- 9
- 284 869
Deeplearning4j - Skymind
United States
Приєднався 7 гру 2014
Lecture 7 | Keras Model Import
This screencasts describes how to import a Neural Network that was created and trained using Keras, into DeepLearning4J
Переглядів: 8 474
Відео
Lecture 0: Getting started with DeepLearning4j
Переглядів 63 тис.7 років тому
This screencast walks through the Quickstart guide at deeplearning4j.org/quickstart. Topics include DeepLearning4j dependencies and tools necessary for the install such as the automated build tool Maven, Git, the IDE Intellij, and working with DeepLearning4j examples. github.com/deeplearning4j/dl4j-examples By the end of this tutorial, you should be able to run an example. Browse all our public...
Lecture 7 | Import a Keras Neural Net Model into Deeplearning4j
Переглядів 23 тис.7 років тому
This screencast showing how to import a Neural Network from the open-source Python framework Keras (keras.io) into DeepLearning4J. Keras, in turn, can import models from Theano, Tensorflow, Caffe and Torch, so through Keras, Deeplearning4j is able to import models from all major deep-learning frameworks. This allows large teams to do data exploration and prototyping in Python, and then deploy t...
Lecture 6 | Testing Wild Images Against Mnist Trained Network
Переглядів 12 тис.7 років тому
In this tutorial, we demonstrate testing images against a Neural Net that had been trained against the MNIST dataset of hand written digits. Browse all our public training courses at skymind.ai/academy.
Lecture 1 | Building a Linear Classifier (MLP) With Deeplearning4j
Переглядів 108 тис.7 років тому
Lecture by Instructor Tom Hanlon on Machine Learning. Tom provides an overview of how to build a simple neural net in this introductory tutorial. This screencast shows how to build a Linear Classifier using Deeplearning4j. Browse all our public training courses at skymind.ai/academy.
Lecture 2 | Preprocessing Data for Machine Learning With Datavec & Spark
Переглядів 25 тис.7 років тому
This screencast shows how to use Skymind's DataVec to ingest Comma Separated Values from a text file, convert the fields to numeric using a DataVec Transform Process in Spark, and save the modified data. Transforming non-numeric data to numeric data is a key preliminary step to using a Neural Network to analyze the data. Browse all our public training courses at skymind.ai/academy.
Lecture 3 | Building an Image Pipeline for Deeplearning4j With DataVec
Переглядів 20 тис.7 років тому
This screencast shows how to use Skymind's DataVec to build an image pipeline to prepare data for processing in a Neural Network using DeepLearning4j. Topics covered include, ParentPathLabelGenerator and Image Scaling. Browse all our public training courses at skymind.ai/academy.
Lecture 4 | Plugging a Neural Network into an Image Pipeline
Переглядів 10 тис.7 років тому
In this tutorial a Neural Network is added to the image pipeline example. The data set is then used to train a Feed Forward Neural Network using DeepLearning4j. Browse all our public training courses at skymind.ai/academy.
Lecture 5 | Save And Load Deep Learning Models With Deeplearning4j
Переглядів 16 тис.7 років тому
In this tutorial, a trained neural network is saved, and then loaded for use with Deeplearning4j. Browse all our public training courses at skymind.ai/academy.
garbage
Could you please tell us what you did at the end of line 91? It's not visible in the video...
github repo doesn't appear to exist?
nowadays no one use this shit language for data science
Thank you for a very clear overview. Also, this guy has most calm, reassuring voice of anyone I've ever heard in a UA-cam tech video. I think my blood pressure is measurably lower now than it was when I started watching.
how can we import a .h5 model file and test in java without the json file
Ah, 2016, no greetings with a guy on cocaine "WHAAATS UP GUYS, WELCOME TO MY CHANNEL, THUMBS UP, SUBSCRIBE AND COMMENT, YES, EVEN BEFORE YOU HAVE SEEN THE FIRST 10 SECONDS OF THE VIDEO, YES, I ON DRUGS, ENERGY DRINKS, I WANT TO BECOME A UA-camR... ". No, instead a calm voice, sovereign, presenting only the facts. What a joy. I miss 2016.
I have to say that it is weird how I see this 6 years old video, is like best Java AI related video in youtube. props for creator, Deeplearning4j Apis have changed little so this is outdated. Java AI promotion is so bad state, first I get terrible Tensor flow videos(it Java api 20% features implemented is worst possible choise for Java dev), Python it is good for it but not Java. So much of AI community has moved using Python that is even hard to find good videos for Java about subject. some small changes to made it work, if this helps someone .. RecordReader rr = new CSVRecordReader('\t'); .. RecordReader rrTest = new CSVRecordReader('\t'); Nesterovs updater = new Nesterovs(); updater.setMomentum(0.9); updater.setLearningRate(learningRate); conf.. .seed(seed).maxNumLineSearchIterations(1).optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT).updater(updater) .. .weightInit(WeightInit.XAVIER).activation(Activation.RELU) .. conf.setBackpropType(BackpropType.Standard); also i ovverride ScoreIterationListener class to print my screen .. INDArray features = t.getFeatures(); worked fine
Great video.
