Issues filed for torch/torch7

View Full Project
Do you use torch7? Leave a review!

Rate of open issues in the last 60 days

torch7 open issues (View Closed Issues)
  • over 1 year OpenBlas could not be compiled
  • over 1 year torch.sum(x,x,2) not returning correct value
  • over 1 year Error:expected arguments: CudaTensor~2D [CudaTensor2D] [float] CudaTensor2D CudaTensor2D
  • over 1 year Convenience function to add a singleton dimension to a torch tensor.
  • over 1 year not having neon does not imply x86
  • over 1 year Calling 'random' method on CudaTensor fails.
  • over 1 year TYPEStorage when given ptr acts weirds
  • over 1 year Missing documents on several functions in torch class
  • almost 2 years ARM64 NEON unknown register name 'r0' in asm
  • almost 2 years Compilation on Raspberry Pi B+ armv6l
  • almost 2 years can you recommend a IDE for me to debug lua?
  • almost 2 years lots of strange issue on ubuntu16.04, never see it on ubuntu14.04
  • almost 2 years VM Ware Player and UBUNTU / MINT installation error: CAN NOT COMPILE
  • almost 2 years torch.mm doesn't resize the result existing tensor
  • almost 2 years How to run torch on 10 GPU system??
  • almost 2 years MacOS 10.12 , qt fail
  • almost 2 years [Feature request] Can we make index() accept 1xN or Nx1 LongTensors?
  • almost 2 years Hi, what's the difference between the files?
  • almost 2 years ./install-deps does not contemplate linux mint
  • almost 2 years clearState() on gModule with moduleFromCriterion does not behave as expected
  • almost 2 years NEON detection on aarch64 / TX1
  • almost 2 years [Port] Portability problem introduced by THVector.c
  • almost 2 years Handling Variable length sequences in training an RNN
  • almost 2 years illegal memory access was encountered
  • almost 2 years test stuck at GPU()
  • almost 2 years [bug] torch.test() fails when the second time calling
  • almost 2 years ../torch/File.lua:370 table index is nil
  • almost 2 years on linux, could use linuxbrew instead of distribution-specific installs?
  • almost 2 years THREAD FATAL ERROR: could not create lua state
  • almost 2 years Architecture not supported
  • almost 2 years Has `convolve_5x5` been used to compute convolution?
  • about 2 years expose real, sys, user time in torch.Timer
  • about 2 years Luarocks install --local is failing
  • about 2 years After :resize() elements are not contiguous in memory
  • about 2 years gels : Lapack library not found in compile time
  • about 2 years Ubuntu 14.04 lzmq module installation problems
  • about 2 years Installation error: identifier "nullptr" is undefined
  • about 2 years Repeatedly creating threads with tensor computation = interruptible sleep state
  • about 2 years Unexpected result for cmul
  • about 2 years luaT_pushlong undefined
  • about 2 years Bug: illegal hardware instruction (core dumped)
  • about 2 years installation to custom location issues
  • over 2 years Misleading torch.svd documentation
  • over 2 years Call for tutorials
  • over 2 years Speed of Torch across various systems following an almost inverse Moore's law effect
  • over 2 years Win32 getAllocSize exception with null pointer
  • over 2 years CudaTensor:topk might overflow and get -1.2762655537074e+38
  • over 2 years torch.chunk fails to split correctly
  • over 2 years Indexing a tensor with non-integer number causes silent rounding
  • over 2 years Use of int32_t and int64_t instead of int and long
  • over 2 years Pluggable C++ Exception wrapper-chain for LUAJIT_MODE_WRAPCFUNC
  • over 2 years Some of Torch7's native libraries segfault when compiled with musl libc
  • over 2 years potrs ** On entry to DPOTRS parameter number 1 had an illegal value
  • over 2 years Implement Numpy-like tensordot operation?
  • over 2 years torch.index doesn't handle inplace
  • over 2 years Overflow in tensor value assignment causes undefined behaviour
  • over 2 years Memory leak in several functions inTHTensorLapack.c
  • over 2 years #434 breaks multinomial.
  • over 2 years torch.eig problem
  • over 2 years Translation (English to Korean)
  • almost 3 years adding -1 support for torch.reshape
  • almost 3 years THTensor.c takes a loooonnngg time to compile. Would be good to factorize a bit
  • almost 3 years THLapackCheck on windows
  • almost 3 years luarocks --local conflicts with system-wide installation of torch
  • almost 3 years Wrong Tensor initialization with LongStorage causes Segmentation fault
  • almost 3 years QR tests are failing because of tolerance
  • almost 3 years Completion of error handling
  • almost 3 years reserved identifier violation
  • almost 3 years Fast nearest neighbor search?
  • almost 3 years Proposals for projects, in the style of Summer of Code
  • almost 3 years [Roadmap] August 2015 - March 2016
  • almost 3 years Torch does not install anymore with gcc-5 on MacOS
  • almost 3 years Use SSE from THVector
  • about 3 years Problem with pointwise math when output tensor is zero-strided with correct size
  • about 3 years ByteTensor is missing mean and std
  • about 3 years Add versioning
  • about 3 years Small incompatibility between C/Cuda maskedFill / maskedSelect
  • about 3 years torch.svd allocates too large a matrix with 'S'
  • about 3 years tensor smart printing
  • about 3 years catch ctrl-c in the REPL
  • about 3 years who() chokes when large string is present
  • over 3 years catching torch error from coroutine
  • over 3 years storage should have byte_size() function
  • over 3 years LongStorage seg fault
  • over 3 years bugs in multinomial?
  • over 3 years Suggestion: torch.Tensor:copy(tensor, [resizeAs])
  • over 3 years PipeFile should handle SIGPIPE
  • almost 4 years broadcasting
  • over 4 years deserialization fails on android if nans are present

