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

FishNet: One Stage to Detect, Segmentation and Pose Estimation

Repository of the paper Compressing Sensor Data for Remote Assistance of Autonomous Vehicles using Deep Generative Models at ML4AD @ NeurIPS 2021.

Unofficial Implementation of MLP-Mixer, gMLP, resMLP, Vision Permutator, S2MLPv2, RaftMLP, ConvMLP, ConvMixer in Jittor and PyTorch.

Yoloxkeypointsegment - An anchor-free version of YOLO, with a simpler design but better performance

💡 Type hints for Numpy

Linear algebra python - Number of operations and problems in Linear Algebra and Numerical Linear Algebra

[ICCV 2021] Released code for Causal Attention for Unbiased Visual Recognition

The official homepage of the COCO-Stuff dataset.

SOFT: Softmax-free Transformer with Linear Complexity, NeurIPS 2021 Spotlight

The implemention of Video Depth Estimation by Fusing Flow-to-Depth Proposals

Code for all the Advent of Code'21 challenges mostly written in python

Yolact-keras实例分割模型在keras当中的实现

Asymmetric metric learning for knowledge transfer

An LSTM based GAN for Human motion synthesis

A deep learning based semantic search platform that computes similarity scores between provided query and documents

Video Swin Transformer - PyTorch

Kalman Filter book using Jupyter Notebook. Focuses on building intuition and experience, not formal proofs. Includes Kalman filters,extended Kalman filters, unscented Kalman filters, particle filters, and more. All exercises include solutions.

Tooling for converting STAC metadata to ODC data model

Bianace Prediction Pytorch Model

Residual Pathway Priors for Soft Equivariance Constraints