AbstractPhil commited on
Commit
10c174c
·
verified ·
1 Parent(s): 0910e65

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -4
README.md CHANGED
@@ -108,7 +108,7 @@ How absurd and difficult anything SD15 has been to debug.
108
  Okay I am correctly converting the valuation and can now properly test the unet for diffusion testing.
109
 
110
  # CKPT Bumbles
111
- Apologies, the safetensors ARE comfyui formatted... CKPT. :V
112
 
113
  Rename the extension to ckpt, because it's clearly incorrect. I'll convert them asap, my apologies for not micro-managing more closely.
114
 
@@ -116,7 +116,7 @@ I renamed epoch 4 in the repo.
116
 
117
  # Training continues.
118
 
119
- Trainer updated aka trainer.py with the updated version that handles checkpoint saving and loading using the correct comfyui script.
120
 
121
  It'll automatically load and run in colab, additionally it's prepared to continue training from the most recent checkpoint here. You can point it at your own repo to load and save.
122
 
@@ -291,8 +291,7 @@ Requires SD1.5 VAE and text encoder (not included in checkpoint).
291
  - ❓ Optimal `block_penalty_weight` unknown (currently 0.05)
292
  - ❓ May require tuning lambda bounds for different datasets
293
  - ❓ Inference quality unvalidated
294
- - ❌ Not compatible with ComfyUI without conversion (details TBD)
295
- - ❌ No SD1.5 components included (VAE, text encoder)
296
 
297
  ## Future Work
298
 
 
108
  Okay I am correctly converting the valuation and can now properly test the unet for diffusion testing.
109
 
110
  # CKPT Bumbles
111
+ Apologies, the safetensors ARE ComfyUI formatted... CKPT. :V
112
 
113
  Rename the extension to ckpt, because it's clearly incorrect. I'll convert them asap, my apologies for not micro-managing more closely.
114
 
 
116
 
117
  # Training continues.
118
 
119
+ Trainer updated aka trainer.py with the updated version that handles checkpoint saving and loading using the correct ComfyUI script.
120
 
121
  It'll automatically load and run in colab, additionally it's prepared to continue training from the most recent checkpoint here. You can point it at your own repo to load and save.
122
 
 
291
  - ❓ Optimal `block_penalty_weight` unknown (currently 0.05)
292
  - ❓ May require tuning lambda bounds for different datasets
293
  - ❓ Inference quality unvalidated
294
+
 
295
 
296
  ## Future Work
297