I am currently at lesson 2. Need help regarding the dogscats redux. When I try to run the code, I get resource exhaust error on my 4 GB graphics card. Any idea on how to deal with this? The code runs fine, albeit very slow when using CPU.
I am using keras with tensorflow backend and theano image ordering.
I am currently in Coimbatore , coming to Bangalore on 17th May . Would anyone be interested in a small meet up. I am currently working part-2 of this course.
@VishnuSubramanian Sure. That works. Meetup on a weekday evening or a weekend sometime? Let me know what works out for you. Anyone else interested in joining?
If few more are interested , then we can probably come up with few topics to discuss. I was thinking about a regular meetup , like the one Jeremy suggested in his last lecture. Couple of months back I attended a meetup on AI in Bangalore. The quality was poor. It would be great if we can work on building a community here at India who can help others .
That’d be great. We can maybe kick off something new here. Before that, we will have to come up with some good presentations so that we could build a community. How do you suggest we take this forward?
Lets wait till the end of the week . Based on that we will decide when to meet next week. I am comfortable either with a week day or weekend. Looking at the forums , I strongly believe that we do not have a lot many from India who are participating . So we may need to start with something basic and then move on. For next week we will try to have a small meetup within us, who are already a part of fast.ai.
I am In Lesson 1 , duing the execution of
vgg = Vgg16()
# Grab a few images at a time for training and validation.
# NB: They must be in subdirectories named based on their category
batches = vgg.get_batches(path+‘train’, batch_size=batch_size)
val_batches = vgg.get_batches(path+‘valid’, batch_size=batch_size*2)
vgg.finetune(batches)
vgg.fit(batches, val_batches, nb_epoch=1)
The Below Error Occured :-
ValueError: Dimension 0 in both shapes must be equal, but are 3 and 64 for ‘Assign_7’ (op: ‘Assign’) with input shapes: [3,3,224,64], [64,3,3,3].
Hi guys! How about we meet this weekend( Saturday) at Starbucks. I would love for us to discuss applications rather than just discussing papers. I would love to talk about my recent experiments with Tensorflow serving, Caffe2 and floydhub. Maybe discuss a simple prisma clone I built with each of them.
Hi Guys, I am also from Bangalore and recently started with the course. Would be amazing to join and discuss as a group either online or in a meetup.
Please let me know if there is already a group where i can join and take things from there!
Hi ,
I am also from Bangalore and started with the course recently.
Please let me know if there is already a group where i can join and take things from there.
As for set up… were there nay specific tweaks for people accessing aws instances from India…?