This is a quantized version of https://huggingface.co/laion/CLIP-ViT-B-32-256x256-DataComp-s34B-b86K that is ready to use with (DeepSparse)[https://github.com/neuralmagic/deepsparse]

It achieves 71.1% one-shot accuracy on ImageNet.

## Usage

First, install DeepSparse with extensions for CLIP:
```
pip install deepsparse-nightly[clip]>=1.7.0.20231210
```

Download some test images of a church, a dog, and elephants:
```
wget -O basilica.jpg https://raw.githubusercontent.com/neuralmagic/deepsparse/main/src/deepsparse/yolo/sample_images/basilica.jpg
wget -O buddy.jpeg https://raw.githubusercontent.com/neuralmagic/deepsparse/main/tests/deepsparse/pipelines/sample_images/buddy.jpeg
wget -O thailand.jpg https://raw.githubusercontent.com/neuralmagic/deepsparse/main/src/deepsparse/yolact/sample_images/thailand.jpg
```

Then make and run a pipeline in Python:
```python
import numpy as np
from deepsparse import Pipeline
from deepsparse.clip import (
    CLIPTextInput,
    CLIPVisualInput,
    CLIPZeroShotInput
)

def new_process_inputs(self, inputs: CLIPTextInput):
    if not isinstance(inputs.text, list):
        inputs.text = [inputs.text]
    if not isinstance(inputs.text[0], str):
        return inputs.text
    tokens = [np.array(t).astype(np.int32) for t in self.tokenizer(inputs.text)]
    tokens = np.stack(tokens, axis=0)
    tokens_lengths = np.array(tokens.shape[0] * [tokens.shape[1] - 1])
    return [tokens, tokens_lengths]

# This overrides the process_inputs function globally for all CLIPTextPipeline classes,
# so when we make a zeroshot pipeline later that uses this class, it will use this edit!
CLIPTextPipeline.process_inputs = new_process_inputs

possible_classes = ["ice cream", "an elephant", "a dog", "a building", "a church"]
images = ["basilica.jpg", "buddy.jpeg", "thailand.jpg"]

pipeline = Pipeline.create(task="clip_zeroshot", visual_model_path="visual.onnx", text_model_path="textual.onnx")

pipeline_input = CLIPZeroShotInput(
    image=CLIPVisualInput(images=images),
    text=CLIPTextInput(text=possible_classes),
)

output = pipeline(pipeline_input).text_scores
for i in range(len(output)):
    prediction = possible_classes[np.argmax(output[i])]
    print(f"Image {images[i]} is a picture of {prediction}")

"""
Image basilica.jpg is a picture of a church
Image buddy.jpeg is a picture of a dog
Image thailand.jpg is a picture of an elephant
"""
```