search

lucidrains / deep-daze

Simple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network). Technique was originally created by https://twitter.com/advadnoun
MIT License
4.37k stars 327 forks source link

Working on the project #124

Closed Fliens closed 3 years ago

Fliens commented 3 years ago

Hi I really wanto to contribute to this project but I'm having some difficulties

I'm using vscode and I've already cloned the project(locally, not in conda) but I'm having difficulties running the project in the console. I've already switched the interpreter to conda but it seems like it's not working

I want to work on the stories so that the text is splitted by a separator and not every 5 words or so:)

NotNANtoN commented 3 years ago

Well, what's your error message? I assume you installed all requirements? Furthermore, make sure you have torch version 1.7.1.

Fliens commented 3 years ago

I think I got it to run The problem was that I tried to run it with the wrong python.exe But through finding the python.exe in the conda env and the executing the project with that .exe it worked

But now I'm getting this error:

from .clip import load, tokenize ImportError: attempted relative import with no known parent package

NotNANtoN commented 3 years ago

That might be a good place to start contributing! ;)

I think this line should have the full path: from deep_daze.clip import load, tokenized

For me it works as it is, but Windows might have some problems there.