tigerneil

in elasticsearch.yml add the
```
analyzer:
  ik:
        alias:
        - ik_analyzer
        type: org.elasticsearch.index.analysis.IkAnalyzerProvider
      ik_max_word:
        type: ik
        use_smart: false

tigerneil

create two directory: `/config/mmseg`, `/plugins/analysis-mmseg`

1. got to [`https://github.com/medcl/elasticsearch-rtf/tree/master/config/mmseg`](https://github.com/medcl/elasticsearch-rtf/tree/master/config/mmseg)`, download the files: 
  chars.dic, units.dic, words-my.dic,words.dic, mv them to `/config/mmseg`
  
2. go to [`https://github.com/medcl/elasticsearch-rtf/tree/master/plugins/analysis-mmseg`](https://github.com/medcl/elasticsearch-rtf/tree/master/plugins/analysis-mmseg), download the jar:
  elasticsearch-analysis-mmseg-1.2.2.jar, mv them to `/plugins/analysis-mmseg`
  
3. add the following configuration to `elasticsearch.yml`:

tigerneil

"""
This is a batched LSTM forward and backward pass
"""
import numpy as np
import code

class LSTM:
  
  @staticmethod
  def init(input_size, hidden_size, fancy_forget_bias_init = 3):

tigerneil

import theano
from pylearn2.models import mlp
from pylearn2.training_algorithms import sgd
from pylearn2.termination_criteria import EpochCounter
from pylearn2.datasets.dense_design_matrix import DenseDesignMatrix
import numpy as np
from random import randint


class XOR(DenseDesignMatrix):

tigerneil

"""
Simple implementation of Identity Recurrent Neural Networks (IRNN)

Reference
    A Simple Way to Initialize Recurrent Networks of Rectified Linear Units
    http://arxiv.org/abs/1504.00941

"""

import numpy as np

tigerneil

"""
This is a batched LSTM forward and backward pass
"""
import numpy as np
import code

class LSTM:
  
  @staticmethod
  def init(input_size, hidden_size, fancy_forget_bias_init = 3):

tigerneil

Prioritized Experience Replay

Experience replay lets online reinforcement learning agents remember and reuse experiences from the past. In prior work, experience transitions were uniformly sampled from a replay memory. However, this approach simply replays transitions at the same frequency that they were originally experienced, regardless of their significance. In this paper we develop a framework for prioritizing experience, so as to replay important transitions more frequently, and therefore learn more efficiently. We use prioritized experience replay in Deep Q-Networks (DQN), a reinforcement learning algorithm that achieved human-level performance across many Atari games. DQN with prioritized experience replay achieves a new state-of-the-art, outperforming DQN with uniform replay on 42 out of 57 games.

Authors: Tom Schaul schaul@gmail.com, John Quan johnquan@google.com, Ioannis Antonoglou ioannisa@google.com, David Silver davidsilver@google.com

Recurrent Reinforcement Learning: A Hybrid Ap

tigerneil

"""Information Retrieval metrics

Useful Resources:
http://www.cs.utexas.edu/~mooney/ir-course/slides/Evaluation.ppt
http://www.nii.ac.jp/TechReports/05-014E.pdf
http://www.stanford.edu/class/cs276/handouts/EvaluationNew-handout-6-per.pdf
http://hal.archives-ouvertes.fr/docs/00/72/67/60/PDF/07-busa-fekete.pdf
Learning to Rank for Information Retrieval (Tie-Yan Liu)
"""
import numpy as np

tigerneil

from __future__ import print_function
from __future__ import absolute_import
import os

os.environ['THEANO_FLAGS'] = 'device=cpu,mode=FAST_COMPILE,optimizer=None'

from rllab.algos.ddpg import DDPG
from rllab.envs.box2d.cartpole_env import CartpoleEnv
from rllab.policies.deterministic_mlp_policy import DeterministicMLPPolicy
from rllab.q_functions.continuous_mlp_q_function import ContinuousMLPQFunction

tigerneil

Code for Keras plays catch blog post

Train

python qlearn.py

Test

1. Generate figures