Deep LearningDeep Learning

This extension provides Deep Learning capabilities for execution on CPU and GPU.

This extension provides operators to create Deep Learning models using different types of layers. Networks can be executed both on CPU and on GPU. It requires the ND4J Back End Extension for configuration of computational resources.

Version 0.9.4:

    Added support for image handling (use the Image Handling Extension for loading images as tensor input)

    Back end handling moved into new extension with additional features (ND4J Back End Extension - automatic dependency)

    Added early stopping mechanism

    Added weights and biases output

    Fixed label mapping bug in TimeSeries to Tensor operator

 

Version 0.9.3:

    Added support for many-to-many classification and regression use-cases, as well as many-to-one regression.

    Added ExampleSet to Tensor operator

    Added logging of test scores to history port

    Added lasso (L1) and ridge (L2) regression loss functions

    Added support for macOS Catalina

    Added support for cuDNN in version 7.4

    Most layers are now taking advantage of cuDNN if installed

    Updated back end to DeepLearning4J Beta6

    Updated CUDA dependency to version 10.0 (if you need support for CUDA 9.2, 10.1, 10.2, please contact us)

    Fixed error, when scoring only one Example

    Fixed bug in displaying parameters of the Add Dropout Layer operator

    Dropped support for GPU on macOS

 

Version 0.9.1:

    Fixed bug during execution on job agents

 

Version 0.9.0:

    Added Load Keras Model Operator (applying sequential Keras models without python)

    Added Recurrent Network (like LSTM) handling

    Added LSTM Layer

    Added Time-Series to Tensor Operator

    Fix of Anaconda blocking extension loading

    Removed Text to Numbers via Word2Vec

    Changed tensor handling (incompatible with previous tensors)

    Lowered CUDA version requirement from 9.1 to 9.0.

 

Version 0.8.1:

    fixed bug causing incompatibility with RapidMiner Studio 9.1

 

Version 0.8.0:

    Deep Learning on ExampleSets with native model handling

    Text Handling using Word2Vec

    Layers:

        fully-connected

        dropout

        batch normalization

        convolutional

        pooling

        global pooling

    GPU usage

    History Port (epoch logging)

    Custom Icons

    No external requirements (except for GPU)

    QuickFixes for switching between regression & classification (loss functions)

    Model Updatability

    Samples Processes (samples/Deep Learning)

 

Remarks:

Execution on GPU is currently only available on NVIDIA GPUs in combination with CUDA version 10.0.

This extension uses the java library DeepLearning4J (version 1.0-beta6).

 

No support for 32-bit.


Product Details

Version 0.9.4
File size 125 MB
Downloads 26234 (86 Today)26234 downloads
Vendor RapidMiner Labs
Category Machine Learning
Released 6/2/20
Last Update 6/2/20 2:11 PM
(Changes)
License AGPL
Product web site www.rapidminer.com
Rating 0.0 stars(0)

Comments

Extension version 0.9: (RM9.1) only supports CUDA 9.0 and CUDNN 7.0 EXACTLY. Ensure that CUDA bin directory is in your path before starting RM. Extension version 0.8.0: (RM9.0) only supports CUDA 9.1 and CUDNN 7.1 EXACTLY

wongcr, 12/21/18 8:12 AM

The data parsing problem occurs when a 32-bit Version of RapidMiner Studio or Java is used. Please check the 64-bit version.

Rapid-Labs, 8/8/18 2:24 PM

I only get a "data parsing problem" error even though with all numerical example sets. Is there a fix for this? Thanks...

Gottfried, 8/8/18 12:00 PM
Log in to post comments.