madlag/bert-base-uncased-squadv1-x1.16-f88.1-d8-unstruct-v1

BERT-base uncased model fine-tuned on SQuAD v1

This model was created using the nn_pruning python library: the linear layers contains 8.0% of the original weights.

The model contains 28.0% of the original weights overall (the embeddings account for a significant part of the model, and they are not pruned by this method).

With a simple resizing of the linear matrices it ran 1.16x as fast as bert-base-uncased on the evaluation. This is possible because the pruning method lead to structured matrices: to visualize them, hover below on the plot to see the non-zero/zero parts of each matrix.

In terms of accuracy, its F1 is 88.11, compared with 88.5 for bert-base-uncased, a F1 drop of 0.39.

Fine-Pruning details

This model was fine-tuned from the HuggingFace model checkpoint on SQuAD1.1, and distilled from the model bert-large-uncased-whole-word-masking-finetuned-squad This model is case-insensitive: it does not make a difference between english and English.

A side-effect of the block pruning is that some of the attention heads are completely removed: 22 heads were removed on a total of 144 (15.3%). Here is a detailed view on how the remaining heads are distributed in the network after pruning.

Details of the SQuAD1.1 dataset

Dataset	Split	# samples
SQuAD1.1	train	90.6K
SQuAD1.1	eval	11.1k

Fine-tuning

• Python: 3.8.5

• Machine specs:

Memory: 64 GiB
GPUs: 1 GeForce GTX 3090, with 24GiB memory
GPU driver: 455.23.05, CUDA: 11.1

Results

Pytorch model file size: 398MB (original BERT: 420MB)

Metric	# Value	# Original (Table 2)	Variation
EM	80.94	80.8	+0.14
F1	88.11	88.5	-0.39

Example Usage

Install nn_pruning: it contains the optimization script, which just pack the linear layers into smaller ones by removing empty rows/columns.

pip install nn_pruning

Then you can use the transformers library almost as usual: you just have to call optimize_model when the pipeline has loaded.

from transformers import pipeline
from nn_pruning.inference_model_patcher import optimize_model

qa_pipeline = pipeline(
    "question-answering",
    model="madlag/bert-base-uncased-squadv1-x1.16-f88.1-d8-unstruct-v1",
    tokenizer="madlag/bert-base-uncased-squadv1-x1.16-f88.1-d8-unstruct-v1"
)

print("bert-base-uncased parameters: 152.0M")
print(f"Parameters count (includes only head pruning, not feed forward pruning)={int(qa_pipeline.model.num_parameters() / 1E6)}M")
qa_pipeline.model = optimize_model(qa_pipeline.model, "dense")

print(f"Parameters count after complete optimization={int(qa_pipeline.model.num_parameters() / 1E6)}M")
predictions = qa_pipeline({
    'context': "Frédéric François Chopin, born Fryderyk Franciszek Chopin (1 March 1810 – 17 October 1849), was a Polish composer and virtuoso pianist of the Romantic era who wrote primarily for solo piano.",
    'question': "Who is Frederic Chopin?",
})
print("Predictions", predictions)