torch7 closed issues

  • over 1 year In Caffe there is EuclideanLoss Layer for regression, what is the corresponding loss in torch?
  • over 1 year Unexpected error
  • over 1 year libreadline fails to load on MacOSX Sierra
  • over 1 year where is the code and data from http://www.clement.farabet.net/?
  • over 1 year luaT interfacing does not know about `torch.FloatTensor` metatable
  • over 1 year Uniformly distributed continuous random numbers in closed interval [0, 1]
  • over 1 year Does torch support atrous convolution layer now?
  • almost 2 years Shared storage allocation error
  • almost 2 years cutorch Error Downloading, Luajit, cuda 7.5
  • almost 2 years Get stack traceback when trying to produce CSV
  • almost 2 years Memory leak vs segfault when pushing Torch Tensor from lua to C
  • almost 2 years Compiling torch on armv7l fails
  • almost 2 years Installation install.sh fails due to many many errors
  • almost 2 years Installation of Torch deps failed on macOS sierra
  • almost 2 years torch.ByteTensor.THNN backend not imported
  • almost 2 years Build failed on ARM with NEON support
  • almost 2 years Error: Build error: Failed building.
  • almost 2 years How to designate tasks to GPUs with torch7?
  • almost 2 years Possible bug in "clone()" and "share()" in the latest torch
  • almost 2 years How to initialize weights?
  • almost 2 years CUDNN_STATUS_ALLOC_FAILED?
  • almost 2 years Multiply-Add
  • almost 2 years I can not do install.sh from 8c44a2a8b3340dc531662efcc02ccc32c5413b06 on JetsonTX1
  • almost 2 years [bug] Tensor operations
  • almost 2 years Unsupported gpu architecture 'compute_61' on luarocks install cutorch
  • almost 2 years Compilation issues on Nvidia TK1
  • almost 2 years Issue regarding installation of torch
  • almost 2 years A very strange problem about indexing cuda tensor.
  • almost 2 years Getting error while installing torch in CENTOS
  • almost 2 years LookupTable.lua:75: bad argument #3 to 'index' (Tensor | LongTensor expected, got torch.CudaLongTensor)