Vaughn  Sauer

Vaughn Sauer

1623428700

Train “undying” Flappy Bird using Reinforcement Learning on Java

Flappy Bird is a mobile game that was introduced in 2013 which became super popular because of its simple way to play (flap/no-flap). With the growth of Deep Learning (DL) and Reinforcement Learning (RL), we can now train an AI agent to control the Flappy Bird actions. Today, we will look at the process to create an AI agent using Java. For the game itself, we used a simple open-source Flappy Bird game using Java. For training, we used Deep Java Library (DJL), a deep learning framework based on Java, to build the training network and RL algorithm.

In this article, we will start from the basis of RL and walk through the key components to build the training architecture. If at anytime you cannot follow our code and would like to try the game, you can refer to our RL repo.

The RL Architecture

In this section, we will introduce some major algorithm and networks we used to help you better understand how we trained the model. This project used a similar approach with DeepLearningFlappyBird, a Python Flappy Bird RL implementation. The main RL architecture is Q-Learning, a Convolutional Neural Network (CNN). In each game action stage, we store the current state of the bird, the action the agent took, and the next state of the bird. These are treated as the training data of the CNN.

CNN Training Overview

The input data for training is a continuous four-frame image. We stack these four images to form an “observation” of the bird. The observation here means time-series data represented by a series of images. The image itself is gray-scaled to reduce the training load. The array representation of the image is (batch size, 4 (frames), 80 (width), 80 (height)). Each element of the array represents the pixel value of each frame. These data are fed into the CNN and compute to an output (batch size, 2). The second dimension of the output represents the confidence of the next action (flap, no-flap).

We use the actual action recorded against the output confidence to compute the loss. After that, the model will be updated through back propagation and parameter optimization. The data used for training are continuously updated by the agent to achieve a better result.

Training data

After the action stage, we create preObservation and currentObservation. As mentioned before, these are just images that represent a series of movement. After that, we just put preObservation, currentObservation, action, reward and terminal together as a step stored into the replayBuffer. The replayBuffer is the training dataset with a limited size and dynamically updated with the latest actions.

public void step(NDList action, boolean training) {
    if (action.singletonOrThrow().getInt(1) == 1) {
        bird.birdFlap();
    }
    stepFrame();
    NDList preObservation = currentObservation;
    currentObservation = createObservation(currentImg);
    FlappyBirdStep step = new FlappyBirdStep(manager.newSubManager(),
            preObservation, currentObservation, action, currentReward, currentTerminal);
    if (training) {
        replayBuffer.addStep(step);
    }
    if (gameState == GAME_OVER) {
        restartGame();
    }
}

Three stages of RL

There are three different stages of RL used to generate better training data:

  • Observation Stage: Most actions are random with a small portion of actions coming from the AI agent
  • Exploration Stage: Random actions and AI agent actions are combined
  • Training Stage: Actions are primarily produced by the AI agent

During the exploration stage, we will choose between random action and AI agent action for the bird. At the beginning of training random actions are primarily used, since the actions generated by the AI agent are generally poor. After that, we gradually increase the probability of taking the AI agent’s action until it ultimately becomes the only decision maker. The parameter that is used to adjust the ratio between random and AI agent actions is called epsilon. It will change constantly through the training process.

public NDList chooseAction(RlEnv env, boolean training) {
    if (training && RandomUtils.random() < exploreRate.getNewValue(counter++)) {
        return env.getActionSpace().randomAction();
    } else return baseAgent.chooseAction(env, training);
}

#java #reinforcement-learning #machine-learning #ai

What is GEEK

Buddha Community

Train “undying” Flappy Bird using Reinforcement Learning on Java
Chloe  Butler

Chloe Butler

1667425440

Pdf2gerb: Perl Script Converts PDF Files to Gerber format

pdf2gerb

Perl script converts PDF files to Gerber format

Pdf2Gerb generates Gerber 274X photoplotting and Excellon drill files from PDFs of a PCB. Up to three PDFs are used: the top copper layer, the bottom copper layer (for 2-sided PCBs), and an optional silk screen layer. The PDFs can be created directly from any PDF drawing software, or a PDF print driver can be used to capture the Print output if the drawing software does not directly support output to PDF.

