https://github.com/edgeimpulse/python-sdk
https://github.com/edgeimpulse/python-sdk
Last synced: 10 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/edgeimpulse/python-sdk
- Owner: edgeimpulse
- License: bsd-3-clause-clear
- Created: 2023-08-03T22:22:32.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2025-03-28T07:03:18.000Z (about 1 year ago)
- Last Synced: 2025-07-17T15:59:12.732Z (11 months ago)
- Language: Python
- Size: 33.4 MB
- Stars: 9
- Watchers: 2
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
# Edge Impulse SDK
The official Python SDK for Edge Impulse is designed to help machine learning practitioners build and deploy models for embedded hardware and edge AI applications.
- Profile your model to estimate RAM, ROM, and inference speed
- Convert your model to C++ to deploy on edge hardware
- Interact with Edge Impulse projects to collect data, train models, and deploy them to edge devices
[Sign up for a free account →](https://studio.edgeimpulse.com/signup)
List of versions and changes can be found [in this changelog](https://github.com/edgeimpulse/python-sdk/blob/main/CHANGELOG.md).
## Getting Started
Install the Edge Impulse Python SDK:
```sh
pip install edgeimpulse
```
Estimate RAM, ROM, and inference speed for a variety of hardware platforms:
```python
import edgeimpulse as ei
# Change to an API key from your Edge Impulse project
ei.API_KEY = "your-api-key"
# Print inference estimates
result = ei.model.profile(model="path/to/model")
result.summary()
```
To learn about the full functionality, see the resources below.
## Resources
- [Overview and getting started](https://docs.edgeimpulse.com/docs/edge-impulse-python-sdk/overview)
- [Tutorial demonstrating how to profile and deploy a model](https://docs.edgeimpulse.com/docs/edge-impulse-python-sdk/01-python-sdk-with-tf-keras)
- [API reference guide](https://docs.edgeimpulse.com/reference/python-sdk/overview)