hasktorch: Torch for tensors and neural networks in Haskell

[ ai, bsd3, library, machine-learning, program, tensors ] [ Propose Tags ]

Hasktorch is a library for tensors and neural networks in Haskell. It is an independent open source community project which leverages the core C libraries shared by Torch and PyTorch. This library leverages cabal v2-build and backpack. *Note that this project is in early development and should only be used by contributing developers. Expect substantial changes to the library API as it evolves. Contributions and PRs are welcome (see details on github).*

Versions 0.0.1.0
Dependencies base (>=4.7 && <5), dimensions (>=1.0), hasktorch, hasktorch-cpu, hasktorch-ffi-th (==0.0.1.*), hasktorch-gpu, hasktorch-types-th (==0.0.1.*), safe-exceptions (>=0.1.0), singletons (>=2.2), text (>=1.2.2) [details]
License BSD-3-Clause
Author Hasktorch dev team
Maintainer Sam Stites <fnz@fgvgrf.vb>, Austin Huang <nhfgvau@nyhz.zvg.rqh> - cipher:ROT13
Category Tensors, Machine Learning, AI
Home page https://github.com/hasktorch/hasktorch#readme
Bug tracker https://github.com/hasktorch/hasktorch/issues
Source repo head: git clone https://github.com/hasktorch/hasktorch
Uploaded by stites at Fri Nov 2 00:59:02 UTC 2018
Distributions NixOS:0.0.1.0
Executables memcheck, isdefinite, isdefinite-gpu, isdefinite-cpu
Downloads 20 total (20 in the last 30 days)
Rating (no votes yet) [estimated by rule of succession]
Your Rating
  • λ
  • λ
  • λ
Status Docs uploaded by user [build log]
All reported builds failed as of 2018-11-02 [all 2 reports]
Hackage Matrix CI

Modules

[Index] [Quick Jump]

  • Torch
    • Torch.Byte
      • Torch.Byte.Dynamic
      • Torch.Byte.Storage
    • Torch.Char
      • Torch.Char.Dynamic
      • Torch.Char.Storage
    • Core
    • Cuda
      • Torch.Cuda.Byte
        • Torch.Cuda.Byte.Dynamic
        • Torch.Cuda.Byte.Storage
      • Torch.Cuda.Char
        • Torch.Cuda.Char.Dynamic
        • Torch.Cuda.Char.Storage
      • Torch.Cuda.Double
        • Torch.Cuda.Double.Dynamic
          • Torch.Cuda.Double.Dynamic.NN
            • Torch.Cuda.Double.Dynamic.NN.Activation
            • Torch.Cuda.Double.Dynamic.NN.Criterion
            • Torch.Cuda.Double.Dynamic.NN.Pooling
        • Torch.Cuda.Double.NN
          • Torch.Cuda.Double.NN.Activation
          • Torch.Cuda.Double.NN.Backprop
          • Torch.Cuda.Double.NN.Conv1d
          • Torch.Cuda.Double.NN.Conv2d
          • Torch.Cuda.Double.NN.Criterion
          • Torch.Cuda.Double.NN.Layers
          • Torch.Cuda.Double.NN.Linear
          • Torch.Cuda.Double.NN.Math
          • Torch.Cuda.Double.NN.Padding
          • Torch.Cuda.Double.NN.Pooling
          • Torch.Cuda.Double.NN.Sampling
        • Torch.Cuda.Double.Storage
      • Torch.Cuda.Float
        • Torch.Cuda.Float.Dynamic
        • Torch.Cuda.Float.Storage
      • Torch.Cuda.Int
        • Torch.Cuda.Int.Dynamic
        • Torch.Cuda.Int.Storage
      • Torch.Cuda.Long
        • Torch.Cuda.Long.Dynamic
        • Torch.Cuda.Long.Storage
      • Torch.Cuda.Short
        • Torch.Cuda.Short.Dynamic
        • Torch.Cuda.Short.Storage
    • Torch.Double
      • Torch.Double.Dynamic
        • Torch.Double.Dynamic.NN
          • Torch.Double.Dynamic.NN.Activation
          • Torch.Double.Dynamic.NN.Criterion
          • Torch.Double.Dynamic.NN.Pooling
      • Torch.Double.NN
        • Torch.Double.NN.Activation
        • Torch.Double.NN.Backprop
        • Torch.Double.NN.Conv1d
        • Torch.Double.NN.Conv2d
        • Torch.Double.NN.Criterion
        • Torch.Double.NN.Layers
        • Torch.Double.NN.Linear
        • Torch.Double.NN.Math
        • Torch.Double.NN.Padding
        • Torch.Double.NN.Pooling
        • Torch.Double.NN.Sampling
      • Torch.Double.Storage
    • Torch.Float
      • Torch.Float.Dynamic
      • Torch.Float.Storage
    • Torch.Int
      • Torch.Int.Dynamic
      • Torch.Int.Storage
    • Torch.Long
      • Torch.Long.Dynamic
      • Torch.Long.Storage
    • Torch.Short
      • Torch.Short.Dynamic
      • Torch.Short.Storage
    • Types
      • Torch.Types.Numeric

Flags

NameDescriptionDefaultType
cuda

build with THC support

DisabledAutomatic
lite

only build with Double and Long support

DisabledAutomatic

Use -f <flag> to enable a flag, or -f -<flag> to disable that flag. More info

Downloads

Maintainer's Corner

For package maintainers and hackage trustees