Update README.md
Browse files
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
|
| 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
|
| 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 |
-
|
| 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 |
|