[theano-users] Re: Compilation error on Windows 8.1 Pro 64-bit

Thanks a lot for the fix!!! On Monday, March 9, 2015 at 8:05:03 AM UTC-4, eseaf...@hotmail.com wrote: > > I had a similar problem. I followed the instructions on the web page >> when >> >> installing (i.e. installed T

Re: [theano-users] which function in theano implement numpy.where?

Thank you Mr. Lamblin, you reply is helpful :) On Wednesday, November 23, 2016 at 8:02:10 AM UTC+8, Pascal Lamblin wrote: > > Hi, > > The only syntax we currently support in tensor.where is the one where > two additional argument are provided, like in: > > >>> numpy.where(a > 3, a, a + 3) > ar

Re: [theano-users] which function in theano implement numpy.where?

Thank you Mr. Pascal, it is very helpful On Wednesday, November 23, 2016 at 8:02:10 AM UTC+8, Pascal Lamblin wrote: > > Hi, > > The only syntax we currently support in tensor.where is the one where > two additional argument are provided, like in: > > >>> numpy.where(a > 3, a, a + 3) > array([3

[theano-users] Re: theano error message: Optimization failure due to: LocalOptGroup(use_c_ger,use_c_gemv) on Mac El Capitan

Well, after switching to the GPU, I no longer get this problem. I'm not sure why, but since I wanted to use a GPU anyway for my real workstation (the laptop was mostly a testbed for small stuff) it's OK. -- --- You received this message because you are subscribed to the Google Groups "theano

Re: [theano-users] Update shared variable after scan

it should not. If this cause problem, it should raise an error. upd isn't a normal python dict, but a special class that will make sure we don't loose updates and give an error if the user would cause that behavior. On Wed, Nov 23, 2016 at 3:13 PM, Shir Gur wrote: > Will it ruin the update dicta

[theano-users] iterate over all shared variables

Hi, I want to iterate over all my shared variables, i.e. by scan. a) is there a way to access them by index in some global attribute of theano? b) is there a way to pack them by tensor-variable (probably no because the have different shapes)? now i iterate them on CPU and execute theano functi

Re: [theano-users] Update shared variable after scan

Will it ruin the update dictation of the scan procedure? On Wednesday, November 23, 2016 at 9:07:53 PM UTC+2, nouiz wrote: > > upd[W] = m > > should work. > > Fred > > On Wed, Nov 23, 2016 at 12:39 PM, Shir Gur > wrote: > >> Hi, >> >> i have my scan function, something like >> >> def _loop(i): >>

Re: [theano-users] Update shared variable after scan

upd[W] = m should work. Fred On Wed, Nov 23, 2016 at 12:39 PM, Shir Gur wrote: > Hi, > > i have my scan function, something like > > def _loop(i): > x = something with shared varibale W > return x, {W:x} > > > res, upd = theano.scan(fn=_loop, sequence=T.arange(10)) > > > so after each

Re: [theano-users] GPU is not available in Theano

Did you reboot? We don't have good and uptodate instruction for GPU on Windows. But with the new gpu back-end, it should get simpler. We aren't sure when it that part will be updated. Otherwise, check theano-user mailing list and theano repo on github and search on the web. There is other people t

[theano-users] Update shared variable after scan

Hi, i have my scan function, something like def _loop(i): x = something with shared varibale W return x, {W:x} res, upd = theano.scan(fn=_loop, sequence=T.arange(10)) so after each iteration the scan function updates *W* to be *x*. Now i have the following: m = T.argmx(res) func =

[theano-users] Update shared variable after scan

Hi, i have my scan function, something like def _loop(i): x = something with shared varibale W return x, {W:x} res, upd = theano.scan(fn=_loop, sequence=T.arange(10)) so after each iteration the scan function updates *W* to be *x*. Now i have the following: m = T.argmx(res) func =

[theano-users] Multiple users access remotely and simultaneously with Nvidia DevBox

Hello everyone, I am wondering if it possible to have multiple users access remotely and simultaneously with Nvidia DevBox? Thanks in advance! Adeel -- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To unsubscribe from this group and stop