torch7 | main package in Torch7 where data structures | Machine Learning library

by torch C Version: Current License: Non-SPDX

Download this library from

kandi X-RAY | torch7 Summary

torch7 is a C library typically used in Artificial Intelligence, Machine Learning, Deep Learning applications. torch7 has no bugs, it has no vulnerabilities and it has medium support. However torch7 has a Non-SPDX License. You can download it from GitHub.

Torch is the main package in Torch7 where data structures for multi-dimensional tensors and mathematical operations over these are defined. Additionally, it provides many utilities for accessing files, serializing objects of arbitrary types and other useful utilities.

Support

Quality

Security

License

Reuse

kandi-support Support

torch7 has a medium active ecosystem.
It has 8619 star(s) with 2375 fork(s). There are 652 watchers for this library.
It had no major release in the last 12 months.
There are 277 open issues and 442 have been closed. On average issues are closed in 83 days. There are 15 open pull requests and 0 closed requests.
It has a neutral sentiment in the developer community.
The latest version of torch7 is current.

torch7 Support

Best in #Machine Learning

Average in #Machine Learning

torch7 Support

Best in #Machine Learning

Average in #Machine Learning

quality kandi Quality

torch7 has 0 bugs and 0 code smells.

torch7 Quality

Best in #Machine Learning

Average in #Machine Learning

torch7 Quality

Best in #Machine Learning

Average in #Machine Learning

security Security

torch7 has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.
torch7 code analysis shows 0 unresolved vulnerabilities.
There are 0 security hotspots that need review.

torch7 Security

Best in #Machine Learning

Average in #Machine Learning

torch7 Security

Best in #Machine Learning

Average in #Machine Learning

license License

torch7 has a Non-SPDX License.
Non-SPDX licenses can be open source with a non SPDX compliant license, or non open source licenses, and you need to review them closely before use.

torch7 License

Best in #Machine Learning

Average in #Machine Learning

torch7 License

Best in #Machine Learning

Average in #Machine Learning

build Reuse

torch7 releases are not available. You will need to build from source code and install.

torch7 Reuse

Best in #Machine Learning

Average in #Machine Learning

torch7 Reuse

Best in #Machine Learning

Average in #Machine Learning

Top functions reviewed by kandi - BETA

Coming Soon for all Libraries!

Currently covering the most popular Java, JavaScript and Python libraries. See a SAMPLE HERE.
kandi's functional review helps you automatically verify the functionalities of the libraries and avoid rework.

torch7 Key Features

http://torch.ch

torch7 Examples and Code Snippets

See all related Code Snippets

Attempt to call local 'callback' (a nil value) error while trying to train SeGAN model

Copy

Download

netG:apply(weights_init)

local function weights_init(m)
   local name = torch.type(m)
   if name:find('Convolution') then
      m.weight:normal(0.0, 0.02)
      m.bias:fill(0)
   elseif name:find('BatchNormalization') then
      if m.weight then m.weight:normal(1.0, 0.02) end
      if m.bias then m.bias:fill(0) end
   end
end

-----------------------

netG:apply(weights_init)

local function weights_init(m)
   local name = torch.type(m)
   if name:find('Convolution') then
      m.weight:normal(0.0, 0.02)
      m.bias:fill(0)
   elseif name:find('BatchNormalization') then
      if m.weight then m.weight:normal(1.0, 0.02) end
      if m.bias then m.bias:fill(0) end
   end
end

See all related Code Snippets

Community Discussions

Trending Discussions on torch7

Attempt to call local 'callback' (a nil value) error while trying to train SeGAN model
torch.CharStorage doesn't read data from a file when size isn't provided

Trending Discussions on torch7

QUESTION

Attempt to call local 'callback' (a nil value) error while trying to train SeGAN model

Asked 2021-Jan-06 at 11:11

I am trying to implement the "SeGAN: Segmenting and Generating the invisible" paper on ubuntu 18.04 with Geforce RTX 2060. I have installed the Driver, CUDA, cuDNN, Torch7 and dependencies and downloaded and extracted the dataset and weights folders and made a link to them. I tried to train the model with this line of code:

th main.lua -baseLR 1e-3 -end2end -istrain "train"

But I got this error:

define model netG...    
/home/darya/distro/install/bin/luajit: /home/darya/distro/install/share/lua/5.1/nn/Module.lua:352: attempt to call local 'callback' (a nil value)
stack traceback:
    /home/darya/distro/install/share/lua/5.1/nn/Module.lua:352: in function 'apply'
    /home/darya/SeGAN/networks/End2EndNetwork.lua:81: in function 'defineG'
    /home/darya/SeGAN/networks/End2EndNetwork.lua:130: in function 'init_networks'
    /home/darya/SeGAN/networks/End2EndNetwork.lua:343: in main chunk
    [C]: in function 'dofile'
    main.lua:265: in main chunk
    [C]: in function 'dofile'
    ...rya/distro/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
    [C]: at 0x55e8fde0f570

I don't know what am I doing wrong and since I don't have much experience I am confused. Can someone help me to fix this problem, please? I appreciate it so much.

github repository for SeGAN segmenting and generating the invisible

ANSWER

Answered 2021-Jan-06 at 11:11

From the linked GitHub repo:

End2EndNetwork.lua line 81

netG:apply(weights_init)

weights_init is supposed to be a function value but is nil.

Which in Module:apply causes callback(self) in Torch nn's Module.lua Line352 to fail.

Looks like someone has to implement that function.

Like in https://github.com/phillipi/pix2pix/blob/89ff2a81ce441fbe1f1b13eca463b87f1e539df8/train.lua, which https://github.com/ehsanik/SeGAN/blob/d29e0a5ac08f093b87dc82af3707d50107900d04/networks/End2EndNetwork.lua#L73 refers to.

where

local function weights_init(m)
   local name = torch.type(m)
   if name:find('Convolution') then
      m.weight:normal(0.0, 0.02)
      m.bias:fill(0)
   elseif name:find('BatchNormalization') then
      if m.weight then m.weight:normal(1.0, 0.02) end
      if m.bias then m.bias:fill(0) end
   end
end

Source https://stackoverflow.com/questions/65594137