TopNode not defined about knet.jl HOT 28 CLOSED

Which version of Julia are you using?

On Fri, Sep 2, 2016 at 5:13 PM Yao Lu [email protected] wrote:

julia> using Knet
WARNING: Base.SparseMatrix is deprecated.
likely near /home/rluser/.julia/v0.5/CUSPARSE/src/CUSPARSE.jl:2
WARNING: Method definition
(::Type{Knet._CudaArray})(CUDArt.CudaArray{#T<:Any, #N<:Any}) in module
Knet at /home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:127 overwritten at
/home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:128.
WARNING: Base.writemime is deprecated.
likely near /home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:133
WARNING: Base.writemime is deprecated.
likely near /home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:133
WARNING: Base.LambdaStaticData is deprecated, use LambdaInfo instead.
likely near /home/rluser/.julia/v0.5/Knet/src/util/deepcopy.jl:21
ERROR: LoadError: LoadError: UndefVarError: TopNode not defined
in include_from_node1(::String) at ./loading.jl:426 (repeats 2 times)
in eval(::Module, ::Any) at ./boot.jl:234
in require(::Symbol) at ./loading.jl:357
while loading /home/rluser/.julia/v0.5/Knet/src/util/deepcopy.jl, in
expression starting on line 21
while loading /home/rluser/.julia/v0.5/Knet/src/Knet.jl, in expression
starting on line 17

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#22, or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpqlN1HPM0TD9fsHUXUo_856JsnMkks5qmC8BgaJpZM4Jzv72
.

from knet.jl.

The current release of Knet is not yet compatible with 0.5. I will release
a compatible version when 0.5 is officially released.

On Fri, Sep 2, 2016 at 6:58 PM Yao Lu [email protected] wrote:

0.5rc3Yao-------- Original Message --------Subject: Re:
[denizyuret/Knet.jl] TopNode not defined (#22)From: denizyuret To:
"denizyuret/Knet.jl" CC: Yao Lu ,Author Which version of Julia are you
using?

On Fri, Sep 2, 2016 at 5:13 PM Yao Lu [email protected] wrote:

julia> using Knet
WARNING: Base.SparseMatrix is deprecated.
likely near /home/rluser/.julia/v0.5/CUSPARSE/src/CUSPARSE.jl:2
WARNING: Method definition
(::Type{Knet._CudaArray})(CUDArt.CudaArray{#T<:Any, #N<:Any}) in module
Knet at /home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:127 overwritten
at
/home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:128.
WARNING: Base.writemime is deprecated.
likely near /home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:133
WARNING: Base.writemime is deprecated.
likely near /home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:133
WARNING: Base.LambdaStaticData is deprecated, use LambdaInfo instead.
likely near /home/rluser/.julia/v0.5/Knet/src/util/deepcopy.jl:21
ERROR: LoadError: LoadError: UndefVarError: TopNode not defined
in include_from_node1(::String) at ./loading.jl:426 (repeats 2 times)
in eval(::Module, ::Any) at ./boot.jl:234
in require(::Symbol) at ./loading.jl:357
while loading /home/rluser/.julia/v0.5/Knet/src/util/deepcopy.jl, in
expression starting on line 21
while loading /home/rluser/.julia/v0.5/Knet/src/Knet.jl, in expression
starting on line 17

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#22, or mute the thread
<
https://github.com/notifications/unsubscribe-auth/ABvNpqlN1HPM0TD9fsHUXUo_856JsnMkks5qmC8BgaJpZM4Jzv72

.

—You are receiving this because you authored the thread.Reply to this
email directly, view it on GitHub, or mute the thread.