EXCELLENT!
The deeplearning4j api has changed so much that this tutorial is now somehow hard to follow using the latest API for DL4J.... To follow the tutorial, make sure you use the 0.7.0 version of DeepLearning4J
Hi. could you post another series with the latest DL4j 1.0.0 Beta 7? I am having major problems with my code. thanks.
There is no pom.xml file in the outer most file directory so when I run mvn clean install it doesn't work, and when you try to run it on the other pom.xml files it runs but get's stuck with no result. Outdated video maybe?
Outdated, most likely. Just tried to run the "mvn clean install" on my own system, and it was just giving me a build error. The exact same thing happened.
@@JLSXMK8 Greetings, I wanted to ask if you figured it out eventually? If yes, then how?
Could I have a link for download?
Idk why but this sound scary
Wtf was that.....I didn't learn anything!!!
getting below exception when i try to run in IntelliJ. followed all instructions in the video. do we have a separate set up for the IRISData ? Exception in thread "main" java.io.IOException: Server returned HTTP response code: 403 for URL: @t at java.base/sun.net.www.protocol.http.HttpURLConnection.getInputStream0(HttpURLConnection.java:1919) at java.base/sun.net.www.protocol.http.HttpURLConnection.getInputStream(HttpURLConnection.java:1515) at java.base/sun.net.www.protocol.https.HttpsURLConnectionImpl.getInputStream(HttpsURLConnectionImpl.java:250)
Most soothing nn tutorial ever
Bots, where?
omg java :(
whats wrong with java ?
We are writing this and then writing this .. . . what is THIS ?? So just give us the codes if you dont explain .
last part of the code is cut off.... incompetency is astounding..
Which software tool is used?
it keeps saying" the java : package com.sun.media.sound does not exist" and gives a bunch of errors when i try to run any of the examples
You need to use JDK 8 instead of JDK 12 or versions above 8.
I want to use Scala, and not Java, but these videos are nice and clear and will help a lot! Thanks!
playlist would be useful edit ua-cam.com/play/PL9iheGibFMtrJZJFrTIdd-QNQyjDX5v2X.html
Looking so long at 'File' without 'new' was a very disturbing experience. Also code formatting in Idea is very fast and accessible, but idk if it exists in CE.
Yes, it is available in CE
sir can u just copying the code on command please
Thanks you for your guiding video. I followed all your steps, and got to the final result. Except, IntelliJ and JDK 11 doest support JavaFx which is used in class ImageDrawer in the folder dataexamples in the folder of the directory src,main,java
I also use JDK 11, and have some problems. Now do you run well?
the website doesn't work I clicked on it and it said 404
Subscribed
Ok, this all is cool, but... How I can extract an image path in evaluation part of code? So maybe I can send to a network the image, resized to 28x28 and get prediction category? I don't understand yet how to extract concrete category(int value)... Does can anybody respond me how ...?
What are hidden nodes, seed, updater, momentum, dense layer builder, iteration, weight init?
do you need to always run mvn clean install?
Here is my code that is updated to the latest version of this software: package net.neural; import org.datavec.api.records.reader.RecordReader; import org.datavec.api.records.reader.impl.csv.CSVRecordReader; import org.datavec.api.split.FileSplit; import org.deeplearning4j.datasets.datavec.RecordReaderDataSetIterator; import org.deeplearning4j.eval.Evaluation; import org.deeplearning4j.nn.api.OptimizationAlgorithm; import org.deeplearning4j.nn.conf.MultiLayerConfiguration; import org.deeplearning4j.nn.conf.NeuralNetConfiguration; import org.deeplearning4j.nn.conf.layers.DenseLayer; import org.deeplearning4j.nn.conf.layers.OutputLayer; import org.deeplearning4j.nn.multilayer.MultiLayerNetwork; import org.deeplearning4j.nn.weights.WeightInit; import org.deeplearning4j.optimize.listeners.ScoreIterationListener; import org.nd4j.linalg.activations.Activation; import org.nd4j.linalg.api.ndarray.INDArray; import org.nd4j.linalg.dataset.DataSet; import org.nd4j.linalg.dataset.api.iterator.DataSetIterator; import org.nd4j.linalg.learning.config.Nesterovs; import org.nd4j.linalg.lossfunctions.LossFunctions; import java.io.File; public class MLPClassifierLinear { public static void main(String[] args) { int seed = 123; double learningRate = 0.01; int batchSize = 50; int numberOfEpochs = 3; //30 is way more than you'll need, you get 100% accuracy, 3 is enough for 96% accuracy int numberOfInputs = 2; int numberOfOutputs = 2; int numberOfHiddenNodes = 20; //load training data RecordReader recordReader = new CSVRecordReader(); File testFile = new File("[pathToWhereIRanGitCloneFromOnWindows]\\dl4j-examples\\dl4j-examples\\src\\main\ esources\\classification\\linear_data_train.csv"); try { recordReader.initialize(new