The general workflow is as follows:

  1. Design the PCB using your favorite CAD or drawing software.
  2. Print the top and bottom copper and top silk screen layers to a PDF file.
  3. Run Pdf2Gerb on the PDFs to create Gerber and Excellon files.
  4. Use a Gerber viewer to double-check the output against the original PCB design.
  5. Make adjustments as needed.
  6. Submit the files to a PCB manufacturer.

Please note that Pdf2Gerb does NOT perform DRC (Design Rule Checks), as these will vary according to individual PCB manufacturer conventions and capabilities. Also note that Pdf2Gerb is not perfect, so the output files must always be checked before submitting them. As of version 1.6, Pdf2Gerb supports most PCB elements, such as round and square pads, round holes, traces, SMD pads, ground planes, no-fill areas, and panelization. However, because it interprets the graphical output of a Print function, there are limitations in what it can recognize (or there may be bugs).

See docs/Pdf2Gerb.pdf for install/setup, config, usage, and other info.


pdf2gerb_cfg.pm

#Pdf2Gerb config settings:
#Put this file in same folder/directory as pdf2gerb.pl itself (global settings),
#or copy to another folder/directory with PDFs if you want PCB-specific settings.
#There is only one user of this file, so we don't need a custom package or namespace.
#NOTE: all constants defined in here will be added to main namespace.
#package pdf2gerb_cfg;

use strict; #trap undef vars (easier debug)
use warnings; #other useful info (easier debug)


##############################################################################################
#configurable settings:
#change values here instead of in main pfg2gerb.pl file

use constant WANT_COLORS => ($^O !~ m/Win/); #ANSI colors no worky on Windows? this must be set < first DebugPrint() call

#just a little warning; set realistic expectations:
#DebugPrint("${\(CYAN)}Pdf2Gerb.pl ${\(VERSION)}, $^O O/S\n${\(YELLOW)}${\(BOLD)}${\(ITALIC)}This is EXPERIMENTAL software.  \nGerber files MAY CONTAIN ERRORS.  Please CHECK them before fabrication!${\(RESET)}", 0); #if WANT_DEBUG

use constant METRIC => FALSE; #set to TRUE for metric units (only affect final numbers in output files, not internal arithmetic)
use constant APERTURE_LIMIT => 0; #34; #max #apertures to use; generate warnings if too many apertures are used (0 to not check)
use constant DRILL_FMT => '2.4'; #'2.3'; #'2.4' is the default for PCB fab; change to '2.3' for CNC

use constant WANT_DEBUG => 0; #10; #level of debug wanted; higher == more, lower == less, 0 == none
use constant GERBER_DEBUG => 0; #level of debug to include in Gerber file; DON'T USE FOR FABRICATION
use constant WANT_STREAMS => FALSE; #TRUE; #save decompressed streams to files (for debug)
use constant WANT_ALLINPUT => FALSE; #TRUE; #save entire input stream (for debug ONLY)

#DebugPrint(sprintf("${\(CYAN)}DEBUG: stdout %d, gerber %d, want streams? %d, all input? %d, O/S: $^O, Perl: $]${\(RESET)}\n", WANT_DEBUG, GERBER_DEBUG, WANT_STREAMS, WANT_ALLINPUT), 1);
#DebugPrint(sprintf("max int = %d, min int = %d\n", MAXINT, MININT), 1); 

