🔒 First time using AI on my machine need some help

[little knot]

import torch
from diffusers import FluxPipeline

pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-schnell", torch_dtype=torch.bfloat16)
pipe.enable_model_cpu_offload()

prompt = "A lot of people takeing a photo of you"
image = pipe(
    prompt,
    height = 1024,
    width = 1024,
    guidance_scale = 3.5,
    num_inference_steps = 60,
    max_sequence_lenth=255,
    generator = torch.Generator("cpu").manual_seed(0)
).images(0)
image.save("photo_of_you.png")

This is my code. I'm trying to make an image to test it. When I try to run it I get this output. I posted an image of it.

[floral lark]

@little knot

Python help channel opened

Remember to:

• Ask your Python question, not if you can ask or if there's an expert who can help.
• Show a code sample as text (rather than a screenshot) and the error message, if you've got one.
• Explain what you expect to happen and what actually happens.

:warning: Do not pip install anything that isn't related to your question, especially if asked to over DMs.

Closes after a period of inactivity, or when you send !close.

[ebon tusk]

did you try to install sentencepiece like the error suggested?

[little knot]

IDK if my code is based on it

[merry oracle]

i swear to god i watch a video on yt and a same example comes in python

isnt that what the guy used?

[little knot]

I didn't watch a yt vid

so idk

[merry oracle]

well u dis same idk why isnt working

[little knot]

OK now all I have is this

aka nothing is being produced and it completes at that

[merry oracle]

do you have the photo?

[little knot]

Nothing is being produced

it stops at that

[merry oracle]

to continue please pay 900 dollars

lol

lemme try it

[little knot]

4/7 pipeline and tells me tokenizer has a problem. It needs to be converted from slow-tokenizer. IDK how to do that

or what is cousing that

[merry oracle]

so whatever the promt is it gonna put on the .png ?

[little knot]

no I named the file output

image.save("photo_of_you.png")

[merry oracle]

thats the output yes /<

[little knot]

so that should be whats on the .png

[merry oracle]

torch taking 90 working days to install 😡

[little knot]

MAn I had to wait for a long time,

[merry oracle]

they install 900 other packages

[merry oracle]

wtf

3.8 gb?

[little knot]

9.96gb

[merry oracle]

😮

[little knot]

I was waiting

[merry oracle]

i should have started an venv

[merry oracle]

pip install transformers

[little knot]

did

[merry oracle]

this needs 9gb?

na man im not doing it

[little knot]

ah, yeah I installed transformers, accelerate, sentencepiece

[merry oracle]

what

[little knot]

with pip

[merry oracle]

xd

lol

okay

im not trusting them

[little knot]

protobuf and this

[merry oracle]

so the problem was you didnt have those packages ?

[little knot]

nope

all four of those got me all the way to where I have one problem left

This one

[merry oracle]

send in text not photo

[little knot]

coping txt from terminal is so fun

You set add_prefix_space. The tokenizer needs to be converted from the slow tokenizers

[merry oracle]

you can just select it and ctrl+c

if you kali linux ctrl+shift+c

[little knot]

I sent it

You set add_prefix_space. The tokenizer needs to be converted from the slow tokenizers

[merry oracle]

ye i saw it 😄

here is a chat

some had same problem but im lazy to read

https://github.com/huggingface/diffusers/issues/9063

maybe you find solution there

[little knot]

nope he seemed to have a problem with his venv, I'm not using one of those

but thx for looking

[merry oracle]

wait for a big brain

[little knot]

will do

[floral lark]

Python help channel closed

This help channel has been closed and it's no longer possible to send messages here. If your question wasn't answered, feel free to create a new post in #1035199133436354600. To maximize your chances of getting a response, check out this guide on asking good questions.