CogView2

CogView2 copied to clipboard

what does parameter max-inference-batch-size truly mean?

1

I don't clearly understand what the parameter max-inference-batch-size means， What is the relationship between it and batch_size?

whdcwsj

How many tokens should be trained when pretraining the text-to-image generation in the paper?

4

runzeer

Image variations

2

Hi! Great research and implementation! Is it possible to make variations of the input image? As I read the paper this should be possible. Maybe you can suggest how to...

KyriaAnnwyn

ask for suggestions on "RuntimeError: CUDA out of memory. Tried to allocate..." error

10

Hi! I am new to Pytorch and would like to try this fantastic text-to-image project. I just tried to clone and run text2image.sh for only predict (not training) on different...

undeathable

Hi, is there any way to run three stages on three separate GPUs? 24G memory still meets out of memory error.

1049451037

Hi, how can I determine the image size when using `--only-first-stage` flag . the images are in 480 pixels but I would expect 256 pixels. Thanks!

Limbicnation

Replicate CUDA out of memory

7

replicate fails after some images CUDA out of memory. Tried to allocate 1.22 GiB (GPU 0; 39.59 GiB total capacity; 30.86 GiB already allocated; 1.07 GiB free; 34.03 GiB reserved...

loboere

Better download links

2

Would it be possible to provide better download links? The downloads from https://model.baai.ac.cn/model-detail/100041 fail before they are finished and the automatic downloads take upwards to 10 hours.

alvitawa

I have not seen this one before: Prediction was canceled. I press the Submit button, it turns light grey with a spinning circle. Error message appears in the right panel....

dza6549

web demo

12

Hi, thanks for releasing the code and models, would you be interested in creating a web demo for CogView2 using Gradio on Hugging Face? The Hub offers free hosting, and...

AK391