tlkh/tf-metal-experiments

TensorFlow Metal Backend on Apple Silicon Experiments (just for fun)

42
/ 100
Emerging

This project helps machine learning engineers and researchers evaluate the performance of common deep learning models on Apple Silicon (M1 series) Macs. It takes various image classification and natural language processing models as input and outputs their training throughput, power consumption, and memory usage. This is for those who want to understand the practical limits and capabilities of their Apple Silicon hardware for deep learning tasks.

280 stars. No commits in the last 6 months.

Use this if you are a machine learning engineer or researcher looking to benchmark the real-world performance of popular deep learning models like ResNet, MobileNet, or BERT on your M1, M1 Max, or M1 Ultra Mac.

Not ideal if you are a developer looking for a production-ready TensorFlow Metal integration or if you need to benchmark models that are not included in these experiments.

deep-learning model-benchmarking machine-learning-engineering apple-silicon-ml model-performance
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

280

Forks

31

Language

Jupyter Notebook

License

MIT

Last pushed

Feb 10, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/tlkh/tf-metal-experiments"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.