There are several alternatives, such as using Theano (http://deeplearning.net/tutorial/), Caffe (http://caffe.berkeleyvision.org/), cuda-convnet2 (https://code.google.com/p/cuda-convnet2/) etc. I am working in Win 7 environment.
I would like to know what is the best starting point for baisc as well as advanced study of deep machine learning.
It depends on your other skills, If you are interested in how CNNs work, you should consider implementing one yourself (its better to start with feed forward ANN). If your interests lie in using it to solve some other problems, learn Theano/ any_other which saves you on ML part so that you can concentrate on your domain.
Overall Theano is more favorable to me.
Hi,
here is a link to a good tutorial about deep learning:
http://danielnouri.org/notes/2014/12/17/using-convolutional-neural-nets-to-detect-facial-keypoints-tutorial/
Best
Hi,
I would start from Yann LeCun puplications http://yann.lecun.com/exdb/publis/index.html
this will help you to understand well the CNN.
You can also take Geoffrey Hinton course of Neural Network available in Coursera https://www.coursera.org/course/neuralnets
Good luck.
I believe the best website for starting is as follows:
https://machinelearningmastery.com/sequence-classification-lstm-recurrent-neural-networks-python-keras/
https://www.datacamp.com/community/tutorials/deep-learning-python
- Badges
- Science topic