# Stable diffusion text\_to\_image.py discussion

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fstable-diffusion-text-to-image-py-discussion%2F38865%3E%5C
**Category:** 🧨 Diffusers\
**Created:** [May 6, 2023, 7:42am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fstable-diffusion-text-to-image-py-discussion%2F38865 "2023-05-06T07:42:14Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![barry556652](https://avatars.discourse-cdn.com/v4/letter/b/b77776/32.png) [@barry556652](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fbarry556652)%5C
**Post date:** [May 6, 2023, 7:42am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fstable-diffusion-text-to-image-py-discussion%2F38865%2F1 "2023-05-06T07:42:14Z")

</div>

Hi!  
I want to change  
**text\_encoder = CLIPTextModel.from\_pretrained(args.pretrained\_model\_name\_or\_path, subfolder=“text\_encoder”, revision=args.revision)** in text\_to\_image.py 464 line to **torch.nn.Embedding(77, 768)**

I don’t want to use CLIP embedding because this dataset is very special.

When i change this code, it ran this error.

 ![image](https://us1.discourse-cdn.com/hellohellohello/original/2X/5/5f8c8cf2edd0fc760d5e134ddc5cfb6d56780313.png)  
 ![image](https://us1.discourse-cdn.com/hellohellohello/original/2X/3/33cadd3c8ca4f4b79dc2492ae7567a13bc1f5ed0.png)

what can i do?Thx

---

<div class="post-metadata">

**Author:** ![williamberman](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/williamberman/32/14112_2.png) [@williamberman](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fwilliamberman)%5C
**Post date:** [May 22, 2023, 6:19pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fstable-diffusion-text-to-image-py-discussion%2F38865%2F2 "2023-05-22T18:19:07Z")

</div>

hey @barry556652 can you post a reproducible code snippet?