{"api_version":"1.0","publisher":{"api_key":"05dde50f1d1a384dd78767c55493e4bb","name":"GitHub"},"entity":{"external_key":"github/denizyuret/Knet.jl","title":"denizyuret/Knet.jl","subtitle":"GitHub
repository","main_image_url":"
https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png
","avatar_image_url":"
https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open
in GitHub","url":"https://github.com/denizyuret/Knet.jl"}},"updates":{"snippets":[{"icon":"PERSON","message":"@denizyuret
in #22: Which version of Julia are you using?\n\nOn Fri, Sep 2, 2016 at
5:13 PM Yao Lu \[email protected]\u003e wrote:\n\n\u003e
julia\u003e using Knet\n\u003e WARNING: Base.SparseMatrix is
deprecated.\n\u003e likely near
/home/rluser/.julia/v0.5/CUSPARSE/src/CUSPARSE.jl:2\n\u003e WARNING: Method
definition\n\u003e (::Type{Knet._CudaArray})(CUDArt.CudaArray{#T\u003c:Any,
#N\u003c:Any}) in module\n\u003e Knet at
/home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:127 overwritten at\n\u003e
/home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:128.\n\u003e WARNING:
Base.writemime is deprecated.\n\u003e likely near
/home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:133\n\u003e WARNING:
Base.writemime is deprecated.\n\u003e likely near
/home/rluser/.julia/v0.5/Knet/src/util/cudart.jl:133\n\u003e WARNING:
Base.LambdaStaticData is deprecated, use LambdaInfo instead.\n\u003e likely
near /home/rluser/.julia/v0.5/Knet/src/util/deepcopy.jl:21\n\u003e ERROR:
LoadError: LoadError: UndefVarError: TopNode not defined\n\u003e in
include_from_node1(::String) at ./loading.jl:426 (repeats 2 times)\n\u003e
in eval(::Module, ::Any) at ./boot.jl:234\n\u003e in require(::Symbol) at
./loading.jl:357\n\u003e while loading
/home/rluser/.julia/v0.5/Knet/src/util/deepcopy.jl, in\n\u003e expression
starting on line 21\n\u003e while loading
/home/rluser/.julia/v0.5/Knet/src/Knet.jl, in expression\n\u003e starting
on line 17\n\u003e\n\u003e —\n\u003e You are receiving this because you are
subscribed to this thread.\n\u003e Reply to this email directly, view it on
GitHub\n\u003e \u003chttps://github.com//issues/22\u003e
http://github.com/denizyuret/Knet.jl/issues/22%5Cu003e, or mute the
thread\n\u003e \u003chttps://
github.com/notifications/unsubscribe-auth/ABvNpqlN1HPM0TD9fsHUXUo_856JsnMkks5qmC8BgaJpZM4Jzv72\u003e\n\u003e
http://github.com/notifications/unsubscribe-auth/ABvNpqlN1HPM0TD9fsHUXUo_856JsnMkks5qmC8BgaJpZM4Jzv72%5Cu003e%5Cn%5Cu003e
.\n\u003e\n"}],"action":{"name":"View Issue","url":"
https://github.com/denizyuret/Knet.jl/issues/22#issuecomment-244414255"}}}

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpofldC0IeYd9PwVdIj8Ql7US8Lb8ks5qmEeegaJpZM4Jzv72
.

from knet.jl.

Hi,

We are testing the Julia port in PowerPC machines. Unfortunately, Julia compiles only in 0.5 Julia. Viral is asking us to test the port and I'm interested to test the GPU support of Julia using deep learning package like Knet. I have similar error above during Knet's building the CUDArt. It will be cool if Knet runs in PowerPC:

using Knet
[29/110]
INFO: Precompiling module CUDArt.
WARNING: @windows is deprecated, use @static is_windows() instead
in depwarn(::String, ::Symbol) at ./deprecated.jl:64
in @windows(::Any, ::Any) at ./deprecated.jl:472
in include_from_node1(::String) at ./loading.jl:481 (repeats 2 times)
in macro expansion; at ./none:2 [inlined]
in anonymous at ./:?
in eval(::Module, ::Any) at ./boot.jl:234
in process_options(::Base.JLOptions) at ./client.jl:239
in _start() at ./client.jl:318
while loading /localhome/paulito/.julia/v0.6/CUDArt/src/libcudart-6.5.jl, in expression starting on line 23
WARNING: Base.WORD_SIZE is deprecated.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/libcudart-6.5.jl:36
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.SparseMatrix is deprecated.
likely near /localhome/paulito/.julia/v0.6/CUSPARSE/src/CUSPARSE.jl:2
ERROR: LoadError: LoadError: UndefVarError: TopNode not defined
in include_from_node1(::String) at ./loading.jl:481 (repeats 2 times)
in eval(::Module, ::Any) at ./boot.jl:234
in require(::Symbol) at ./loading.jl:409
while loading /localhome/paulito/.julia/v0.6/Knet/src/util/deepcopy.jl, in expression starting on line 21
while loading /localhome/paulito/.julia/v0.6/Knet/src/Knet.jl, in expression starting on line 17

from knet.jl.

