Tarot Learner

Introduction

Front Player

Description

API

Description

Requirements

apt-get install golang

# export GOPATH

go get -u github.com/gorilla/mux

go run rest.go

API Doc

# To gen api doc, go inside rest.go dir
# https://github.com/apidoc/apidoc/tree/dev
apidoc -i API/src/rest/ -o API/docs/

# Display doc
google-chrome API/docs/index.html

Front Learner

SVM

It takes 2 years with 144000 tricks

Random Forest

Faster, but cannot load in RAM...

### Q Learning

Description

Go further with doInteraction()

Experiment Env (EE)
Experiment Task (ET)
Q Learner (QL)
Q Agent (QA)

Observations :

QL --getObservation--> ET --getSensors--> EE
QL --integrateObservation--> QA

Action :

QL --getAction--> QA
QL --performAction--> ET --performAction--> EE

Reward :

QL --getReward()--> ET (--reset--> EE)
QL --giveReward --> QA

Add PyBrain repository

sudo pip install --upgrade pybrain

# Apply fix https://github.com/pybrain/pybrain/issues/211
# Files are present in pybrain_fix/

http://cs229.stanford.edu/proj2014/Ivan%20Leung,%20Pedro%20Milani,%20Ben-han%20Sung,%20SkatBot.pdf