DL4S
Accelerated tensor operations and dynamic neural networks based on reverse mode automatic differentiation for every device that can run Swift - from watchOS to Linux (by palle-k)
Bender
Easily craft fast Neural Networks on iOS! Use TensorFlow models. Metal under the hood. (by xmartlabs)
Our great sponsors
The number of mentions indicates the total number of mentions that we've tracked plus the number of user suggested alternatives.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
DL4S
Posts with mentions or reviews of DL4S.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2021-04-05.
-
Machine learning modules for swift
Lastly, there are some third party libraries that you could try. I wrote a machine learning / deep learning library for Swift a while ago: DL4S. It does not have GPU acceleration (yet), so it won't allow you to do large ML workloads, but it has no problem with datasets like MNIST and it has an API similar to PyTorch/Tensorflow 2.
-
Should I use accelerate or another library for simple, np.sum-like, matrix operations?
Shameless plug: If you're looking for a more user friendly method for accelerated operations on vectors, matrices and tensors: I built Deep Learning for Swift a while ago, which implements a lot of numpy functions. It's primarily made for deep learning but you can also do number crunching with it.
-
Anyone taking part or has taken part in the Swift Student Challenge?
Last year I won by building a chat bot with seq2seq and attention using my own deep learning library. The whole thing wasn't all that impressive from a visual standpoint but I guess the technical achievement was good enough for them. Also, I wrote a lot of stuff into the beyond WWDC field.
-
Recommendations for Books on Deep Learning.
If you want to go the latter route, Apple provides a bunch of low level frameworks for this: Accelerate, BNNS, ML Compute and MetalPerformanceShaderGraph. CoreML also supports some limited fine tuning capabilities. There are also 3rd party solutions, like DL4S (which I created).
-
Any good open source projects that uses Swift?
I actually have one project in this list myself (DL4S), but the project is probably not very beginner friendly to work on.
Bender
Posts with mentions or reviews of Bender.
We have used some of these posts to build our list of alternatives
and similar projects.
We haven't tracked posts mentioning Bender yet.
Tracking mentions began in Dec 2020.
What are some alternatives?
When comparing DL4S and Bender you can also consider the following projects:
Swift-AI - The Swift machine learning library.
CoreML-Models - Largest list of models for Core ML (for iOS 11+)
CoreML-samples - Sample code for Core ML using ResNet50 provided by Apple and a custom model generated by coremltools.
MLKit - A simple machine learning framework written in Swift 🤖
Tensorflow-iOS
AIToolbox - A toolbox of AI modules written in Swift: Graphs/Trees, Support Vector Machines, Neural Networks, PCA, K-Means, Genetic Algorithms
Caffe2
Serrano - A Swift deep learning library with Accelerate and Metal support.
spokestack-ios - Spokestack: give your iOS app a voice interface!