Model card auto-generated by SimpleTuner

README.md

@@ -20,7 +20,7 @@ widget:
     negative_prompt: 'blurry, cropped, ugly'
   output:
     url: ./assets/image_0_0.png
-- text: 'A 3D vfx of flame effect in red and yellow, glazing against black background'
+- text: 'A 2D vfx of flame effect in red and yellow, glazing against black background'
   parameters:
     negative_prompt: 'blurry, cropped, ugly'
   output:
@@ -33,7 +33,7 @@ This is a standard PEFT LoRA derived from [black-forest-labs/FLUX.1-dev](https:/
 
 The main validation prompt used during training was:
 ```
-A 3D vfx of flame effect in red and yellow, glazing against black background
+A 2D vfx of flame effect in red and yellow, glazing against black background
 ```
 
 
@@ -59,8 +59,8 @@ You may reuse the base model text encoder for inference.
 
 ## Training settings
 
-- Training epochs: 40
-- Training steps: 20000
+- Training epochs: 0
+- Training steps: 1000
 - Learning rate: 0.0001
   - Learning rate schedule: polynomial
   - Warmup steps: 100
@@ -85,9 +85,9 @@ You may reuse the base model text encoder for inference.
 
 ## Datasets
 
-### mapledata_3D
+### mapledata_2D
 - Repeats: 5
-- Total number of images: ~664
+- Total number of images: ~2208
 - Total number of aspect buckets: 1
 - Resolution: 1.0 megapixels
 - Cropped: False
@@ -108,7 +108,7 @@ adapter_id = 'binarydaddy/simpletuner-lora'
 pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
 pipeline.load_lora_weights(adapter_id)
 
-prompt = "A 3D vfx of flame effect in red and yellow, glazing against black background"
+prompt = "A 2D vfx of flame effect in red and yellow, glazing against black background"
 
 
 ## Optional: quantise the model to save on vram.