I will take a look...
On Fri, Sep 2, 2016 at 7:31 PM Paulito Palmes [email protected]
wrote:

Hi,

We are testing the Julia port in PowerPC machines. Unfortunately, Julia
compiles only in 0.5 Julia. Viral is asking us to test the port and I'm
interested to test the GPU support of Julia using deep learning package
like Knet. I have similar error above during Knet's building the CUDArt. It
will be cool if Knet runs in PowerPC:

using Knet

[29/110]
INFO: Precompiling module CUDArt.
WARNING: @windows is deprecated, use @static is_windows() instead
in depwarn(::String, ::Symbol) at ./deprecated.jl:64
in @windows(::Any, ::Any) at ./deprecated.jl:472
in include_from_node1(::String) at ./loading.jl:481 (repeats 2 times)
in macro expansion; at ./none:2 [inlined]
in anonymous at ./:?

in eval(::Module, ::Any) at ./boot.jl:234

in process_options(::Base.JLOptions) at ./client.jl:239
in

_start() at ./client.jl:318 while loading
/localhome/paulito/.julia/v0.6/CUDArt/src/libcudart-6.5.jl, in expression
starting on line 23 WARNING: Base.WORD_SIZE is deprecated. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/libcudart-6.5.jl:36 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 in precompile at
/localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.SparseMatrix is deprecated.
likely near /localhome/paulito/.julia/v0.6/CUSPARSE/src/CUSPARSE.jl:2

ERROR: LoadError: LoadError: UndefVarError: TopNode not defined

in include_from_node1(::String) at ./loading.jl:481 (repeats 2 times)

in eval(::Module, ::Any) at ./boot.jl:234

in require(::Symbol) at ./loading.jl:409
while loading /localhome/paulito/.julia/v0.6/Knet/src/util/deepcopy.jl, in
expression starting on line 21
while loading /localhome/paulito/.julia/v0.6/Knet/src/Knet.jl, in
expression starting on line 17

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNptuhei9wj4b_Dg6KmdaQCF14liqLks5qmE9BgaJpZM4Jzv72
.

from knet.jl.

I switched the latest master to 0.8 development. You should be able to
test it on 0.5 Julia:

julia> Pkg.checkout("Knet")
julia> using Knet
julia> include(Pkg.dir("Knet/examples/mnist.jl"))
julia> MNIST.main() # trains on cpu, gpu is not activated at initialization
julia> gpu(true)
julia> MNIST.main() # trains on gpu
julia> MNIST.main("--help") # prints out the options

Let me know if this solves your problem.

Knet8 is under development, the documentation and updates are not up to
date, but the code should work.

best,
deniz

On Sat, Sep 3, 2016 at 11:00 PM Deniz Yuret [email protected] wrote:

I will take a look...
On Fri, Sep 2, 2016 at 7:31 PM Paulito Palmes [email protected]
wrote:

Hi,

We are testing the Julia port in PowerPC machines. Unfortunately, Julia
compiles only in 0.5 Julia. Viral is asking us to test the port and I'm
interested to test the GPU support of Julia using deep learning package
like Knet. I have similar error above during Knet's building the CUDArt. It
will be cool if Knet runs in PowerPC:

using Knet

[29/110]
INFO: Precompiling module CUDArt.
WARNING: @windows is deprecated, use @static is_windows() instead
in depwarn(::String, ::Symbol) at ./deprecated.jl:64
in @windows(::Any, ::Any) at ./deprecated.jl:472
in include_from_node1(::String) at ./loading.jl:481 (repeats 2 times)
in macro expansion; at ./none:2 [inlined]
in anonymous at ./:?

in eval(::Module, ::Any) at ./boot.jl:234

in process_options(::Base.JLOptions) at ./client.jl:239
in

_start() at ./client.jl:318 while loading
/localhome/paulito/.julia/v0.6/CUDArt/src/libcudart-6.5.jl, in expression
starting on line 23 WARNING: Base.WORD_SIZE is deprecated. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/libcudart-6.5.jl:36 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.UTF8String is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 WARNING:
Base.ASCIIString is deprecated, use String instead. likely near
/localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57 in precompile
at /localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at
/localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at
/localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.UTF8String is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at
/localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at
/localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.ASCIIString is deprecated, use String instead.
likely near /localhome/paulito/.julia/v0.6/CUDArt/src/CUDArt.jl:57
in precompile at
/localhome/paulito/.julia/v0.6/CUDArt/src/precompile.jl
WARNING: Base.SparseMatrix is deprecated.
likely near /localhome/paulito/.julia/v0.6/CUSPARSE/src/CUSPARSE.jl:2

