https://github.com/vladimirrotariu/surrealml-vs-onnx-vs-pytorch
We will compare the performance of PyTorch with ONNX and SurrealML, a library designed for uploading and inferring using ML models directly inside a SurrealDB instance.
https://github.com/vladimirrotariu/surrealml-vs-onnx-vs-pytorch
ai artificial-intelligence artificial-neural-networks database hacktoberfest iot-database machine-learning ml onnx python pytorch realtime-database surreal surrealdb
Last synced: 6 months ago
JSON representation
We will compare the performance of PyTorch with ONNX and SurrealML, a library designed for uploading and inferring using ML models directly inside a SurrealDB instance.
- Host: GitHub
- URL: https://github.com/vladimirrotariu/surrealml-vs-onnx-vs-pytorch
- Owner: vladimirrotariu
- Created: 2024-05-10T22:25:26.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-11-01T13:05:47.000Z (11 months ago)
- Last Synced: 2025-02-02T07:13:44.973Z (8 months ago)
- Topics: ai, artificial-intelligence, artificial-neural-networks, database, hacktoberfest, iot-database, machine-learning, ml, onnx, python, pytorch, realtime-database, surreal, surrealdb
- Language: Jupyter Notebook
- Homepage: https://github.com/surrealdb/surrealml
- Size: 24.2 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# SurrealML versus PyTorch versus ONNX
SurrealML simplifies storing, loading, and executing trained ML models, working alongside existing frameworks like PyTorch, TensorFlow, scikit-learn, and linfa, while supporting execution in both Python and Rust.
In this repo, you can find [the Jupyter notebook](https://github.com/vladimirrotariu/surrealml-vs-onnx-vs-pytorch/blob/main/surrealml_vs_onnx_vs_pytorch/comparing_surreal_ml_onnx_pytorch.ipynb) behind the following comparison, which shows execution times for `SurrealML`, as contrasted with `PyTorch` and `ONNX`.
To start, clone this repository, and then execute:
```bash
poetry install
```
and then follow the Jupyter notebook attached.
For convenience, you can find here [a PDF version](https://github.com/vladimirrotariu/surrealml-vs-onnx-vs-pytorch/blob/main/surrealml_vs_onnx_vs_pytorch/comparing_surreal_ml_onnx_pytorch.pdf) of the Jupyter notebook.
## System specs
We used the following setup to create the benchmark:
```text
Number of physical CPU cores: 10
Number of logical CPU cores: 20
Total Memory (RAM): 15.47 GB
Operating System: Linux 5.15.153.1-microsoft-standard-WSL2
Processor: x86_64
Python Version: 3.11.10
```
as well as `SurrealDB 1.5.5`.