#define standard trace and pad sizes to reduce scaling or PDF rendering errors:
#This avoids weird aperture settings and replaces them with more standardized values.
#(I'm not sure how photoplotters handle strange sizes).
#Fewer choices here gives more accurate mapping in the final Gerber files.
#units are in inches
use constant TOOL_SIZES => #add more as desired
(
#round or square pads (> 0) and drills (< 0):
    .010, -.001,  #tiny pads for SMD; dummy drill size (too small for practical use, but needed so StandardTool will use this entry)
    .031, -.014,  #used for vias
    .041, -.020,  #smallest non-filled plated hole
    .051, -.025,
    .056, -.029,  #useful for IC pins
    .070, -.033,
    .075, -.040,  #heavier leads
#    .090, -.043,  #NOTE: 600 dpi is not high enough resolution to reliably distinguish between .043" and .046", so choose 1 of the 2 here
    .100, -.046,
    .115, -.052,
    .130, -.061,
    .140, -.067,
    .150, -.079,
    .175, -.088,
    .190, -.093,
    .200, -.100,
    .220, -.110,
    .160, -.125,  #useful for mounting holes
#some additional pad sizes without holes (repeat a previous hole size if you just want the pad size):
    .090, -.040,  #want a .090 pad option, but use dummy hole size
    .065, -.040, #.065 x .065 rect pad
    .035, -.040, #.035 x .065 rect pad
#traces:
    .001,  #too thin for real traces; use only for board outlines
    .006,  #minimum real trace width; mainly used for text
    .008,  #mainly used for mid-sized text, not traces
    .010,  #minimum recommended trace width for low-current signals
    .012,
    .015,  #moderate low-voltage current
    .020,  #heavier trace for power, ground (even if a lighter one is adequate)
    .025,
    .030,  #heavy-current traces; be careful with these ones!
    .040,
    .050,
    .060,
    .080,
    .100,
    .120,
);
#Areas larger than the values below will be filled with parallel lines:
#This cuts down on the number of aperture sizes used.
#Set to 0 to always use an aperture or drill, regardless of size.
use constant { MAX_APERTURE => max((TOOL_SIZES)) + .004, MAX_DRILL => -min((TOOL_SIZES)) + .004 }; #max aperture and drill sizes (plus a little tolerance)
#DebugPrint(sprintf("using %d standard tool sizes: %s, max aper %.3f, max drill %.3f\n", scalar((TOOL_SIZES)), join(", ", (TOOL_SIZES)), MAX_APERTURE, MAX_DRILL), 1);

#NOTE: Compare the PDF to the original CAD file to check the accuracy of the PDF rendering and parsing!
#for example, the CAD software I used generated the following circles for holes:
#CAD hole size:   parsed PDF diameter:      error:
#  .014                .016                +.002
#  .020                .02267              +.00267
#  .025                .026                +.001
#  .029                .03167              +.00267
#  .033                .036                +.003
#  .040                .04267              +.00267
#This was usually ~ .002" - .003" too big compared to the hole as displayed in the CAD software.
#To compensate for PDF rendering errors (either during CAD Print function or PDF parsing logic), adjust the values below as needed.
#units are pixels; for example, a value of 2.4 at 600 dpi = .0004 inch, 2 at 600 dpi = .0033"
use constant
{
    HOLE_ADJUST => -0.004 * 600, #-2.6, #holes seemed to be slightly oversized (by .002" - .004"), so shrink them a little
    RNDPAD_ADJUST => -0.003 * 600, #-2, #-2.4, #round pads seemed to be slightly oversized, so shrink them a little
    SQRPAD_ADJUST => +0.001 * 600, #+.5, #square pads are sometimes too small by .00067, so bump them up a little
    RECTPAD_ADJUST => 0, #(pixels) rectangular pads seem to be okay? (not tested much)
    TRACE_ADJUST => 0, #(pixels) traces seemed to be okay?
    REDUCE_TOLERANCE => .001, #(inches) allow this much variation when reducing circles and rects
};

#Also, my CAD's Print function or the PDF print driver I used was a little off for circles, so define some additional adjustment values here:
#Values are added to X/Y coordinates; units are pixels; for example, a value of 1 at 600 dpi would be ~= .002 inch
use constant
{
    CIRCLE_ADJUST_MINX => 0,
    CIRCLE_ADJUST_MINY => -0.001 * 600, #-1, #circles were a little too high, so nudge them a little lower
    CIRCLE_ADJUST_MAXX => +0.001 * 600, #+1, #circles were a little too far to the left, so nudge them a little to the right
    CIRCLE_ADJUST_MAXY => 0,
    SUBST_CIRCLE_CLIPRECT => FALSE, #generate circle and substitute for clip rects (to compensate for the way some CAD software draws circles)
    WANT_CLIPRECT => TRUE, #FALSE, #AI doesn't need clip rect at all? should be on normally?
    RECT_COMPLETION => FALSE, #TRUE, #fill in 4th side of rect when 3 sides found
};