ERROR: LoadError: LoadError: UndefVarError: TopNode not defined

in include_from_node1(::String) at ./loading.jl:481 (repeats 2 times)

in eval(::Module, ::Any) at ./boot.jl:234

in require(::Symbol) at ./loading.jl:409
while loading /localhome/paulito/.julia/v0.6/Knet/src/util/deepcopy.jl,
in expression starting on line 21
while loading /localhome/paulito/.julia/v0.6/Knet/src/Knet.jl, in
expression starting on line 17

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNptuhei9wj4b_Dg6KmdaQCF14liqLks5qmE9BgaJpZM4Jzv72
.

from knet.jl.

Thanks.

I tried running in 0.5 but CUDArt has issues. Viral sent us 0.6.0 for testing.

I did Pkg.build("Knet") after checkout and it compiled successfully in 0.6 Julia.

gpu(true) also worked but running MNIST.main() result to errors: it may indicate I'll convert Float32 to Float64?

opts=(:seed,-1)(:batchsize,100)(:hidden,Int64[])(:epochs,10)(:lr,0.5)(:atype,"KnetArray{Float32}")(:winit,0.1)(:fast,false)
ERROR: ccall: could not find function mul_32_01 in library
in .*(::Float32, ::Knet.KnetArray{Float32,2}) at /localhome/paulito/.julia/v0.6/Knet/src/cuda01.jl:62
in ./(::Knet.KnetArray{Float32,2}, ::Float64) at /localhome/paulito/.julia/v0.6/Knet/src/cuda01.jl:32
in (::MNIST.#xbatch#6{DataType,Int64,Float64})(::Array{UInt8,1}) at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:110
in #minibatch#5(::Type{T}, ::Type{T}, ::Int64, ::Int64, ::Float64, ::Function, ::Array{UInt8,1}, ::Array{UInt8,1}, ::Int64) at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:117
in (::MNIST.#kw##minibatch)(::Array{Any,1}, ::MNIST.#minibatch, ::Array{UInt8,1}, ::Array{UInt8,1}, ::Int64) at ./:0
in main(::Array{String,1}) at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:43
in main() at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:20

from knet.jl.

It does not look like it is able to locate the libknet8 library. Is it in
the Pkg.dir("Knet/src") directory? What is the value of the const libknet8?

deniz

On Tue, Sep 6, 2016 at 4:55 PM Paulito Palmes [email protected]
wrote:

Thanks.

I tried running in 0.5 but CUDArt has issues. Viral sent us 0.6.0 for
testing.

I did Pkg.build("Knet") after checkout and it compiled successfully in 0.6
Julia.

gpu(true) also worked but running MNIST.main() result to errors: it may
indicate I'll convert Float32 to Float64?

opts=(:seed,-1)(:batchsize,100)(:hidden,Int64[])(:epochs,10)(:lr,0.5)(:atype,"KnetArray{Float32}")(:winit,0.1)(:fast,false)
ERROR: ccall: could not find function mul_32_01 in library
in .*(::Float32, ::Knet.KnetArray{Float32,2}) at
/localhome/paulito/.julia/v0.6/Knet/src/cuda01.jl:62
in ./(::Knet.KnetArray{Float32,2}, ::Float64) at
/localhome/paulito/.julia/v0.6/Knet/src/cuda01.jl:32
in (::MNIST.#xbatch#6{DataType,Int64,Float64})(::Array{UInt8,1}) at
/localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:110
in #minibatch#5(::Type{T}, ::Type{T}, ::Int64, ::Int64, ::Float64,
::Function, ::Array{UInt8,1}, ::Array{UInt8,1}, ::Int64) at
/localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:117
in (::MNIST.#kw##minibatch)(::Array{Any,1}, ::MNIST.#minibatch,
::Array{UInt8,1}, ::Array{UInt8,1}, ::Int64) at ./:0
in main(::Array{String,1}) at
/localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:43
in main() at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:20

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNptR581M4N8RK7q3hdf5Q-Po7l7VOks5qnXDNgaJpZM4Jzv72
.

from knet.jl.

yes, libknet8.so and libknet.so are in the Pkg.dir("Knet/src")

the value of Knet.libknet8 is ""

from knet.jl.

sorry, still not working.

