Skip to content

Instantly share code, notes, and snippets.

Eugenio Culurciello culurciello

Block or report user

Report or block culurciello

Hide content and notifications from this user.

Learn more about blocking users

Contact Support about this user’s behavior.

Learn more about reporting abuse

Report abuse
View GitHub Profile
View pytorchcheat.txt
http://pytorch.org/
https://discuss.pytorch.org/
https://github.com/pytorch/
https://github.com/bfortuner/pytorch-cheatsheet/blob/master/pytorch-cheatsheet.ipynb
View pytorch failure Titan X (Maxwell)
elab@gpu5 ~/pytorch-examples/imagenet [master*]$ python3 main.py -a alexnet /media/SuperSSD/test-dataset-train-val/
=> creating model 'alexnet'
Traceback (most recent call last):
File "main.py", line 286, in <module>
main()
File "main.py", line 129, in main
train(train_loader, model, criterion, optimizer, epoch)
File "main.py", line 165, in train
output = model(input_var)
File "/usr/local/lib/python3.4/dist-packages/torch/nn/modules/module.py", line 210, in __call__
View table-test.lua
-- E. Culurciello, November 2016
-- test to see if table insert/remove is slower than table addressing
iters = 1e5
tablesize = 1e4
tab1={}
item = torch.randn(iters,1024)
sys.tic()
View xception.lua
-- Xception model
-- a Torch7 implementation of: https://arxiv.org/abs/1610.02357
-- E. Culurciello, October 2016
require 'nn'
local nClasses = 1000
function nn.SpatialSeparableConvolution(nInputPlane, nOutputPlane, kW, kH)
local block = nn.Sequential()
block:add(nn.SpatialConvolutionMap(nn.tables.oneToOne(nInputPlane), kW,kH, 1,1, 1,1))
View RL-example.py
#! /usr/bin/env python3
# E. Culurciello, example of reinforcement learning in Python
#
# http://mnemstudio.org/path-finding-q-learning-tutorial.htm
# Game: 5 rooms connected through doors. One room is the goal-room.
# the goal of the game is to get to the goal-room
import numpy as np
# this is how rooms are connected:
View recurrent-multilayer.lua
-- Eugenio Culurciello
-- August 2016
-- a test to learn to code PredNet-like nets in Torch7
require 'nn'
require 'nngraph'
torch.setdefaulttensortype('torch.FloatTensor')
nngraph.setDebug(true)
View LineSegmentCetector.lua
local cv = require 'cv'
require 'cv.highgui'
require 'cv.imgproc'
require 'cv.imgcodecs'
require 'image'
-- local image = cv.imread{arg[1] or 'demo/lena.jpg', cv.IMREAD_GRAYSCALE}
imgT = image.lena()
imgT = image.lena()
imgTg = imgT[2] -- convert to grayscale and remove the first dimension
View ffmpeg.rtf
to resize 1080p videos:
ffmpeg -i pool.mp4 -vf scale=640:360 pool-small.mp4
to match odroid fps (12fps): need to speed up video:
https://trac.ffmpeg.org/wiki/How%20to%20speed%20up%20/%20slow%20down%20a%20video
1:
ffmpeg -i dog.mp4 -vf scale=640:360 -filter:v "setpts=0.5*PTS" dog-2x.mp4
because it does not scale with this command, so we need to do more:
2:
You can’t perform that action at this time.