PyTorch source code for Distilling Knowledge by Mimicking Features

Overview

LSHFM.detection

This is the PyTorch source code for Distilling Knowledge by Mimicking Features. And this project contains code for object detection with mimicking features. For image classification, please visit LSHFM.classification.

dependence

  • python
  • pytorch 1.7.1
  • torchvision 0.8.2

Prepare the dataset

Please prepare the COCO and VOC datasets by youself. Then you need to fix the get_data_path function in src/dataset/coco_utils.py and src/dataset/voc_utils.py.

Run

You can run the experiments by

PORT=4444 bash experiments/[script name].sh 0,1,2,3 

the training set contains VOC2007 trainval and VOC2012 trainval, while the testing set is VOC2007 test.

We train all models by 24 epochs while the learning rate decays at the 18th and 22th epoch.

Faster R-CNN

Before you run the KD experiments, please make sure the teacher model weight have been saved in pretrained. You can first run ResNet101 baseline and VGG16 baseline to train the teacher model, and then move the model to pretrained and edit --teacher-ckpt in the training shell scripts. You can also download voc0712_fasterrcnn_r101_83.6 and voc0712_fasterrcnn_vgg16fpn_79.0 directly, and move them to pretrained.

[email protected] [email protected]
Teacher 83.6 79.0
Student 82.0 75.1
L2 83.0 76.8
LSH 82.6 76.7
LSHL2 83.0 77.2

RetinaNet

As mentioned in Faster R-CNN, please make sure there are teacher models in pretrained. You can download the teacher models in voc0712_retinanet_r101_83.0.ckpt and voc0712_retinanet_vgg16fpn_76.6.ckpt.

[email protected] [email protected]
Teacher 83.0 76.6
Student 82.5 73.2
L2 82.6 74.8
LSHL2 83.0 75.2

We find that it is easy to get NaN loss when training by LSH KD.

visualize

visualize the ground truth label

python src/visual.py --dataset voc07 --idx 1 --gt

visualize the model prediction

python src/visual.py --dataset voc07 --idx 2 --model fasterrcnn_resnet50_fpn --checkpoint results/voc0712/fasterrcnn_resnet50_fpn/2020-12-11_20\:14\:09/model_13.pth

Citing this repository

If you find this code useful in your research, please consider citing us:

@article{LSHFM,
  title={Distilling knowledge by mimicking features},
  author={Wang, Guo-Hua and Ge, Yifan and Wu, Jianxin},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2021},
}

Acknowledgement

This project is based on https://github.com/pytorch/vision/tree/master/references/detection. This project aims at object detection, so I remove the code about segmentation and keypoint detection.

Owner
Guo-Hua Wang
Guo-Hua Wang
CTRL-C: Camera calibration TRansformer with Line-Classification

CTRL-C: Camera calibration TRansformer with Line-Classification This repository contains the official code and pretrained models for CTRL-C (Camera ca

57 Nov 14, 2022
Predictive AI layer for existing databases.

MindsDB is an open-source AI layer for existing databases that allows you to effortlessly develop, train and deploy state-of-the-art machine learning

MindsDB Inc 12.2k Jan 03, 2023
Implementation of the pix2pix model on satellite images

This repo shows how to implement and use the pix2pix GAN model for image to image translation. The model is demonstrated on satellite images, and the

3 May 24, 2022
Pytorch implementation of "Neural Wireframe Renderer: Learning Wireframe to Image Translations"

Neural Wireframe Renderer: Learning Wireframe to Image Translations Pytorch implementation of ideas from the paper Neural Wireframe Renderer: Learning

Yuan Xue 7 Nov 14, 2022
Source Code for our paper: Understand me, if you refer to Aspect Knowledge: Knowledge-aware Gated Recurrent Memory Network

KaGRMN-DSG_ABSA This repository contains the PyTorch source Code for our paper: Understand me, if you refer to Aspect Knowledge: Knowledge-aware Gated

XingBowen 4 May 20, 2022
OpenIPDM is a MATLAB open-source platform that stands for infrastructures probabilistic deterioration model

Open-Source Toolbox for Infrastructures Probabilistic Deterioration Modelling OpenIPDM is a MATLAB open-source platform that stands for infrastructure

CIVML 0 Jan 20, 2022
Post-Training Quantization for Vision transformers.

PTQ4ViT Post-Training Quantization Framework for Vision Transformers. We use the twin uniform quantization method to reduce the quantization error on

Zhihang Yuan 61 Dec 28, 2022
City-seeds - A random generator of cultural characteristics intended to spark ideas and help draw threads

City Seeds This is a random generator of cultural characteristics intended to sp

Aydin O'Leary 2 Mar 12, 2022
In this project I played with mlflow, streamlit and fastapi to create a training and prediction app on digits

Fastapi + MLflow + streamlit Setup env. I hope I covered all. pip install -r requirements.txt Start app Go in the root dir and run these Streamlit str

76 Nov 23, 2022
Implementation of the master's thesis "Temporal copying and local hallucination for video inpainting".

Temporal copying and local hallucination for video inpainting This repository contains the implementation of my master's thesis "Temporal copying and

David Álvarez de la Torre 1 Dec 02, 2022
DL course co-developed by YSDA, HSE and Skoltech

Deep learning course This repo supplements Deep Learning course taught at YSDA and HSE @fall'21. For previous iteration visit the spring21 branch. Lec

Yandex School of Data Analysis 1.3k Dec 30, 2022
JupyterNotebook - C/C++, Javascript, HTML, LaTex, Shell scripts in Jupyter Notebook Also run them on remote computer

JupyterNotebook Read, write and execute C, C++, Javascript, Shell scripts, HTML, LaTex in jupyter notebook, And also execute them on remote computer R

1 Jan 09, 2022
Python Assignments for the Deep Learning lectures by Andrew NG on coursera with complete submission for grading capability.

Python Assignments for the Deep Learning lectures by Andrew NG on coursera with complete submission for grading capability.

Utkarsh Agiwal 1 Feb 03, 2022
Viperdb - A tiny log-structured key-value database written in pure Python

ViperDB 🐍 ViperDB is a lightweight embedded key-value store written in pure Pyt

17 Oct 17, 2022
Code for the paper: Adversarial Training Against Location-Optimized Adversarial Patches. ECCV-W 2020.

Adversarial Training Against Location-Optimized Adversarial Patches arXiv | Paper | Code | Video | Slides Code for the paper: Sukrut Rao, David Stutz,

Sukrut Rao 32 Dec 13, 2022
Universal Adversarial Examples in Remote Sensing: Methodology and Benchmark

Universal Adversarial Examples in Remote Sensing: Methodology and Benchmark Yong

19 Dec 17, 2022
iPOKE: Poking a Still Image for Controlled Stochastic Video Synthesis

iPOKE: Poking a Still Image for Controlled Stochastic Video Synthesis iPOKE: Poking a Still Image for Controlled Stochastic Video Synthesis Andreas Bl

CompVis Heidelberg 36 Dec 25, 2022
Code for NeurIPS 2021 paper: Invariant Causal Imitation Learning for Generalizable Policies

Invariant Causal Imitation Learning for Generalizable Policies Ioana Bica, Daniel Jarrett, Mihaela van der Schaar Neural Information Processing System

Ioana Bica 17 Dec 01, 2022
Self-supervised spatio-spectro-temporal represenation learning for EEG analysis

EEG-Oriented Self-Supervised Learning and Cluster-Aware Adaptation This repository provides a tensorflow implementation of a submitted paper: EEG-Orie

Wonjun Ko 4 Jun 09, 2022
Wider or Deeper: Revisiting the ResNet Model for Visual Recognition

ademxapp Visual applications by the University of Adelaide In designing our Model A, we did not over-optimize its structure for efficiency unless it w

Zifeng Wu 338 Dec 12, 2022