I thought compiling files in src7 will help but it did not work. it compiled successfully but MNIST.main() still has similar errors described above.

from knet.jl.

src7 is deprecated, I am only keeping it around temporarily, no need to
compile there...

On Tue, Sep 6, 2016 at 5:18 PM Paulito Palmes [email protected]
wrote:

ok, i got it working now.

i have to compile (using make) manually in the Knet/src and Knet/src7. I
think the Pkg.build("Knet") lacks the command to compile src7 directory.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNplnpvHchvybZyvmdQ7g4-fjkXOAPks5qnXYOgaJpZM4Jzv72
.

from knet.jl.

is there a manual way to make sure libknet8 will be available to Knet Julia? the libs are there so I think we just have to export the location?

from knet.jl.

Normally the info under Knet/deps should cause the compilation libknet8
automatically when you install and when you run Pkg.build("Knet"). Then
the package looks for it under Pkg.dir("Knet/src") -- see the first line of
"src/gpu.jl". Maybe your Knet is not at a standard location?

On Tue, Sep 6, 2016 at 5:29 PM Paulito Palmes [email protected]
wrote:

is there a manual way to make sure libknet8 will be available to Knet
Julia? the libs are there so I think we just have to export the location?

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpkl5Pojk7M9Ofy05FcT0iq9QfOthks5qnXjdgaJpZM4Jzv72
.

from knet.jl.

i just followed the steps:

Pkg.checkout("Knet")
Pkg.build("Knet")
and the location of Knet is: /localhome/paulito/.julia/v0.6/Knet
which I think is correct.

from knet.jl.

Hmm, I should test with 0.6 as well. Is there a downloadable version
somewhere or did you compile?

On Tue, Sep 6, 2016 at 5:47 PM Paulito Palmes [email protected]
wrote:

i just followed the steps:

Pkg.checkout("Knet")
Pkg.build("Knet")
and the location of Knet is: /localhome/paulito/.julia/v0.6/Knet
which I think is correct.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNptHQhGeGTxz1dgrpGRnQw5Sv3_Gsks5qnX0dgaJpZM4Jzv72
.

from knet.jl.

let me send the link. viral sent us the link but i'm afraid he sent us binary version for powerpc machines. let me check the email...

from knet.jl.

here's a snapshot of the email from Viral. I think the first link is the binary version for ppc while the second link is the source code.

Can you try this one? I believe this error is addressed now.
http://s3.amazonaws.com/julianightlies/bin/linux/ppc64le/julia-latest-linuxppc64.tar.gz

Also, Elliot enabled the nightlies:
https://build.julialang.org/builders/package_tarballppc64le

from knet.jl.

One of my students had the same problem -- apparently the precompiled
Knet.ji file is the culprit. If you remove that file it recompiles it with
the right path and it should work.

On Tue, Sep 6, 2016 at 6:45 PM Paulito Palmes [email protected]
wrote:

here's a snapshot of the email from Viral. I think the first link is the
binary version for ppc while the second link is the source code.

Can you try this one? I believe this error is addressed now.

http://s3.amazonaws.com/julianightlies/bin/linux/ppc64le/julia-latest-linuxppc64.tar.gz

Also, Elliot enabled the nightlies:
https://build.julialang.org/builders/package_tarballppc64le

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpjGEk6lNg_ZHUv_lE2_rHgwTMkb9ks5qnYqNgaJpZM4Jzv72
.

from knet.jl.

thanks.

I removed Knet.ji but got different error:

