Skip to content

Instantly share code, notes, and snippets.

View jaeoh2's full-sized avatar
๐Ÿ˜ƒ

Jaeoh Lee jaeoh2

๐Ÿ˜ƒ
View GitHub Profile
@sbarratt
sbarratt / geo.py
Last active January 26, 2024 12:55
This script provides coordinate transformations between geodetic, ecef and enu in python. Based on https://gist.github.com/govert/1b373696c9a27ff4c72a.
"""
This script provides coordinate transformations from Geodetic -> ECEF, ECEF -> ENU
and Geodetic -> ENU (the composition of the two previous functions). Running the script
by itself runs tests.
based on https://gist.github.com/govert/1b373696c9a27ff4c72a.
"""
import math
a = 6378137
b = 6356752.3142
@melgor
melgor / linknet_tf.py
Created August 10, 2017 10:26
LinkNet implemenation in TensorFlow
import tensorflow as tf
from tensorflow.contrib.layers.python.layers import initializers
slim = tf.contrib.slim
'''
============================================================================
LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation
============================================================================
Based on the paper: https://arxiv.org/pdf/1707.03718.pdf
'''
@karpathy
karpathy / pg-pong.py
Created May 30, 2016 22:50
Training a Neural Network ATARI Pong agent with Policy Gradients from raw pixels
""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym
# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward
@haje01
haje01 / ๋„์ปค์™€ AWS๋ฅผ ํ™œ์šฉํ•œ ํด๋ผ์šฐ๋“œ ๋”ฅ๋Ÿฌ๋‹ ํ™˜๊ฒฝ ๊ตฌ์ถ•.md
Last active December 20, 2020 08:56
๋„์ปค์™€ AWS๋ฅผ ํ™œ์šฉํ•œ ํด๋ผ์šฐ๋“œ ๋”ฅ๋Ÿฌ๋‹ ํ™˜๊ฒฝ ๊ตฌ์ถ•

๋„์ปค์™€ AWS๋ฅผ ํ™œ์šฉํ•œ ํด๋ผ์šฐ๋“œ ๋”ฅ๋Ÿฌ๋‹ ํ™˜๊ฒฝ ๊ตฌ์ถ•

๊ธ€์“ด์ด: ๊น€์ •์ฃผ(haje01@gmail.com)

์ตœ๊ทผ ๋”ฅ๋Ÿฌ๋‹ ๊ด€๋ จ ํŒจํ‚ค์ง€๋“ค์€ ๋Œ€๋ถ€๋ถ„ CPU์™€ GPU๋ฅผ ํ•จ๊ป˜ ์ง€์›ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. GPU๋ฅผ ์‚ฌ์šฉํ•˜๋ฉด ๋ณด๋‹ค ๋น ๋ฅด๊ฒŒ ํ•™์Šต ๊ฒฐ๊ณผ๋ฅผ ๋‚ผ ์ˆ˜ ์žˆ์ง€๋งŒ, GPU๋ฅผ ํ™œ์šฉํ•˜๊ธฐ ์œ„ํ•ด์„œ๋Š” NVIDIA๊ณ„์—ด์˜ ๊ทธ๋ž˜ํ”ฝ ์นด๋“œ, ๋“œ๋ผ์ด๋ฒ„ S/W ๊ทธ๋ฆฌ๊ณ  CUDA์˜ ์„ค์น˜๋ฅผ ํ•„์š”๋กœ ํ•ฉ๋‹ˆ๋‹ค.

์ด ๊ธ€์—์„œ๋Š” AWS์˜ GPU ์ธ์Šคํ„ด์Šค์™€ ๋„์ปค๋ฅผ ํ™œ์šฉํ•ด ๋”ฅ๋Ÿฌ๋‹ ํŒจํ‚ค์ง€(Caffe)๋ฅผ ํŽธ๋ฆฌํ•˜๊ฒŒ ์‚ฌ์šฉํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์†Œ๊ฐœํ•ฉ๋‹ˆ๋‹ค.


@baraldilorenzo
baraldilorenzo / readme.md
Last active September 13, 2025 12:17
VGG-16 pre-trained model for Keras

##VGG16 model for Keras

This is the Keras model of the 16-layer network used by the VGG team in the ILSVRC-2014 competition.

It has been obtained by directly converting the Caffe model provived by the authors.

Details about the network architecture can be found in the following arXiv paper:

Very Deep Convolutional Networks for Large-Scale Image Recognition

K. Simonyan, A. Zisserman

@thearn
thearn / capture.py
Created June 15, 2013 21:31
Screen capture using python PIL
import time
import Image
import ImageGrab
while True:
time.sleep(0.5)
t = str(time.time()).replace('.','-')
tt = time.time()
img=ImageGrab.grab()
img = img.resize((800,600))