Skip to content

[Bug]: Training embeddings stop after 1 epoch #10778

Description

@RONNYKHALIL

Is there an existing issue for this?

  • I have searched the existing issues and checked the recent builds/commits

What happened?

Training embeddings suddenly stopped working after the 1st epoch.

After some digging, I narrowed it down to the "image_embeddings", and perhaps a font issue. I was able to solve the issue (train embeds through completion) by unchecking "Save images with embedding in PNG chunks"

Steps to reproduce the problem

Train embeddings as usual, but they stop after the first epoch (error logs below)

What should have happened?

Training should've completed as per normal

Commit where the problem happens

20ae71f

What Python version are you running on ?

Python 3.10.x

What platforms do you use to access the UI ?

Linux, Other/Cloud

What device are you running WebUI on?

AMD GPUs (RX 6000 above)

What browsers do you use to access the UI ?

Google Chrome

Command Line Arguments

No

List of extensions

Auto-Photoshop-StableDiffusion-Plugin
SadTalker
adetailer
clip-interrogator-ext deforum-for-automatic1111-webui
gif2gif openOutpaint-webUl-extension sd-dynamic-prompts
sd-webui-controlnet
sd-webui-infinite-image-browsing sd-webui-text2video
sd_civitai_extension sd_web_ui_preset_utils stable-diffusion-webui-state
ultimate-upscale-for-automatic1111
unprompted
LDSR
Lora
ScuNET
SwinIR
prompt-bracket-checker

Console logs

Traceback (most recent call last):
  File "/notebooks/sd/stable-diffusion-webui/modules/textual_inversion/textual_inversion.py", line 612, in train_embedding
    captioned_image = caption_image_overlay(image, title, footer_left, footer_mid, footer_right)
  File "/notebooks/sd/stable-diffusion-webui/modules/textual_inversion/image_embedding.py", line 150, in caption_image_overlay
    font = ImageFont.truetype(textfont, fontsize)
  File "/usr/local/lib/python3.9/dist-packages/PIL/ImageFont.py", line 976, in truetype
    return freetype(font)
  File "/usr/local/lib/python3.9/dist-packages/PIL/ImageFont.py", line 973, in freetype
    return FreeTypeFont(font, size, index, encoding, layout_engine)
  File "/usr/local/lib/python3.9/dist-packages/PIL/ImageFont.py", line 253, in __init__
    load_from_bytes(font)
  File "/usr/local/lib/python3.9/dist-packages/PIL/ImageFont.py", line 233, in load_from_bytes
    self.font_bytes = f.read()
AttributeError: 'FreeTypeFont' object has no attribute 'read'

Additional information

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugReport of a confirmed bugbug-reportReport of a bug, yet to be confirmed

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions