PromptClip with Gemini, output modified

[Cheez22]

This version of the code uses Google's Gemini. The output has been modified to improve video editing efficiency, turning PromptClip into more of a search engine for videos.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[Cheez22]

@codeAshu
@ashish-spext

Hi, just bumping this PR, please let me know if there's anything else I need to do on my side.

[ashish-spext]

@Cheez22 Copying all the existing files into a new notebook as cells is not a good idea, as it unnecessarily duplicates code that could be reused from the existing files.

There is no need to create a new notebook file as well just for Gemini.

Ideal implementation: Users should be able to pass a Gemini LLM instance to the text_prompter, scene_prompter, and multimodal_prompter functions in video_prompter.py called from existing notebooks.

This approach will enable users to select their preferred LLM in the existing notebooks (PromptClip_spoken.ipynb, PromptClip_visual.ipynb, PromptClip_multimodal.ipynb, and others that may be added in the future) without having to create separate copies of each notebook for different LLMs.

Could you delete the Gemini file and, instead, try passing the Gemini LLM instance directly into the functions of the existing notebooks? This approach should unlock the support without needing a separate file.

If passing the Gemini LLM works correctly, there will be no further changes needed. However, if any fixes are required to enable this functionality, it would be appreciated if you could include those changes in this PR.