control-edgedrawing-cv480edpf-rect-fp16-batch32-checkpoint-6696

README.md

@@ -174,11 +174,15 @@ restarted from 0 and `--mixed-precision="fp16"`.
 
 TODO: Why did I end up with less images after I added more images? fastdup suddenly finds even more duplicates. Is fastdup default threshold=0.9 too aggressive?
 
-**Experiment 6.1 - control-edgedrawing-cv480edpf-rect-fp16-gas16-checkpoint-XXXXX**
+**Experiment 6.1 - control-edgedrawing-cv480edpf-rect-fp16-batch32-checkpoint-6696**
 
-see experiment 6.0. `--gradient_accumulation_steps=16`.
+see experiment 6.0. restarted from 0 with `--train_batch_size=2 --gradient_accumulation_steps=16`.
 
-1 epoch = 107122 * 2 / 16 = 13390 steps per epoch
+1 epoch = 107122 * 2 / 32 = 6696 steps per epoch
+
+**Experiment 6.2 - control-edgedrawing-cv480edpf-rect-fp16-batch32-drop50-checkpoint-6696**
+
+see experiment 6.1. restarted from 0 with `--proportion_empty_prompts=0.5`.
 
 # Ideas
 

control-edgedrawing-cv480edpf-rect-fp16-batch32-checkpoint-6696.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a26d8771ec9084d7949db8bc044f384f7c0cd9bdf2d97b91e4039469ca76a21
+size 722598616

control-edgedrawing-cv480edpf-rect-fp16-batch32-checkpoint-6696.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efdf2702b64948c893057d8ccfe3e39ad9fe9e75cd9f139b393263ccdb146719
+size 59653528