AI is becoming increasingly important for retail use cases. It can provide retailers with advanced capabilities to personalize customer experiences, optimize operations, and increase sales.
This retail use case shows an example of fine-tuning a YOLOX-PyTorch* model to manage shelf space and shelf inventory.
This example takes a dataset of store shelf images as input and uses computer vision algorithms to analyze the images to identify the products and their attributes. See how to quickly fine-tune a deep learning model based on existing source code in the Model References GitHub* repository to apply this retail data. Retailers can use this data for shelf analytics, out-of-stock detection, and promotional compliance.
This use case is available on the Intel Gaudi accelerator solutions repository in a Jupyter* Notebook. To run this use case, start with one of the following two options. For more information, see the Installation Guide.
Access Intel® Gaudi® accelerators or Intel Gaudi 2 accelerators in two ways:
- Amazon Elastic Compute Cloud (Amazon EC2*) DL1 instances based on first-generation Intel Gaudi accelerators.
For instructions on how to start a DL1 instance, see our Quick Start Guide. An Amazon Web Services (AWS)* user account is required.
- Intel® Tiber™ Developer Cloud using Intel Gaudi 2 accelerators.
To get access to the Intel Tiber Developer Cloud, see Get Started. For instructions on how to connect to and run models on the Intel Tiber Developer Cloud, see the Intel Tiber Developer Cloud Quick Start Guide. A user account is required.