Skip to content

Instantly share code, notes, and snippets.

View ryogrid's full-sized avatar

Ryo Kanbayashi ryogrid

View GitHub Profile
@ryogrid
ryogrid / calm_dji.csv
Created March 25, 2017 02:51
validation of "Twitter mood predicts the stock market"
2009/1/1 0 0
2009/1/2 0 9034.69043
2009/1/3 0 0
2009/1/4 0 0
2009/1/5 0 8952.889648
2009/1/6 0 9015.099609
2009/1/7 0 8769.700195
2009/1/8 0 8742.459961
2009/1/9 0 8599.179688
2009/1/10 0 0
from __future__ import print_function
import gym
import numpy as np
import itertools
import os
from neat import nn, population, statistics
np.set_printoptions(threshold=np.inf)
@ryogrid
ryogrid / readme.txt
Last active February 22, 2017 01:31
agent using NEAT method
above codes emulated https://gist.github.com/stefanopalmieri/2efec0e09c14de06fb93fbb91e18a93a
[Run]
git clone https://github.com/CodeReclaimers/neat-python.git
cd neat-python
sudo python setup.py install
# at dir contains script and xor2.config
python walker_neat.py
@ryogrid
ryogrid / AI_gym_DQN.py
Created December 12, 2016 15:53
DQN algorithm for BipedalWalkerHardcore-v2 on Open AI Gym
# coding: utf-8
import numpy as np
import time
import chainer
from chainer import cuda, Function, gradient_check, Variable, optimizers, serializers, utils
from chainer import Link, Chain, ChainList
import chainer.functions as F
import chainer.links as L