julia> using Knet
INFO: Precompiling module Knet.
julia> include(Pkg.dir("Knet/examples/mnist.jl"))
INFO: Loading MNIST...
MNIST
julia> gpu(true)
0
julia> MNIST.main()
opts=(:seed,-1)(:batchsize,100)(:hidden,Int64[])(:epochs,10)(:lr,0.5)(:atype,"KnetArray{Float32}")(:winit,0.1)(:fast,false)
(:epoch,0,:trn,0.06503333f0,:tst,0.0552f0)
ERROR: MethodError: no method matching tofloat(::Knet.KnetArray{Float32,2})
Closest candidates are:
tofloat(::Number) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:544
tofloat{T<:Number}(::AbstractArray{T<:Number,N}) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:545
tofloat(::Tuple) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:547
...
in map(::AutoGrad.#tofloat, ::Array{Any,1}) at ./essentials.jl:124
in tofloat(::Array{Any,1}) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:549
in forward_pass(::Function, ::Tuple{Array{Any,1},Knet.KnetArray{Float32,2},Knet.KnetArray{Float32,2}}, ::Array{Any,1}, ::Int64) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:68
in (::AutoGrad.##gradfun#1#3{MNIST.#loss,Int64})(::Array{Any,1}, ::Function, ::Array{Any,1}, ::Vararg{Any,N}) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:37
in (::AutoGrad.#gradfun#2)(::Array{Any,1}, ::Vararg{Any,N}) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:37
in #train#3(::Float64, ::Int64, ::Function, ::Array{Any,1}, ::Array{Any,1}) at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:79
in (::MNIST.#kw##train)(::Array{Any,1}, ::MNIST.#train, ::Array{Any,1}, ::Array{Any,1}) at ./:0
in macro expansion at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:51 [inlined]
in macro expansion at ./util.jl:184 [inlined]
in main(::Array{String,1}) at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:50
in main() at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:20

from knet.jl.

Can you do a Pkg.update(), I don't think you have the latest AutoGrad
version. (Should use v0.0.2).

On Tue, Sep 6, 2016 at 8:43 PM Paulito Palmes [email protected]
wrote:

thanks.

I removed Knet.ji but got different error:

julia> using Knet
INFO: Precompiling module Knet.
julia> include(Pkg.dir("Knet/examples/mnist.jl"))
INFO: Loading MNIST...
MNIST
julia> gpu(true)
0
julia> MNIST.main()

opts=(:seed,-1)(:batchsize,100)(:hidden,Int64[])(:epochs,10)(:lr,0.5)(:atype,"KnetArray{Float32}")(:winit,0.1)(:fast,false)

(:epoch,0,:trn,0.06503333f0,:tst,0.0552f0)
ERROR: MethodError: no method matching tofloat(::Knet.KnetArray{Float32,2})
Closest candidates are:
tofloat(::Number) at
/localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:544
tofloat{T<:Number}(::AbstractArray{T<:Number,N}) at
/localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:545
tofloat(::Tuple) at /localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:547
...
in map(::AutoGrad.#tofloat, ::Array{Any,1}) at ./essentials.jl:124
in tofloat(::Array{Any,1}) at
/localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:549
in forward_pass(::Function,
::Tuple{Array{Any,1},Knet.KnetArray{Float32,2},Knet.KnetArray{Float32,2}},
::Array{Any,1}, ::Int64) at
/localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:68
in (::AutoGrad.##gradfun#1#3{MNIST.#loss,Int64})(::Array{Any,1},
::Function, ::Array{Any,1}, ::Vararg{Any,N}) at
/localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:37
in (::AutoGrad.#gradfun#2)(::Array{Any,1}, ::Vararg{Any,N}) at
/localhome/paulito/.julia/v0.6/AutoGrad/src/core.jl:37
in #train#3(::Float64, ::Int64, ::Function, ::Array{Any,1},
::Array{Any,1}) at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:79
in (::MNIST.#kw##train)(::Array{Any,1}, ::MNIST.#train, ::Array{Any,1},
::Array{Any,1}) at ./:0
in macro expansion at
/localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:51 [inlined]
in macro expansion at ./util.jl:184 [inlined]
in main(::Array{String,1}) at
/localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:50

in main() at /localhome/paulito/.julia/v0.6/Knet/examples/mnist.jl:20

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpjL-yOj9LA8volWhuqcFgZfXtTedks5qnaYlgaJpZM4Jzv72
.

from knet.jl.

MNIST.main() with gpu(true) works now with the latest AutoGrad.

I timed gpu(false) vs gpu(true) for MNIST.main().

They finished both around 3 to 4 seconds. It feels no difference between the two. Even the printing of results look similar speed.

By the way, latest Knet no longer uses compile and other macros. Is there a documentation for the latest Knet?

Thanks.

from knet.jl.

I am writing the documentation as we speak!!!

In summary, Knet8 does not compile models any more, it uses AutoGrad to
differentiate ordinary Julia code!

When I test with MNIST.main("--fast") I get 2.82 for the cpu and 2.12 for
the gpu. (--fast disables the loss printing and gets a better estimate of
real training cost). This is 10 epochs on a softmax model. To see some
more difference you can try MNIST.main("--fast --hidden 64") for a single
hidden layer with 64 units. I get 6.23 with the cpu and 2.96 with the gpu.

best,
deniz

On Tue, Sep 6, 2016 at 9:04 PM Paulito Palmes [email protected]
wrote:

MNIST.main() with gpu(true) works now with the latest AutoGrad.

