It is surprising that Python was not yet mentioned here. Many Python programmers have contributed key packages for machine learning, including neural networks. Python is fast, esp. when its abilities to run on graphics cards are exploited. It is open source. Packages like Theano help you design different NN architectures, as it does differentiation for you.
MATLAB will guide you easily through the training process. MATLAB has built-in neural network toolbox that saves you from the hassle of coding and setting parameters. Later on, advanced code can also be generated from where you can change the parameters.
it depends on what neural network do you want to work with if you just want to implement it on some data set without programming i say SPSS Clementine but if you want to improve or develop some algorithm Matlab as other said too
I would recommend NETLAB made by Ian Nabney which has some of the standard algorithms for neural networks and an accompanying book: "Netlab: Algorithms for Pattern Recognition" and is the applied companion book for Bishops "Pattern Recognition and Machine Learning"
Nonetheless, most advanced and recent algorithms have not been integrated into standard packages so I would suggest writing to specific authors to see if they can share their code with you.
Weka (Waikato Environment for Knowledge Analysis) is a popular suite of machine learning software written in Java.User friendly software for training an ANN models.
It is surprising that Python was not yet mentioned here. Many Python programmers have contributed key packages for machine learning, including neural networks. Python is fast, esp. when its abilities to run on graphics cards are exploited. It is open source. Packages like Theano help you design different NN architectures, as it does differentiation for you.
MATLAB based training can be slow even when using parallel computing toolbox if there is a 'large' training set and there need to be multiple networks trained for sake of experiments. Some training algorithms that make use of regularization are not compatible with GPU so far, which seems like a promising pathway otherwise. I haven't tried playing with this much yet. Apart from this, universities have produced number of open-source softwares although installation and user friendly part of this is bit tedious and not great respectively... Any follow up thoughts from others? For using pre-trained networks, yes, can do with MATLAB and then go to third party via coder...so agree with that.
It depends upon the Area in which you want to use also. Kindly refer to
1. Answers given in Researchgate for the Question,"Which commercial software is user friendly for artificial neural network training? " asked by Lila Khatiwada.Answers by Arturo Geigel and Ylermi Cabrera are interesting.
When used without qualification, the terms “Neural Network” (NN) and “Artificial Neural Network” (ANN) usually refer to a Multilayer Perceptron Network. However, there are many other types of neural networks including Probabilistic Neural Networks, General Regression Neural Networks, Radial Basis Function Networks, Cascade Correlation, Functional Link Networks, Kohonen networks, Gram-Charlier networks, Learning Vector Quantization, Hebb networks, Adaline networks, Heteroassociative networks, Recurrent Networks and Hybrid Networks.
DTREG implements the most widely used types of neural networks: Multilayer Perceptron Networks (also known as multilayer feed-forward network), Cascade Correlation Neural Networks, Probabilistic Neural Networks (PNN) and General Regression Neural Networks (GRNN).
Neuroph is lightweight Java neural network framework to develop common neural network architectures. It contains well designed, open source Java library with small number of basic classes which correspond to basic NN concepts. Also has nice GUI neural network editor to quickly create Java neural network components. It has been released as open source under the Apache 2.0 license, and it's FREE for you to use it.