AI HUB

 "Streamline your AI deployment process with Qualcomm® AI Hub, your gateway to effortless deployment of AI models on edge devices. Qualcomm® AI Hub simplifies the deployment of AI models for vision, audio, and speech applications, enabling you to optimize, validate, and deploy your models on hosted Qualcomm platform devices within minutes.

How does it work?

Simply upload your model, and Qualcomm® AI Hub handles the rest: auto-optimizing for target devices, validating performance and numerics, and repeating the process until your model is ready to deploy.

Installation

The Qualcomm AI Hub library for optimization, profiling, and validation can be installed via PyPI. We recommend using Miniconda to manage your python versions and environments. To install, run the following command in your terminal. We recommend Python > 3.8 and < 3.10

you may face SSL version issue due to conflict of python of yours and server so for different model try to use python version which suits that model.

run the following commands in terminal :

pip3 install qai-hub
Sign in to Qualcomm AI Hub with your Qualcomm® ID. After signing in, navigate to[your Qualcomm ID] -> Settings -> API Token. This should provide an API token that you can use to configure your client.
qai-hub configure --api_token API_TOKEN
One configured, you can check that your API token is installed correctly by fetching a list of available devices with the following command
qai-hub list-devices
Run your first PyTorch model on a hosted device
Once you have set up your Qualcomm AI Hub environment, the next step is to optimize, validate, and deploy a PyTorch model on a cloud hosted device. This example requires some extra dependencies which can be installed using the following:
pip3 install "qai-hub[torch]"
Now, you can request an automated performance analysis of the MobileNet v2 network. This example compiles and profiles the model on a real hosted device:

import qai_hub as hub
import torch
from torchvision.models import mobilenet_v2

# Using pre-trained MobileNet
torch_model = mobilenet_v2(pretrained=True)
torch_model.eval()

# Trace model (for on-device deployment)
input_shape = (1, 3, 224, 224)
example_input = torch.rand(input_shape)
torch_model = torch.jit.trace(torch_model, example_input)

# Choose a device
devices = hub.get_devices()
for device_num in range(1, len(devices)+1):
    print(f"{ device_num }. { devices[device_num-1].name }")

device_choice = 0
while device_choice < 1 or device_choice > len(devices):
    try:
        device_choice = int(input("Choose a device: "))
    except: pass
device = devices[device_choice-1]

# Optimize model for the chosen device
compile_job = hub.submit_compile_job(
    model=torch_model,
    name="MyMobileNetModel",
    device=device,
    input_specs=dict(image=input_shape),
)

# Run the model on a hosted device
profile_job = hub.submit_profile_job(
  model=compile_job.get_target_model(),
  device=device,
)
This will submit a compilation job and then a profiling job .
Easily select from Qualcomm AI Hub Models
Qualcomm AI Hub Models are a collection of state-of-the-art machine learning models optimized for performance and ready to deploy on Qualcomm platform devices. You can explore models optimized for on-device deployment for vision, speech, text, and generative AI applications. All models come with open-source recipes on GitHub and Hugging Face.
Here is a simple example of real-time selfie segmentation that can be easily exported and deployed on-device. First, we install the Qualcomm AI Hub Models python package.
pip3 install "qai-hub-models[ffnet_40s]"
Run a local PyTorch based CLI demo to validate the model off-device with some sample input.
python -m qai_hub_models.models.ffnet_40s.demo
Once verified off-device, you can then run this model on a hosted Qualcomm platform device within minutes.
python -m qai_hub_models.models.ffnet_40s.export
The above script optimizes the model for on-device execution, profiles the model on a cloud hosted Qualcomm platform devices, runs the model on-device, and compares accuracy between a local CPU based PyTorch run and the on-device run. Finally, you can run a full demo with the inference running on-device.
python -m qai_hub_models.models.ffnet_40s.demo --on-device
With Qualcomm® AI Hub, you can enjoy optimized translation, with automatic model conversion from PyTorch or ONNX to TensorFlow Lite or Qualcomm AI Engine Direct. The resulting models are finely tuned for deployment on CPU, GPU, or NPU.
Access a vast library of over 100 optimized AI models covering various applications, including vision, speech, audio, and text, deployable on devices powered by Snapdragon® and Qualcomm platforms within minutes. Plus, all the recipes are fully open-sourced and available on GitHub and Hugging Face.
Experience real devices in action by easily running models on hosted Qualcomm platform devices provisioned within minutes. Perform performance profiling and on-device inference with just a few lines of code.
Getting started is a breeze. Install the Qualcomm AI Hub library for optimization, profiling, and validation via PyPI, and configure it with your API token. Once set up, you can start deploying your PyTorch models on cloud-hosted devices effortlessly.