nagataka

import java.net.URI;
import java.net.URISyntaxException;
import java.sql.*;

public class PrestoJDBC {
   // JDBC driver name and database URL
   static final String JDBC_DRIVER = "com.facebook.presto.jdbc.PrestoDriver";
   //static final String JDBC_DRIVER = "com.teradata.presto.jdbc42.Driver";

   static final String DB_URL = "jdbc:presto://ec2-xx-xx-xxx-xxx.ap-northeast-1.compute.amazonaws.com:8889/hive/default";

nagataka

import random

class EpsilonGreedy():
    def __init__(self, epsilon, counts, values):
        self.epsilon = epsilon # probability of explore
        self.counts = counts # number of pulls for each arms
        self.values = values # average amount of reward we've gotten from each arms
        return

    def initialize(self, n_arms):

nagataka

たまに色々と忘れるので便利な資料をメモ

• How to Read Figures, Mathematical Expressions and Equations, and Glossary
• 英語による数学表現

nagataka

• ブログなど
  • 日本語なら 今更聞けないLSTMの基本 がとても参考になる。
  • In English: Understanding LSTM Networks
• Implementation
  • Docs > torch.nn > LSTM
  • SEQUENCE MODELS AND LONG-SHORT TERM MEMORY NETWORKS

その他、未読だけれど参考になりそうなリンク (随時読んだら上へ)

• The Unreasonable Effectiveness of Recurrent Neural Networks

nagataka

# OpenAI gym custom environment mimicking Blocking Maze
# See Sutton and Barto "Reinforcement Learning an Introduction"
#     Example 8.2: Blocking Maze
from enum import Enum
import sys
import copy

import gym
from gym import error, spaces, utils
from gym.utils import seeding

nagataka

{
    "python.formatting.provider": "black",
    "python.linting.pylintEnabled": false,
    "python.linting.flake8Enabled": true,
    "python.linting.flake8Args": [
        "--ignore=E501,W503"
    ],
    "python.sortImports.args": [
        "-m 3"
    ],

nagataka

import random
import numpy as np
np.random.seed(0)

def kerri(p, b):
    """https://en.wikipedia.org/wiki/Kelly_criterion
    """
    return (p*(b+1)-1 )/b

N = 300

nagataka

import gym
import ray
from ray.rllib.agents.ppo import PPOTrainer, DEFAULT_CONFIG

import pprint as pp

#tune.run(PPOTrainer, config={"env": "Breakout-v0", "use_pytorch": True})
ray.init(num_gpus=1, ignore_reinit_error=True, log_to_driver=False)

# https://github.com/ray-project/ray/blob/master/rllib/agents/ppo/ppo.py#L15

nagataka

Dataset for Reinforcement Learning

• DQN Replay Dataset by Google AI
  • context: An Optimistic Perspective on Offline Reinforcement Learning

labeled video datasets

Human activity

• ActivityNet

nagataka

"""A template to implement RL agent with OpenAI Gym

Usage: python ./gym_template.py --env=CarRacing-v0 --algo=policy_gradient --epochs 1

implementation of algorithms need to be ./algorithms/ directory, or change the following line to your env
> algo = import_module('algorithms.'+args.algo)
"""
import argparse
import numpy as np