23
edits
Summer of Code - Getting Started: Difference between revisions
→Neural Networks package: Convolutional Neural Networks
| Line 231: | Line 231: | ||
=== Neural Networks package: Convolutional Neural Networks === | === Neural Networks package: Convolutional Neural Networks === | ||
Convolutional Neural Networks (CNNs) have recently become the state-of-the-art for image recognition and are widely used for solving classification and regression problems and for [https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/tutorials/deepdream/deepdream.ipynb | Convolutional Neural Networks (CNNs) have recently become the state-of-the-art for image recognition and are widely used for solving classification and regression problems and for [https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/tutorials/deepdream/deepdream.ipynb image generation]. The goal of the project is to implement a [https://it.mathworks.com/help/nnet/convolutional-neural-networks.html Matlab compatible] CNN toolbox using Google's library [https://www.tensorflow.org/ TensorFlow], which has a Python and C++ interface. As execution environment for the training function, the user will be able to choose between single/multiple CPUs and GPUs. | ||
* '''Required skills''' | * '''Required skills''' | ||
: C, C++, Python, m-file scripting, familiarity with Machine Learning algorithms, basic knowledge of Makefiles. | : C, C++, Python, m-file scripting, familiarity with Machine Learning algorithms, basic knowledge of Makefiles, experience with parallel computing and distributed systems. | ||
* '''Difficulty''' | * '''Difficulty''' | ||
: Difficult. | : Difficult. | ||