Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

Last update: Dec 05, 2022

Overview

Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

This is the Pytorch implementation for sparse progressive distillation (SPD). For more details about the motivation, techniques and experimental results, refer to our paper here.

Running

Environment Preparation (using python3)
```
pip install -r requirements.txt
```
Dataset Preparation

The original GLUE dataset could be downloaded here.

BERT_base fine-tuning on GLUE

We use finetuned BERT_base as the teacher. For each task of GLUE benchmark, we obtain the finetuned model using the original huggingface transformers code with the following script.

python run_glue.py \
          --model_name_or_path $INT_DIR \
          --task_name $TASK_NAME \
          --do_train \
          --do_eval \
          --data_dir $GLUE_DIR/$TASK_NAME/ \
          --max_seq_length 128 \
          --per_gpu_train_batch_size 32 \
          --per_gpu_eval_batch_size 32 \
          --learning_rate 3e-5 \
          --num_train_epochs 4.0 \
          --output_dir $OUT_DIR \
          --evaluate_during_training \
          --overwrite_output_dir \
          --logging_steps 400 \
          --logging_dir $OUT_DIR \
          --save_steps 10000

Sparse Progressive Distillation

We use run_glue.py to run the sparse progressive distillation. --num_prune_epochs is the epochs for pruning. --num_train_epochs is the total number of epochs (pruning, progressive distillation, finetuning).

python run_glue.py \
  --model_name_or_path PATH_TO_FINETUNED_MODEL \
  --task_name $TASK_NAME \
  --do_train \
  --do_eval \
  --do_lower_case \
  --data_dir $GLUE_DIR/$TASK_NAME/ \
  --max_seq_length 128 \
  --per_gpu_train_batch_size 32 \
  --per_gpu_eval_batch_size 32 \
  --learning_rate 6.4e-4 \
  --save_steps 50 \
  --num_prune_epochs 30 \
  --num_train_epochs 60 \
  --sparsity 0.9 \
  --output_dir $OUT_DIR \
  --evaluate_during_training \
  --replacing_rate 0.8 \
  --overwrite_output_dir \
  --steps_for_replacing 0 \
  --scheduler_type linear

To Dos

Provide our teacher model for each task.
Provide best performed model checkpoint for each task.

Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

Related tags

Overview

Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

Running

BERT_base fine-tuning on GLUE

Sparse Progressive Distillation

To Dos

Owner

High-resolution networks and Segmentation Transformer for Semantic Segmentation

Code to train models from "Paraphrastic Representations at Scale".

Generalized Decision Transformer for Offline Hindsight Information Matching

Problem-943.-ACMP - Problem 943. ACMP

The mini-MusicNet dataset

HarDNeXt: Official HarDNeXt repository

PyTorch implementation of CloudWalk's recent work DenseBody

Tensorflow implementation of "BEGAN: Boundary Equilibrium Generative Adversarial Networks"

Tensorflow implementation of Swin Transformer model.

Official repository for Natural Image Matting via Guided Contextual Attention

World Models with TensorFlow 2

Decoding the Protein-ligand Interactions Using Parallel Graph Neural Networks

PyTorch implementation of "A Simple Baseline for Low-Budget Active Learning".

[arXiv22] Disentangled Representation Learning for Text-Video Retrieval

A Python framework for developing parallelized Computational Fluid Dynamics software to solve the hyperbolic 2D Euler equations on distributed, multi-block structured grids.

This is an official implementation for "SimMIM: A Simple Framework for Masked Image Modeling".

A Python package for time series augmentation

Official implementation of the PICASO: Permutation-Invariant Cascaded Attentional Set Operator

Prototypical python implementation of the trust-region algorithm presented in Sequential Linearization Method for Bound-Constrained Mathematical Programs with Complementarity Constraints by Larson, Leyffer, Kirches, and Manns.

Activity tragle - Google is tracking everything, we just look at it