Renamed to Avalon.jl
44 Stars
Updated Last
1 Year Ago
Started In
September 2019


⚠️ This package has been renamed to Avalon.jl, all new features will be added there

Build Status

Lilith is a deep learning library in Julia with focus on high performance and interoperability with existing DL frameworks. Its main features include:

  • tracing autograd engine - models are just structs, transformations are just functions
  • optimizing code generator based on hackable computational graph
  • GPU support
  • layer API similar to PyTorch's to ease translation of existing Python code to Julia
  • high backward compatibility to allow accumulation of models


To get you a feeling of what Lilith is like, here's a definition of a small convolutional neural network:

using Lilith

mutable struct Net

Net() = Net(
    Conv2d(1, 20, 5),
    Conv2d(20, 50, 5),
    Linear(4 * 4 * 50, 500),
    Linear(500, 10)

function (m::Net)(x::AbstractArray)
    x = maxpool2d(relu.(m.conv1(x)), (2, 2))
    x = maxpool2d(relu.(m.conv2(x)), (2, 2))
    x = reshape(x, 4*4*50, :)
    x = relu.(m.fc1(x))
    x = logsoftmax(m.fc2(x))
    return x

For detailed explanation of this and other models see the tutorial. Some predefined models are also available in the zoo.


Performance comparison between different libraries is hard and benchmarks are rarely fair, but here's our best shot in this direction:

Convolutional neural network

Code available here

training 1 epoch training total time* prediction
Lilith (CPU) 170 s 1742 s 39 ms
Flux (CPU) 250 s 2515 s 42 ms
------------- ---------------- -------------------- ----------
Lilith (GPU) 10 s 164 s 5 ms
Flux (GPU) 12 s 150 s 5 ms
PyTorch (GPU) 12 s 120 s 2 ms

* - total time includes 10 epochs + compilation time

Note that in the test on GPU Lilith has longest compilation time and thus longest total training time after 10 epochs. However, time per epoch is the lowest, so Lilith is typically the fastest one in longer run.

Variational Autoencoder

Code available here

training 1 epoch training total time prediction
Lilith (CPU) 50 s 535 s 395 μs
Flux (CPU) 948 s 158 min 81 ms
------------- ---------------- -------------------- ----------
Lilith (GPU) 3 s 93 s 194 μs
Flux (GPU)** --- --- ---
PyTorch (GPU) 7 s 66 s 501 µs

** - VAE example from the Flux zoo doesn't work on GPU

API Stability

One of the central ideas behind Lilith is the ability to reuse existing code instead of writing everything from scratch. To facilitate it, Lilith is committed to high, although not absolute backward compatibility. The following table outlines stability level you should expect from various components of the library.

Component API Stable?
Basic layers Yes
Losses Mostly
Activations Yes
Initializations Mostly
Optimizers Yes
Device API Yes
Fitting API No**

* - currently Lilith provides only basic implementations of vanilla RNN, LSTM and GRU; this implementation will be improved in future version and made more compatible with PyTorch version, but currently it cannot be considered stable

** - function fit!() provides a convenient shortcut for training supervised learning models, but in its current state it's too basic for most real use cases; for more durable code consider writing your own method for training using fit!() as a template

Please note that until version 1.0 "stable API" means that we will try our best to keep it unchanged, but we reserve the right to the break the rule in some rare and exceptional cases.

Used By Packages

No packages found.