#allow .012 clearance around pads for solder mask:
#This value effectively adjusts pad sizes in the TOOL_SIZES list above (only for solder mask layers).
use constant SOLDER_MARGIN => +.012; #units are inches

#line join/cap styles:
use constant
{
    CAP_NONE => 0, #butt (none); line is exact length
    CAP_ROUND => 1, #round cap/join; line overhangs by a semi-circle at either end
    CAP_SQUARE => 2, #square cap/join; line overhangs by a half square on either end
    CAP_OVERRIDE => FALSE, #cap style overrides drawing logic
};
    
#number of elements in each shape type:
use constant
{
    RECT_SHAPELEN => 6, #x0, y0, x1, y1, count, "rect" (start, end corners)
    LINE_SHAPELEN => 6, #x0, y0, x1, y1, count, "line" (line seg)
    CURVE_SHAPELEN => 10, #xstart, ystart, x0, y0, x1, y1, xend, yend, count, "curve" (bezier 2 points)
    CIRCLE_SHAPELEN => 5, #x, y, 5, count, "circle" (center + radius)
};
#const my %SHAPELEN =
#Readonly my %SHAPELEN =>
our %SHAPELEN =
(
    rect => RECT_SHAPELEN,
    line => LINE_SHAPELEN,
    curve => CURVE_SHAPELEN,
    circle => CIRCLE_SHAPELEN,
);

#panelization:
#This will repeat the entire body the number of times indicated along the X or Y axes (files grow accordingly).
#Display elements that overhang PCB boundary can be squashed or left as-is (typically text or other silk screen markings).
#Set "overhangs" TRUE to allow overhangs, FALSE to truncate them.
#xpad and ypad allow margins to be added around outer edge of panelized PCB.
use constant PANELIZE => {'x' => 1, 'y' => 1, 'xpad' => 0, 'ypad' => 0, 'overhangs' => TRUE}; #number of times to repeat in X and Y directions

# Set this to 1 if you need TurboCAD support.
#$turboCAD = FALSE; #is this still needed as an option?

#CIRCAD pad generation uses an appropriate aperture, then moves it (stroke) "a little" - we use this to find pads and distinguish them from PCB holes. 
use constant PAD_STROKE => 0.3; #0.0005 * 600; #units are pixels
#convert very short traces to pads or holes:
use constant TRACE_MINLEN => .001; #units are inches
#use constant ALWAYS_XY => TRUE; #FALSE; #force XY even if X or Y doesn't change; NOTE: needs to be TRUE for all pads to show in FlatCAM and ViewPlot
use constant REMOVE_POLARITY => FALSE; #TRUE; #set to remove subtractive (negative) polarity; NOTE: must be FALSE for ground planes

#PDF uses "points", each point = 1/72 inch
#combined with a PDF scale factor of .12, this gives 600 dpi resolution (1/72 * .12 = 600 dpi)
use constant INCHES_PER_POINT => 1/72; #0.0138888889; #multiply point-size by this to get inches

# The precision used when computing a bezier curve. Higher numbers are more precise but slower (and generate larger files).
#$bezierPrecision = 100;
use constant BEZIER_PRECISION => 36; #100; #use const; reduced for faster rendering (mainly used for silk screen and thermal pads)

# Ground planes and silk screen or larger copper rectangles or circles are filled line-by-line using this resolution.
use constant FILL_WIDTH => .01; #fill at most 0.01 inch at a time

# The max number of characters to read into memory
use constant MAX_BYTES => 10 * M; #bumped up to 10 MB, use const

use constant DUP_DRILL1 => TRUE; #FALSE; #kludge: ViewPlot doesn't load drill files that are too small so duplicate first tool

my $runtime = time(); #Time::HiRes::gettimeofday(); #measure my execution time

print STDERR "Loaded config settings from '${\(__FILE__)}'.\n";
1; #last value must be truthful to indicate successful load


#############################################################################################
#junk/experiment:

#use Package::Constants;
#use Exporter qw(import); #https://perldoc.perl.org/Exporter.html

#my $caller = "pdf2gerb::";

#sub cfg
#{
#    my $proto = shift;
#    my $class = ref($proto) || $proto;
#    my $settings =
#    {
#        $WANT_DEBUG => 990, #10; #level of debug wanted; higher == more, lower == less, 0 == none
#    };
#    bless($settings, $class);
#    return $settings;
#}

#use constant HELLO => "hi there2"; #"main::HELLO" => "hi there";
#use constant GOODBYE => 14; #"main::GOODBYE" => 12;

#print STDERR "read cfg file\n";

#our @EXPORT_OK = Package::Constants->list(__PACKAGE__); #https://www.perlmonks.org/?node_id=1072691; NOTE: "_OK" skips short/common names

#print STDERR scalar(@EXPORT_OK) . " consts exported:\n";
#foreach(@EXPORT_OK) { print STDERR "$_\n"; }
#my $val = main::thing("xyz");
#print STDERR "caller gave me $val\n";
#foreach my $arg (@ARGV) { print STDERR "arg $arg\n"; }

Download Details:

Author: swannman
Source Code: https://github.com/swannman/pdf2gerb

License: GPL-3.0 license

#perl 

Seamus  Quitzon

Seamus Quitzon

1602637135

Learning by Doing: How to Learn Java Basics by Building Your Own Project

Java is not the hardest language to start with. So, it becomes way popular among novice developers joining the ranks of Java coders every single day. If you are reading this blog post, you might be interested in learning Java.

Java is widely used across industry, and especially in the area of Enterprise software, which results in many high paying job opportunities and makes this programming language a common language for newbies. A general promotion of it within colleges and other institutions providing a formal Computer Science education also contributes to its popularity.

However, these are not the only advantages of Java — among other things, it allows you to adopt good practices and makes it way easier to learn other languages in the future. And with no doubt, you can easily learn it if you’re following the right approach. In this post, I am going to share some of them with you.

The Importance of Practice in Programming

Beyond all doubt, practice is important and valuable. But, before we get to the advantages of hands-on experience, I want to draw your attention to one essential thing I often tell my students.

New programmers who are just learning and start implementing things, without being supervised, often end up adapting bad practices. To avoid that, especially when you are making your first steps in programming, I recommend looking for a person who will supervise you and teach you. A strong mentorship with someone engaged in a serious project, as well as communication within the community in the form of sharing code and asking for feedback, is worth the effort. Similarly, when you are applying for your first job, you want to be looking for a company with a strong team and a good leader who would be keen on investing into your learning.

Now, let’s return to practical experience. Learning by doing is different from learning by passively consuming the information. To make sure we can use all the newly acquired technology, we should put our skills to test and write tons of code. The benefits of hands-on experience are almost endless.

Efficiency and Productivity

By practicing, you get a clear understanding of what programming is. Consequently, you start doing better with each new hands-on task, complete it faster, and thus become more productive.

Even if you are not working on real-world projects yet, it’s important to get used to having deadlines. They are inextricably linked to the programming process. My recommendation is to set up your own deadlines while practicing stage and follow them as closely as possible.

#java #learn java #java code #learn java in easy way #learn java course #learn java development

Tyrique  Littel

Tyrique Littel

1600135200

How to Install OpenJDK 11 on CentOS 8

What is OpenJDK?

OpenJDk or Open Java Development Kit is a free, open-source framework of the Java Platform, Standard Edition (or Java SE). It contains the virtual machine, the Java Class Library, and the Java compiler. The difference between the Oracle OpenJDK and Oracle JDK is that OpenJDK is a source code reference point for the open-source model. Simultaneously, the Oracle JDK is a continuation or advanced model of the OpenJDK, which is not open source and requires a license to use.

In this article, we will be installing OpenJDK on Centos 8.

#tutorials #alternatives #centos #centos 8 #configuration #dnf #frameworks #java #java development kit #java ee #java environment variables #java framework #java jdk #java jre #java platform #java sdk #java se #jdk #jre #open java development kit #open source #openjdk #openjdk 11 #openjdk 8 #openjdk runtime environment

Vaughn  Sauer

Vaughn Sauer

1623428700

Train “undying” Flappy Bird using Reinforcement Learning on Java

Flappy Bird is a mobile game that was introduced in 2013 which became super popular because of its simple way to play (flap/no-flap). With the growth of Deep Learning (DL) and Reinforcement Learning (RL), we can now train an AI agent to control the Flappy Bird actions. Today, we will look at the process to create an AI agent using Java. For the game itself, we used a simple open-source Flappy Bird game using Java. For training, we used Deep Java Library (DJL), a deep learning framework based on Java, to build the training network and RL algorithm.

In this article, we will start from the basis of RL and walk through the key components to build the training architecture. If at anytime you cannot follow our code and would like to try the game, you can refer to our RL repo.

The RL Architecture

In this section, we will introduce some major algorithm and networks we used to help you better understand how we trained the model. This project used a similar approach with DeepLearningFlappyBird, a Python Flappy Bird RL implementation. The main RL architecture is Q-Learning, a Convolutional Neural Network (CNN). In each game action stage, we store the current state of the bird, the action the agent took, and the next state of the bird. These are treated as the training data of the CNN.

CNN Training Overview

The input data for training is a continuous four-frame image. We stack these four images to form an “observation” of the bird. The observation here means time-series data represented by a series of images. The image itself is gray-scaled to reduce the training load. The array representation of the image is (batch size, 4 (frames), 80 (width), 80 (height)). Each element of the array represents the pixel value of each frame. These data are fed into the CNN and compute to an output (batch size, 2). The second dimension of the output represents the confidence of the next action (flap, no-flap).

We use the actual action recorded against the output confidence to compute the loss. After that, the model will be updated through back propagation and parameter optimization. The data used for training are continuously updated by the agent to achieve a better result.

Training data

After the action stage, we create preObservation and currentObservation. As mentioned before, these are just images that represent a series of movement. After that, we just put preObservation, currentObservation, action, reward and terminal together as a step stored into the replayBuffer. The replayBuffer is the training dataset with a limited size and dynamically updated with the latest actions.

public void step(NDList action, boolean training) {
    if (action.singletonOrThrow().getInt(1) == 1) {
        bird.birdFlap();
    }
    stepFrame();
    NDList preObservation = currentObservation;
    currentObservation = createObservation(currentImg);
    FlappyBirdStep step = new FlappyBirdStep(manager.newSubManager(),
            preObservation, currentObservation, action, currentReward, currentTerminal);
    if (training) {
        replayBuffer.addStep(step);
    }
    if (gameState == GAME_OVER) {
        restartGame();
    }
}

Three stages of RL

There are three different stages of RL used to generate better training data:

  • Observation Stage: Most actions are random with a small portion of actions coming from the AI agent
  • Exploration Stage: Random actions and AI agent actions are combined
  • Training Stage: Actions are primarily produced by the AI agent

During the exploration stage, we will choose between random action and AI agent action for the bird. At the beginning of training random actions are primarily used, since the actions generated by the AI agent are generally poor. After that, we gradually increase the probability of taking the AI agent’s action until it ultimately becomes the only decision maker. The parameter that is used to adjust the ratio between random and AI agent actions is called epsilon. It will change constantly through the training process.

public NDList chooseAction(RlEnv env, boolean training) {
    if (training && RandomUtils.random() < exploreRate.getNewValue(counter++)) {
        return env.getActionSpace().randomAction();
    } else return baseAgent.chooseAction(env, training);
}

#java #reinforcement-learning #machine-learning #ai

Samanta  Moore

Samanta Moore

1621826659

Important Things For Java Developers To Learn In 2021

If you are looking to learn Java, you may be wondering where to start. Which technologies should you focus on? Whether you are new to the language, a middle-level learner, or already using Java at work, this article explores the essentials that you need to know.

Learning a programming language is a technological process that requires serious preparation. Otherwise, you can easily “choke” on the learning process itself.

I work for a company that created an interactive Java online course. From time to time, our graduates tell us about what they are required to know in interviews, and also about what technologies they use in their work. Based on these surveys, a shortlist of such technologies can be compiled.

#java #java-development #learn-to-code-java #tech-trends #learn-java #learning #learning-to-code #education