tilarids

#include <iostream>
#include <vector>
#include <set>
#include <boost/iterator/counting_iterator.hpp>
#include <algorithm>
#include <functional>

typedef std::vector<int>    int_vec;
typedef std::set<int>       int_set;

tilarids

open Batteries_uni

let regroup e n =
  let next () =
    if Enum.is_empty e then
      raise Enum.No_more_elements
    else
      Enum.take n e in Enum.from next;;

let load_ciphers fname =

tilarids

open Batteries_uni;;

let get_hash x = let s = Cryptokit.Hash.sha256 () in
                 begin
                   s#add_string x;
                   s#result
                 end;;

let lsb x n =
  let complete_bytes_count = n / 8 in

tilarids

# SRC should be specified
OUTPUT_FILES=$(shell find ${SRC} -type f -exec echo {}.html \;)

all: ${OUTPUT_FILES}
	echo "All is done"

%.md.html : %.md
	python -c "import markdown; md = markdown.Markdown(extensions=['meta']); md.convertFile('$^', '$@')"

tilarids

o  changeset:   1:6c03dec0bcf6
|  tag:         tip
|  user:        Sergey Kishchenko <sergey.kishchenko@quickoffice.com>
|  date:        Mon Oct 08 14:33:43 2012 +0300
|  summary:     file.txt was fixed
|
|  diff -r 6fb9917bb523 -r 6c03dec0bcf6 file.txt
|  --- a/file.txt	Mon Oct 08 13:36:59 2012 +0300
|  +++ b/file.txt	Mon Oct 08 14:33:43 2012 +0300
|  @@ -1,2 +1,2 @@

tilarids

@  changeset:   3:927d19ecf38e
|  tag:         tip
|  user:        Sergey Kishchenko <sergey.kishchenko@quickoffice.com>
|  date:        Mon Oct 08 14:34:17 2012 +0300
|  summary:     file.txt was fixed again
|
|  diff -r f2f4866c5fa1 -r 927d19ecf38e file.txt
|  --- a/file.txt	Mon Oct 08 14:33:43 2012 +0300
|  +++ b/file.txt	Mon Oct 08 14:34:17 2012 +0300
|  @@ -1,3 +1,3 @@

tilarids

/*
gcc camera-shot.c -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include/ -I/usr/include/clutter-gtk-1.0 -I /usr/include/gtk-3.0/ -I /usr/include/pango-1.0 -I /usr/include/cairo/ -I /usr/include/gdk-pixbuf-2.0 -I /usr/include/atk-1.0 -I /usr/include/clutter-1.0 -I /usr/include/cogl -I /usr/include/json-glib-1.0 -I /usr/include/gstreamer-0.10/ -I /usr/include/libxml2 -I /usr/include/cheese -lcheese -lgtk-3 -lclutter-gtk-1.0 -o camera-shot
*/

#define HAVE_DLFCN_H 1
#define HAVE_INTTYPES_H 1
#define HAVE_MEMORY_H 1
#define HAVE_STDINT_H 1
#define HAVE_STDLIB_H 1
#define HAVE_STRINGS_H 1

tilarids

tilarids

Vanilla policy gradients with ValueFunction to estimate value for the specific state (I use current observation, previous observation and previous action as a state). This same algorithm works fine without ValueFunction if you don't stop the learning process at step 200 and continue learning after that. OpenAI Gym's monitor stops the game at step 200 so you can't use monitor at the same time as training on more than 200 steps.

Reproducing:

• Use this commit: https://github.com/tilarids/reinforcement_learning_playground/commit/7faeff3ff0c0397f84f391b75e301ac7dea8e69b
• Consider changing the API key :)
• python pg_agent.py

tilarids

main
main.dSYM/