FileSplit(testFile)); } catch (Exception excecption) { excecption.printStackTrace(); } DataSetIterator trainingDataSetIterator = new RecordReaderDataSetIterator(recordReader, batchSize, 0, 2); //load evaluation data RecordReader recordReaderEvaluation = new CSVRecordReader(); File evaluationFile = new File("[pathToWhereIRanGitCloneFromOnWindows]]\\dl4j-examples\\dl4j-examples\\src\\main\ esources\\classification\\linear_data_eval.csv"); try { recordReaderEvaluation.initialize(new FileSplit(evaluationFile)); } catch (Exception excecption) { excecption.printStackTrace(); } DataSetIterator evaluationDataSetIterator = new RecordReaderDataSetIterator(recordReaderEvaluation, batchSize, 0, 2); MultiLayerConfiguration multiLayerConfiguration = new NeuralNetConfiguration.Builder() .seed(seed) .maxNumLineSearchIterations(1) .optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT) .updater(new Nesterovs(learningRate,0.5)) .list() .layer(0, new DenseLayer.Builder() .nIn(numberOfInputs) .nOut(numberOfHiddenNodes) .weightInit(WeightInit.XAVIER) .activation(Activation.RELU) .build() ) .layer(1, new OutputLayer.Builder(LossFunctions.LossFunction.NEGATIVELOGLIKELIHOOD) .nIn(numberOfHiddenNodes) .nOut(numberOfOutputs) .weightInit(WeightInit.XAVIER) .activation(Activation.SOFTMAX) .build() ) .pretrain(false).backprop(true).build(); MultiLayerNetwork multiLayerNetwork = new MultiLayerNetwork(multiLayerConfiguration); multiLayerNetwork.init(); multiLayerNetwork.setListeners(new ScoreIterationListener(10)); for(int n = 0; n < numberOfEpochs; n++){ multiLayerNetwork.fit(trainingDataSetIterator); } Evaluation evaluation = new Evaluation(numberOfOutputs); while(evaluationDataSetIterator.hasNext()){ DataSet dataSet = evaluationDataSetIterator.next(); INDArray features = dataSet.getFeatureMatrix(); INDArray labels = dataSet.getLabels(); INDArray predicted = multiLayerNetwork.output(features, false); evaluation.eval(labels, predicted); } System.out.println(evaluation.stats()); } }
Note: I changed some of the names of the variables from the tutorial because I thought it would be clearer with better names. If you're new to programming, this is probably more complex than you want to start out with.
None of these terminal commands work on windows. What should Windows users do?
Get a virtual Linux machine
Or download and use gitbash. Google for gitbash.
Where can I get the image data folder in training and testing?
How do we go about importing the Keras Tokenizer that has been fit on files for certain vocabulary size? Appreciate if you can provide any pointers
Just coding without any semantic information about what he is actually doing. Very bad!!! Especially for beginners in ML.
Like... actually wtf did he do omegalul
This is not for beginners, I recommend these videos for people who already know ML techniques and experience with other tools like Python tools, If you are starting you should start with Statistics fundamentals and any book about ML fundamentals.
Thank You! excellent tutorial, I changed the speed to x1.5 much better
Hi, I wanna know how to convert hand written text to binary text using java. Thanks
Is that working with Android Studio??
I don't know, but i think mobile devices arent really fast in calculating thing like Deep learning
No it is not Android Studio but I believe it's from the same company JetBrains, the IDE he is using is called IntelliJ Community Version which is a free IDE highly recommend over Eclipse if you are writing Java.
Okay, if this is outdated is there a (link to a) current one available that could be posted here?
Can you please leave out unnecessary dependencies as well? There are a lot of dependencies in this project, and I don't see the necessity of it all. I want to know as bare bones as possible how to use DeepLearning4j.
is this outdated? I get an error for org.deeplearning4j.nn.conf.Updator.NESTEROVS
Yes I would say this is pretty outdated now. There's been a lot of deprecation since this video was made.
I did everything step by step, but it gives me "UnsatisfiedLinkError: no gfortran in java.library.path" while running. I am using LinuxMint
Sadly, I can't figure out how to set up the project. It simply doesn't work for me. It seems to start from the middle of no where.
how do we save a word2vec model? It looks like Modelserializer is not able to handle a model when it is created as: Word2Vec model = new Word2Vec.Builder();
Hello again, I just wanted to ask if the DenseLayer is more like an input layer oder a hidden layer? And if it's not an input layer, if we do not have to configure an input layer?
Error:"method call expected java" Just found my mistake after quite some time desperate search. XD Hope it helps for those typing while watching. In 06:38 "File(...)" was used, in 14:02 you'll see the correct version. "new File(...)".
Oh yeah and I really appreciate you making the video! Big thanks!