Metadata-Version: 2.1
Name: nnio
Version: 0.2.4.3
Summary: Neural network inference on accelerators simplified
Home-page: https://github.com/FastSense/nnio
Author: Ruslan Baynazarov
Author-email: ruslan.baynazarov@fastsense.tech
License: MIT
Description: # nnio
        
        Please refer to the [project's documentation](https://nnio.readthedocs.io/).
        
        ## What is it
        
        **nnio** is a light-weight python package for easily running neural networks.
        
        It supports running models on CPU as well as some of the edge devices:
        
        * Google USB Accelerator
        * Intel Compute Stock
        * Intel integrated GPUs
        
        For each device there exists an own library and a model format. We wrap all those in a single well-defined python package.
        
        Look at this simple example:
        
        ```python
        import nnio
        
        # Create model and put it on a Google Coral Edge TPU device
        model = nnio.EdgeTPUModel(
            model_path='path/to/model_quant_edgetpu.tflite',
            device='TPU',
        )
        # Create preprocessor
        preproc = nnio.Preprocessing(
            resize=(224, 224),
            batch_dimension=True,
        )
        
        # Preprocess your numpy image
        image = preproc(image_rgb)
        
        # Make prediction
        class_scores = model(image)
        ```
        
        **nnio** was developed for the [Fast Sense X](https://fastsense.readthedocs.io/en/latest/) microcomputer.
        It has **six neural accelerators**, which are all supported by nnio:
        
        * 3 x [Google Coral Edge TPU](https://coral.ai/)
        * 2 x [Intel VPU](https://www.intel.ru/content/www/ru/ru/products/processors/movidius-vpu/movidius-myriad-x.html)
        * an integrated Intel GPU
        
        [More usage examples](https://nnio.readthedocs.io/en/latest/basic_usage.html) can be found in the [documentation](https://nnio.readthedocs.io/).
Platform: UNKNOWN
Description-Content-Type: text/markdown
