Hugging Face's logo

qualcomm
/
OpenAI-Clip

Image Classification
PyTorch
LiteRT
ONNX
foundation
android
Model card Files Community
1

OpenAI-Clip: Optimized for Mobile Deployment

Multi-modal foundational model for vision and language tasks like image/text similarity and for zero-shot image classification

Contrastive Language-Image Pre-Training (CLIP) uses a ViT like transformer to get visual features and a causal language model to get the text features. Both the text and visual features can then be used for a variety of zero-shot learning tasks.

This model is an implementation of OpenAI-Clip found here.

This repository provides scripts to run OpenAI-Clip on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Image classification
  • Model Stats:
    • Model checkpoint: ViT-B/16
    • Image input resolution: 224x224
    • Text context length: 77
    • Number of parameters (CLIPTextEncoder): 76.0M
    • Model size (CLIPTextEncoder): 290 MB
    • Number of parameters (CLIPImageEncoder): 115M
    • Model size (CLIPImageEncoder): 437 MB
Model Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Precision Primary Compute Unit Target Model
OpenAI-Clip Samsung Galaxy S23 Snapdragon® 8 Gen 2 TFLITE 24.855 ms 0 - 23 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip Samsung Galaxy S23 Snapdragon® 8 Gen 2 QNN 19.238 ms 0 - 22 MB FP16 NPU OpenAI-Clip.so
OpenAI-Clip Samsung Galaxy S23 Snapdragon® 8 Gen 2 ONNX 25.697 ms 0 - 104 MB FP16 NPU OpenAI-Clip.onnx
OpenAI-Clip Samsung Galaxy S24 Snapdragon® 8 Gen 3 TFLITE 18.007 ms 0 - 383 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip Samsung Galaxy S24 Snapdragon® 8 Gen 3 QNN 13.619 ms 1 - 391 MB FP16 NPU OpenAI-Clip.so
OpenAI-Clip Samsung Galaxy S24 Snapdragon® 8 Gen 3 ONNX 18.132 ms 1 - 473 MB FP16 NPU OpenAI-Clip.onnx
OpenAI-Clip Snapdragon 8 Elite QRD Snapdragon® 8 Elite TFLITE 14.586 ms 0 - 383 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip Snapdragon 8 Elite QRD Snapdragon® 8 Elite QNN 11.09 ms 0 - 388 MB FP16 NPU Use Export Script
OpenAI-Clip Snapdragon 8 Elite QRD Snapdragon® 8 Elite ONNX 17.609 ms 1 - 443 MB FP16 NPU OpenAI-Clip.onnx
OpenAI-Clip SA7255P ADP SA7255P TFLITE 368.401 ms 0 - 383 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip SA7255P ADP SA7255P QNN 306.68 ms 1 - 9 MB FP16 NPU Use Export Script
OpenAI-Clip SA8255 (Proxy) SA8255P Proxy TFLITE 24.877 ms 0 - 20 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip SA8255 (Proxy) SA8255P Proxy QNN 20.983 ms 1 - 3 MB FP16 NPU Use Export Script
OpenAI-Clip SA8295P ADP SA8295P TFLITE 30.504 ms 0 - 336 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip SA8295P ADP SA8295P QNN 24.978 ms 1 - 17 MB FP16 NPU Use Export Script
OpenAI-Clip SA8650 (Proxy) SA8650P Proxy TFLITE 24.962 ms 0 - 23 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip SA8650 (Proxy) SA8650P Proxy QNN 21.068 ms 1 - 3 MB FP16 NPU Use Export Script
OpenAI-Clip SA8775P ADP SA8775P TFLITE 34.905 ms 0 - 382 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip SA8775P ADP SA8775P QNN 28.903 ms 1 - 11 MB FP16 NPU Use Export Script
OpenAI-Clip QCS8275 (Proxy) QCS8275 Proxy TFLITE 368.401 ms 0 - 383 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip QCS8275 (Proxy) QCS8275 Proxy QNN 306.68 ms 1 - 9 MB FP16 NPU Use Export Script
OpenAI-Clip QCS8550 (Proxy) QCS8550 Proxy TFLITE 24.987 ms 0 - 24 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip QCS8550 (Proxy) QCS8550 Proxy QNN 21.05 ms 1 - 3 MB FP16 NPU Use Export Script
OpenAI-Clip QCS9075 (Proxy) QCS9075 Proxy TFLITE 34.905 ms 0 - 382 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip QCS9075 (Proxy) QCS9075 Proxy QNN 28.903 ms 1 - 11 MB FP16 NPU Use Export Script
OpenAI-Clip QCS8450 (Proxy) QCS8450 Proxy TFLITE 26.982 ms 0 - 351 MB FP16 NPU OpenAI-Clip.tflite
OpenAI-Clip QCS8450 (Proxy) QCS8450 Proxy QNN 22.097 ms 1 - 394 MB FP16 NPU Use Export Script
OpenAI-Clip Snapdragon X Elite CRD Snapdragon® X Elite QNN 21.709 ms 1 - 1 MB FP16 NPU Use Export Script
OpenAI-Clip Snapdragon X Elite CRD Snapdragon® X Elite ONNX 26.59 ms 293 - 293 MB FP16 NPU OpenAI-Clip.onnx

Installation

Install the package via pip:

pip install "qai-hub-models[openai-clip]"

Configure Qualcomm® AI Hub to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.openai_clip.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.openai_clip.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

  • Performance check on-device on a cloud-hosted device
  • Downloads compiled assets that can be deployed on-device for Android.
  • Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.openai_clip.export

Profiling Results
------------------------------------------------------------
OpenAI-Clip
Device                          : Samsung Galaxy S23 (13)   
Runtime                         : TFLITE                    
Estimated inference time (ms)   : 24.9                      
Estimated peak memory usage (MB): [0, 23]                   
Total # Ops                     : 1320                      
Compute Unit(s)                 : NPU (1318 ops) CPU (2 ops)

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.openai_clip import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S24")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub. Sign up for access.

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

  • TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.

  • QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on OpenAI-Clip's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of OpenAI-Clip can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Image Classification
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support