engindeniz / vitis Goto Github PK
View Code? Open in Web Editor NEW[ICCV 2023 CLVL Workshop] Zero-Shot and Few-Shot Video Question Answering with Multi-Modal Prompts
Home Page: https://arxiv.org/abs/2309.15915
License: Apache License 2.0
[ICCV 2023 CLVL Workshop] Zero-Shot and Few-Shot Video Question Answering with Multi-Modal Prompts
Home Page: https://arxiv.org/abs/2309.15915
License: Apache License 2.0
First of all, thanks to your great work!
I wonder about the arguments related with "visual prompts".
To run your code on custom dataset I'm preparing to extract features.
When I tried to extract visual features, i got some troubles.
self.encoder = CLIPEncoder(config, self.args.prompt_type, self.args.visual_prompt_projection_factor,
AttributeError: 'Namespace' object has no attribute 'visual_prompt_projection_factor'
Actually, not only about 'visual_prompt_projection_factor'.
I also can't find about these informations in your args file.
self.encoder = CLIPEncoder(config, self.args.prompt_type, self.args.visual_prompt_projection_factor,
self.args.visual_prompt_projection_layer, self.args.visual_model_hidden_size,
self.args.visual_prompt_num_layer, self.args.visual_prompt_num_tokens,
self.args.visual_prompt_initialization, self.args.visual_prompt_patch_size,
self.args.visual_prompt_dropout_rate, self.args.visual_prompt_like_linear_layer_each_transformer_layer)
And when extract the features, I think we would not use promts and projections right?
Would you help me to shoot this troubles?
I think this will affect to follow-up tasks like training and testing.
Great work!
Can you provide trained weights and code to run inference locally on my own videos?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.