Awesome
Torch.rb
:fire: Deep learning for Ruby, powered by LibTorch
Check out:
- TorchVision for computer vision tasks
- TorchText for text and NLP tasks
- TorchAudio for audio tasks
- TorchRec for recommendation systems
- TorchData for data loading
Installation
First, download LibTorch. For Mac arm64, use:
curl -L https://download.pytorch.org/libtorch/cpu/libtorch-macos-arm64-2.5.1.zip > libtorch.zip
unzip -q libtorch.zip
For Linux x86-64, use the cxx11 ABI
version. For other platforms, build LibTorch from source.
Then run:
bundle config build.torch-rb --with-torch-dir=/path/to/libtorch
And add this line to your application’s Gemfile:
gem "torch-rb"
It can take 5-10 minutes to compile the extension. Windows is not currently supported.
Getting Started
A good place to start is Deep Learning with Torch.rb: A 60 Minute Blitz.
Tutorials
Examples
- Image classification with MNIST (日本語版)
- Collaborative filtering with MovieLens
- Generative adversarial networks
API
This library follows the PyTorch API. There are a few changes to make it more Ruby-like:
- Methods that perform in-place modifications end with
!
instead of_
(add!
instead ofadd_
) - Methods that return booleans use
?
instead ofis_
(tensor?
instead ofis_tensor
) - Numo is used instead of NumPy (
x.numo
instead ofx.numpy()
)
You can follow PyTorch tutorials and convert the code to Ruby in many cases. Feel free to open an issue if you run into problems.
Overview
Some examples below are from Deep Learning with PyTorch: A 60 Minutes Blitz
Tensors
Create a tensor from a Ruby array
x = Torch.tensor([[1, 2, 3], [4, 5, 6]])
Get the shape of a tensor
x.shape
There are many functions to create tensors, like
a = Torch.rand(3)
b = Torch.zeros(2, 3)
Each tensor has four properties
dtype
- the data type -:uint8
,:int8
,:int16
,:int32
,:int64
,:float32
,:float64
, or:bool
layout
-:strided
(dense) or:sparse
device
- the compute device, like CPU or GPUrequires_grad
- whether or not to record gradients
You can specify properties when creating a tensor
Torch.rand(2, 3, dtype: :float64, layout: :strided, device: "cpu", requires_grad: true)
Operations
Create a tensor
x = Torch.tensor([10, 20, 30])
Add
x + 5 # tensor([15, 25, 35])
Subtract
x - 5 # tensor([5, 15, 25])
Multiply
x * 5 # tensor([50, 100, 150])
Divide
x / 5 # tensor([2, 4, 6])
Get the remainder
x % 3 # tensor([1, 2, 0])
Raise to a power
x**2 # tensor([100, 400, 900])
Perform operations with other tensors
y = Torch.tensor([1, 2, 3])
x + y # tensor([11, 22, 33])
Perform operations in-place
x.add!(5)
x # tensor([15, 25, 35])
You can also specify an output tensor
result = Torch.empty(3)
Torch.add(x, y, out: result)
result # tensor([15, 25, 35])
Numo
Convert a tensor to a Numo array
a = Torch.ones(5)
a.numo
Convert a Numo array to a tensor
b = Numo::NArray.cast([1, 2, 3])
Torch.from_numo(b)
Autograd
Create a tensor with requires_grad: true
x = Torch.ones(2, 2, requires_grad: true)
Perform operations
y = x + 2
z = y * y * 3
out = z.mean
Backprop
out.backward
Get gradients
x.grad # tensor([[4.5, 4.5], [4.5, 4.5]])
Stop autograd from tracking history
x.requires_grad # true
(x**2).requires_grad # true
Torch.no_grad do
(x**2).requires_grad # false
end
Neural Networks
Define a neural network
class MyNet < Torch::NN::Module
def initialize
super()
@conv1 = Torch::NN::Conv2d.new(1, 6, 3)
@conv2 = Torch::NN::Conv2d.new(6, 16, 3)
@fc1 = Torch::NN::Linear.new(16 * 6 * 6, 120)
@fc2 = Torch::NN::Linear.new(120, 84)
@fc3 = Torch::NN::Linear.new(84, 10)
end
def forward(x)
x = Torch::NN::F.max_pool2d(Torch::NN::F.relu(@conv1.call(x)), [2, 2])
x = Torch::NN::F.max_pool2d(Torch::NN::F.relu(@conv2.call(x)), 2)
x = Torch.flatten(x, 1)
x = Torch::NN::F.relu(@fc1.call(x))
x = Torch::NN::F.relu(@fc2.call(x))
@fc3.call(x)
end
end
Create an instance of it
net = MyNet.new
input = Torch.randn(1, 1, 32, 32)
net.call(input)
Get trainable parameters
net.parameters
Zero the gradient buffers and backprop with random gradients
net.zero_grad
