zhanghang1989

MODEL:
  META_ARCHITECTURE: "Detr"
  WEIGHTS: "detectron2://ImageNetPretrained/torchvision/R-50.pkl"
  PIXEL_MEAN: [123.675, 116.280, 103.530]
  PIXEL_STD: [58.395, 57.120, 57.375]
  MASK_ON: False
  BACKBONE:
    NAME: "build_resnet_backbone"
  RESNETS:
    DEPTH: 50

zhanghang1989

wget https://raw.githubusercontent.com/zhanghang1989/AutoGluon-Tutorial-CVPR2020/master/6.pytorch_hpo.ipynb quick_start.ipynb
wget https://raw.githubusercontent.com/zhanghang1989/AutoGluon-Tutorial-CVPR2020/master/7.enas_mnist.ipynb proxylessnas.ipynb

mkdir -p ILSVRC2012 && cd ILSVRC2012/
wget https://hanlab.mit.edu/files/OnceForAll/ofa_cvpr_tutorial/imagenet_subtrain.tar.gz
tar -xzvf imagenet_subtrain.tar.gz && rm imagenet_subtrain.tar.gz

mkdir -p val && cd val/
wget https://hangzh.s3-us-west-1.amazonaws.com/encoding/legacy/ILSVRC2012_img_val.tar
tar -xf ILSVRC2012_img_val.tar

zhanghang1989

#!/bin/bash
sudo -u ec2-user -i <<'EOF'

cd ~
wget https://autogluon.s3-us-west-2.amazonaws.com/autogluon_tutorial/build_contrib.yml

conda env create --name autogluon -f build_contrib.yml
wget https://autogluon.s3-us-west-2.amazonaws.com/autogluon_tutorial/prefetch.py

source activate autogluon

zhanghang1989

aws s3 cp s3://zhiz-data/datasets/imagenet/ILSVRC2012_img_train.tar .
aws s3 cp s3://zhiz-data/datasets/imagenet/ILSVRC2012_img_val.tar .

cd ResNeSt/
python scripts/dataset/prepare_imagenet.py --download-dir ~/

zhanghang1989

###########################################################################
# Created by: Hang Zhang 
# Email: zhang.hang@rutgers.edu 
# Copyright (c) 2018
###########################################################################
from __future__ import division
import os
import numpy as np
import torch
import torch.nn as nn

zhanghang1989

printf "\e[?2004l"

zhanghang1989

sudo mkfs -t ext4 /dev/nvme1n1 
sudo mkfs -t ext4 /dev/nvme2n1 
mkdir drive1
mkdir drive2
sudo mount /dev/nvme1n1 drive1/
sudo mount /dev/nvme2n1 drive2/
sudo chmod a+rwx  drive1/
sudo chmod a+rwx  drive2/

zhanghang1989

find . -name \*.swp -type f -delete

zhanghang1989

import os
from os import listdir
from os.path import isfile, join
import hashlib
import shutil
import zipfile

mypath = './'
newpath = './zip'

zhanghang1989

PyTorch to MXNet

This cheatsheet serves as a quick reference for PyTorch users who are interested in trying MXNet, and vice versa.

Pytorch is a deep learning framework provides imperative tensor manipulation and neural network training. MXNet provides similar imperative tensor manipulation through the ndarray package and neural network training through gluon. This cheatsheet maps functions one-by-one between these two frameworks.

Note that MXNet has a symbolic interface similar to Keras and Tensorflow that may provide better performance and portability. This cheatsheet mainly focus on MXNet's imperative interface.

Installation