsum	✅	N/A	N/A
cnn_dailymail	t5_base_cnn	✅	N/A	N/A
cnn_dailymail	t5_large_cnn	✅	N/A	N/A
samsum	flan_t5_large_samsum	✅	✅	N/A