MachineLearningBenchMarkingToolkit
Cross-platform machine-learning benchmarking toolkit to compare Windows PCs and MacBook (Pro/Air) machines. It reports system specs (CPU cores/frequency, RAM) and runs lightweight ML-style benchmarks on:
- CPU (always)
- NVIDIA CUDA GPU (Windows/Linux, if available)
- Apple Silicon GPU via MPS (macOS, if available)
Outputs are saved as JSON files so you can easily compare multiple machines.
Features
- ✅ Machine specs: hostname, OS, CPU model, cores, CPU freq, RAM totals/available
- ✅ Accelerator info:
- CUDA: GPU name, total VRAM, compute capability
- MPS: Apple Silicon (unified memory note)
- ✅ Benchmarks:
- PyTorch matmul on CPU
- PyTorch matmul on CUDA/MPS (if available)
- Optional: scikit-learn RandomForest training benchmark
Installation
Minimal
pip install mlbenchkit
With PyTorch benchmarks
# With PyTorch benchmarks
pip install "mlbenchkit[torch]"
With scikit-learn benchmark
pip install "mlbenchkit[sklearn]"
Everything
pip install "mlbenchkit[torch,sklearn]"
## How to use it?
```bash
mlbench specs
Run benchmark suite (saves a JSON file):
mlbench run
# Run with sklearn benchmark:
mlbench run --with-sklearn
## Customize workload:
mlbench run --cpu-N 2048 --gpu-N 4096 --iters 50 --warmup 20 --gpu-dtype float16
1 maintainer
Release files for mlbenchkit 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| mlbenchkit-0.1.0.tar.gz | 6.5 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| mlbenchkit-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 14.5 kB
Release files / mlbenchkit-0.1.0.tar.gz
| Download URL | mlbenchkit-0.1.0.tar.gz |
|---|---|
| Size | 6.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
18f0d5ae7e92d018244fe88803f8ad1821a4252b1adb05ed5a13e2a7eef18e7f
|
|
BLAKE2b-256 checksum How to use checksums |
6322205ebe8864db1f927644b8c3cc4678e6831e229af67b661db66cd94bb0a0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.9.0
|
Release files / mlbenchkit-0.1.0-py3-none-any.whl
| Download URL | mlbenchkit-0.1.0-py3-none-any.whl |
|---|---|
| Size | 8.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
5486bb0a26255c3728f6ae1b98d07f7a446bbb7ebb7a9adbd97ae54acb13602c
|
|
BLAKE2b-256 checksum How to use checksums |
2764715f4a027b4c586ff4328d26fb827ba3c5b174bcc8628cc439be86628abc
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.9.0
|