I timed gpu(false) vs gpu(true) for MNIST.main().

They finished both around 3 to 4 seconds. It feels no difference between
the two. Even the printing of results look similar speed.

By the way, latest Knet no longer uses compile and other macros. Is there
a documentation for the latest Knet?

Thanks.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpjUG5gJcqI2RRT8Cp7jS7sk5eJuCks5qnasmgaJpZM4Jzv72
.

from knet.jl.

thanks for the explanation.

I'm just trying to make sure that GPU is really working ;).

it is great to have Knet GPU working now in Power. So far, this is the only deep learning package easy to install relative to other deep learning packages I tried and I very much prefer the programming logic of Knet. the documentation of the previous version is great and really looking forward to use Knet in my workflow. thanks for the great package!!!

Sent from my iPhone

On 6 Sep 2016, at 19:14, denizyuret [email protected] wrote:

I am writing the documentation as we speak!!!

In summary, Knet8 does not compile models any more, it uses AutoGrad to
differentiate ordinary Julia code!

When I test with MNIST.main("--fast") I get 2.82 for the cpu and 2.12 for
the gpu. (--fast disables the loss printing and gets a better estimate of
real training cost). This is 10 epochs on a softmax model. To see some
more difference you can try MNIST.main("--fast --hidden 64") for a single
hidden layer with 64 units. I get 6.23 with the cpu and 2.96 with the gpu.

best,
deniz

On Tue, Sep 6, 2016 at 9:04 PM Paulito Palmes [email protected]
wrote:

MNIST.main() with gpu(true) works now with the latest AutoGrad.

I timed gpu(false) vs gpu(true) for MNIST.main().

They finished both around 3 to 4 seconds. It feels no difference between
the two. Even the printing of results look similar speed.

By the way, latest Knet no longer uses compile and other macros. Is there
a documentation for the latest Knet?

Thanks.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpjUG5gJcqI2RRT8Cp7jS7sk5eJuCks5qnasmgaJpZM4Jzv72
.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or mute the thread.

from knet.jl.

You are very welcome. You can sign up on the knet-users mailing list (very
low volume) to get updates as I try to finish v0.8.

BTW, I couldn't compile Julia 6 on the intel servers I have (Centos 6.5)
and the nightlies seem to be built for a different architecture. Let me
know if you come across an image that might work.

deniz

On Tue, Sep 6, 2016 at 11:40 PM Paulito Palmes [email protected]
wrote:

thanks for the explanation.

I'm just trying to make sure that GPU is really working ;).

it is great to have Knet GPU working now in Power. So far, this is the
only deep learning package easy to install relative to other deep learning
packages I tried and I very much prefer the programming logic of Knet. the
documentation of the previous version is great and really looking forward
to use Knet in my workflow. thanks for the great package!!!

Sent from my iPhone

On 6 Sep 2016, at 19:14, denizyuret [email protected] wrote:

I am writing the documentation as we speak!!!

In summary, Knet8 does not compile models any more, it uses AutoGrad to
differentiate ordinary Julia code!

When I test with MNIST.main("--fast") I get 2.82 for the cpu and 2.12 for
the gpu. (--fast disables the loss printing and gets a better estimate of
real training cost). This is 10 epochs on a softmax model. To see some
more difference you can try MNIST.main("--fast --hidden 64") for a single
hidden layer with 64 units. I get 6.23 with the cpu and 2.96 with the
gpu.

best,
deniz

On Tue, Sep 6, 2016 at 9:04 PM Paulito Palmes [email protected]
wrote:

MNIST.main() with gpu(true) works now with the latest AutoGrad.

I timed gpu(false) vs gpu(true) for MNIST.main().

They finished both around 3 to 4 seconds. It feels no difference
between
the two. Even the printing of results look similar speed.

By the way, latest Knet no longer uses compile and other macros. Is
there
a documentation for the latest Knet?

Thanks.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
<
https://github.com/denizyuret/Knet.jl/issues/22#issuecomment-245036837>,
or mute the thread
<
https://github.com/notifications/unsubscribe-auth/ABvNpjUG5gJcqI2RRT8Cp7jS7sk5eJuCks5qnasmgaJpZM4Jzv72

.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or mute the thread.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpvyB-en39tZsI7wTz-CtsbLv6r8dks5qnc-6gaJpZM4Jzv72
.

from knet.jl.

