echarlaix's picture
echarlaix HF staff
Remove pruning plot.
23103fd
metadata
language: en
license: apache-2.0
tags:
  - summarization
datasets:
  - cnn_dailymail
metrics:
  - R1
  - R2
  - RL

facebook/bart-base model fine-tuned on CNN/DailyMail

This model was created using the nn_pruning python library: the linear layers contains 23% of the original weights.

The model contains 45% of the original weights overall (the embeddings account for a significant part of the model, and they are not pruned by this method).

Fine-Pruning details

This model was fine-tuned from the HuggingFace model. A side-effect of block pruning is that some of the attention heads are completely removed: 61 heads were removed on a total of 216 (28.2%).

Details of the CNN/DailyMail dataset

Dataset Split # samples
CNN/DailyMail train 287K
CNN/DailyMail eval 13K

Results

Metric # Value
Rouge 1 41.43
Rouge 2 18.72
Rouge L 38.35