Skip to content

Problems using BridgeTower Local in Multimodal RAG without PredictionGuard #16

@ixn3rd3mxn

Description

@ixn3rd3mxn

I need help. I am currently studying Multimodal RAG: Chat with Videos
https://www.deeplearning.ai/short-courses/multimodal-rag-chat-with-videos/
In the course, there is a use of bridgetower-large-itm-mlm-itc using predictionguard. When I want to try it on a local laptop, following all the examples in the course, I am currently working on the chapter L4_Multimodal Retrieval from Vector Stores. I am having trouble with bridgetower-large-itm-mlm-itc using predictionguard, which I do not have an API KEY for. So I searched for information on huggingface and found https://huggingface.co/BridgeTower/bridgetower-large-itm-mlm-itc. But the next problem I encountered is how do I make a function to solve this problem?

# helper function to compute the joint embedding of a prompt and a base64-encoded image through PredictionGuard
def bt_embedding_from_prediction_guard(prompt, base64_image):
    # get PredictionGuard client
    client = _getPredictionGuardClient()
    message = {"text": prompt,}
    if base64_image is not None and base64_image != "":
        if not isBase64(base64_image): 
            raise TypeError("image input must be in base64 encoding!")
        message['image'] = base64_image
    response = client.embeddings.create(
        model="bridgetower-large-itm-mlm-itc",
        input=[message]
    )
    return response['data'][0]['embedding']

Can you suggest how I should modify the function to successfully use bridgetower-large-itm-mlm-itc locally?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions