How can users get started using the BIC’s new GPU processors in their image pipelines?
Using the SGE batch system with the queue gpu.q:
qsub -q gpu.q your_script.sh
I’ve used Theano for training artificial neural networks on this machine, if anyone has more specific questions. There’s also some documentation here: http://deeplearning.net/software/theano/tutorial/using_gpu.html
The link above for the tutorial doesn’t appear to respond at times, but the following should/might work:
https://theano-pymc.readthedocs.io/en/latest/tutorial/using_gpu.html
Otherwise I asked ChatGPT (https://chat.openai.com/chat) The following question:
Give me a plan to learn about theano and deep learning.
I’ll spare you the answer as it was quite long and elaborate, not exactly what is needed here, but then I asked the following:
Do you know of any online tutorials regarding theano and deep learning ?
-
Theano documentation: The official Theano documentation is a great resource for learning about the library. It includes tutorials, examples, and API documentation.
-
Deep Learning with Theano: This is a free online book that provides an introduction to deep learning with Theano. It covers the basics of neural networks and how to implement them using Theano.
-
Udemy courses: Udemy offers several courses on deep learning with Theano, including “Deep Learning with Theano and TensorFlow” and “Theano: A Comprehensive Tutorial.”
-
Coursera courses: Coursera also offers several courses on deep learning with Theano, including “Deep Learning” and “Neural Networks and Deep Learning.”
-
TensorFlow and Theano tutorials by Brandon Rohrer: Brandon Rohrer has created several YouTube tutorials on deep learning with TensorFlow and Theano. They cover a range of topics, from basics to advanced techniques.
-
Kaggle tutorials: Kaggle offers several tutorials on deep learning using Theano. They cover a range of topics, from image classification to natural language processing.