I'll ask Viral if he has Intel binaries also.

I like to contribute someday on the aspects of running Knet in PowerPC machines. I like the simplicity of Knet which is true to the KISS principles of UNIX ;)

Sent from my iPhone

On 7 Sep 2016, at 10:40, denizyuret [email protected] wrote:

You are very welcome. You can sign up on the knet-users mailing list (very
low volume) to get updates as I try to finish v0.8.

BTW, I couldn't compile Julia 6 on the intel servers I have (Centos 6.5)
and the nightlies seem to be built for a different architecture. Let me
know if you come across an image that might work.

deniz

On Tue, Sep 6, 2016 at 11:40 PM Paulito Palmes [email protected]
wrote:

thanks for the explanation.

I'm just trying to make sure that GPU is really working ;).

it is great to have Knet GPU working now in Power. So far, this is the
only deep learning package easy to install relative to other deep learning
packages I tried and I very much prefer the programming logic of Knet. the
documentation of the previous version is great and really looking forward
to use Knet in my workflow. thanks for the great package!!!

Sent from my iPhone

On 6 Sep 2016, at 19:14, denizyuret [email protected] wrote:

I am writing the documentation as we speak!!!

In summary, Knet8 does not compile models any more, it uses AutoGrad to
differentiate ordinary Julia code!

When I test with MNIST.main("--fast") I get 2.82 for the cpu and 2.12 for
the gpu. (--fast disables the loss printing and gets a better estimate of
real training cost). This is 10 epochs on a softmax model. To see some
more difference you can try MNIST.main("--fast --hidden 64") for a single
hidden layer with 64 units. I get 6.23 with the cpu and 2.96 with the
gpu.

best,
deniz

On Tue, Sep 6, 2016 at 9:04 PM Paulito Palmes [email protected]
wrote:

MNIST.main() with gpu(true) works now with the latest AutoGrad.

I timed gpu(false) vs gpu(true) for MNIST.main().

They finished both around 3 to 4 seconds. It feels no difference
between
the two. Even the printing of results look similar speed.

By the way, latest Knet no longer uses compile and other macros. Is
there
a documentation for the latest Knet?

Thanks.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
<
https://github.com/denizyuret/Knet.jl/issues/22#issuecomment-245036837>,
or mute the thread
<
https://github.com/notifications/unsubscribe-auth/ABvNpjUG5gJcqI2RRT8Cp7jS7sk5eJuCks5qnasmgaJpZM4Jzv72

.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or mute the thread.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpvyB-en39tZsI7wTz-CtsbLv6r8dks5qnc-6gaJpZM4Jzv72
.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or mute the thread.

from knet.jl.

by the way, the latest version of Knet's REQUIRE commented out CUDArt, ArgParse, and GZip which are needed to run MNIST.main()

from knet.jl.

Yes, I decided to minimize the requirements to run everything under src and
write your own models. I will add something to the documentation of
additional requirements to run examples, or use the gpu.

On Wed, Sep 7, 2016 at 5:10 PM Paulito Palmes [email protected]
wrote:

by the way, the latest version REQUIRE commented out CUDArt, ArgParse, and
GZip which are needed to run MNIST.main()

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNpm3ofeRjOckzVqfomQ7d1y--KmcPks5qnsXMgaJpZM4Jzv72
.

from knet.jl.

@denizyuret Thank you! Knet8 is usable for me.
Will knet7 and knet8 be developed in parallel or knet8 will be a successor of knet7? If it's the latter case, can AutoGrad emulate the performance of knet7 on gpu?

from knet.jl.

knet7 will be maintained with bugfixes, future work will focus on knet8.
I am working on knet8 performance on the gpu, hope to make it better than
knet7.

On Thu, Sep 8, 2016, 07:18 Yao Lu [email protected] wrote:

@denizyuret https://github.com/denizyuret Thank you! Knet8 is usable
for me.
Will knet7 and knet8 be developed in parallel or knet8 will be a successor
of knet7? If it's the latter case, can AutoGrad emulate the performance of
knet7 on gpu?

—
You are receiving this because you were mentioned.

Reply to this email directly, view it on GitHub
#22 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/ABvNphUWUxAem4JNcS4j9sMK7uiPjSZUks5qn4x4gaJpZM4Jzv72
.

from knet.jl.