out.backward(Torch.randn(1, 10))
Define a loss function
output = net.call(input)
target = Torch.randn(10)
target = target.view(1, -1)
criterion = Torch::NN::MSELoss.new
loss = criterion.call(output, target)
Backprop
net.zero_grad
p net.conv1.bias.grad
loss.backward
p net.conv1.bias.grad
Update the weights
learning_rate = 0.01
net.parameters.each do |f|
f.data.sub!(f.grad.data * learning_rate)
end
Use an optimizer
optimizer = Torch::Optim::SGD.new(net.parameters, lr: 0.01)
optimizer.zero_grad
output = net.call(input)
loss = criterion.call(output, target)
loss.backward
optimizer.step
Saving and Loading Models
Save a model
Torch.save(net.state_dict, "net.pth")
Load a model
net = MyNet.new
net.load_state_dict(Torch.load("net.pth"))
net.eval
When saving a model in Python to load in Ruby, convert parameters to tensors (due to outstanding bugs in LibTorch)
state_dict = {k: v.data if isinstance(v, torch.nn.Parameter) else v for k, v in state_dict.items()}
torch.save(state_dict, "net.pth")
Tensor Creation
Here’s a list of functions to create tensors (descriptions from the C++ docs):
-
arange
returns a tensor with a sequence of integersTorch.arange(3) # tensor([0, 1, 2])
-
empty
returns a tensor with uninitialized valuesTorch.empty(3) # tensor([7.0054e-45, 0.0000e+00, 0.0000e+00])
-
eye
returns an identity matrixTorch.eye(2) # tensor([[1, 0], [0, 1]])
-
full
returns a tensor filled with a single valueTorch.full([3], 5) # tensor([5, 5, 5])
-
linspace
returns a tensor with values linearly spaced in some intervalTorch.linspace(0, 10, 5) # tensor([0, 5, 10])
-
logspace
returns a tensor with values logarithmically spaced in some intervalTorch.logspace(0, 10, 5) # tensor([1, 1e5, 1e10])
-
ones
returns a tensor filled with all onesTorch.ones(3) # tensor([1, 1, 1])
-
rand
returns a tensor filled with values drawn from a uniform distribution on [0, 1)Torch.rand(3) # tensor([0.5444, 0.8799, 0.5571])
-
randint
returns a tensor with integers randomly drawn from an intervalTorch.randint(1, 10, [3]) # tensor([7, 6, 4])
-
randn
returns a tensor filled with values drawn from a unit normal distributionTorch.randn(3) # tensor([-0.7147, 0.6614, 1.1453])
-
randperm
returns a tensor filled with a random permutation of integers in some intervalTorch.randperm(3) # tensor([2, 0, 1])
-
zeros
returns a tensor filled with all zerosTorch.zeros(3) # tensor([0, 0, 0])
LibTorch Compatibility
Here’s the list of compatible versions.
Torch.rb | LibTorch |
---|---|
0.18.x | 2.5.x |
0.17.x | 2.4.x |
0.16.x | 2.3.x |
0.15.x | 2.2.x |
0.14.x | 2.1.x |
0.13.x | 2.0.x |
0.12.x | 1.13.x |
Performance
Deep learning is significantly faster on a GPU.
Linux
With Linux, install CUDA and cuDNN and reinstall the gem.
Check if CUDA is available
Torch::CUDA.available?
Move a neural network to a GPU
net.cuda
If you don’t have a GPU that supports CUDA, we recommend using a cloud service. Paperspace has a great free plan. We’ve put together a Docker image to make it easy to get started. On Paperspace, create a notebook with a custom container. Under advanced options, set the container name to:
ankane/ml-stack:torch-gpu
And leave the other fields in that section blank. Once the notebook is running, you can run the MNIST example.
Mac
With Apple silicon, check if Metal Performance Shaders (MPS) is available
Torch::Backends::MPS.available?
Move a neural network to a GPU
device = Torch.device("mps")
net.to(device)
History
View the changelog
Contributing
Everyone is encouraged to help improve this project. Here are a few ways you can help:
- Report bugs
- Fix bugs and submit pull requests
- Write, clarify, or fix documentation
- Suggest or add new features
To get started with development:
git clone https://github.com/ankane/torch.rb.git
cd torch.rb
bundle install
bundle exec rake compile -- --with-torch-dir=/path/to/libtorch
bundle exec rake test
You can use this script to test on GPUs with the AWS Deep Learning Base AMI (Ubuntu 18.04).
Here are some good resources for contributors: