Model card auto-generated by SimpleTuner

README.md

@@ -13,7 +13,7 @@ inference: true
 
 ---
 
-# SahinFlux
+# SahinFLUX
 
 This is a LoRA derived from [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev).
 
@@ -24,7 +24,7 @@ The main validation prompt used during training was:
 
 
 ```
-A man in front of a Sahin car
+A man in front of a white Sahin car
 ```
 
 ## Validation settings
@@ -48,9 +48,9 @@ You may reuse the base model text encoder for inference.
 
 ## Training settings
 
-- Training epochs: 399
-- Training steps: 2000
-- Learning rate: 8e-07
+- Training epochs: 5
+- Training steps: 100
+- Learning rate: 1.0
 - Effective batch size: 1
   - Micro-batch size: 1
   - Gradient accumulation steps: 1
@@ -60,7 +60,7 @@ You may reuse the base model text encoder for inference.
 - Optimizer: AdamW, stochastic bf16
 - Precision: Pure BF16
 - Xformers: Not used
-- LoRA Rank: 16
+- LoRA Rank: 4
 - LoRA Alpha: None
 - LoRA Dropout: 0.1
 - LoRA initialisation style: default
@@ -70,7 +70,7 @@ You may reuse the base model text encoder for inference.
 
 ### Sahin
 - Repeats: 0
-- Total number of images: 5
+- Total number of images: 18
 - Total number of aspect buckets: 1
 - Resolution: 1 megapixels
 - Cropped: True
@@ -86,11 +86,11 @@ import torch
 from diffusers import DiffusionPipeline
 
 model_id = 'black-forest-labs/FLUX.1-dev'
-adapter_id = 'adaozer/SahinFlux'
+adapter_id = 'adaozer/SahinFLUX'
 pipeline = DiffusionPipeline.from_pretrained(model_id)
 pipeline.load_lora_weights(adapter_id)
 
-prompt = "A man in front of a Sahin car"
+prompt = "A man in front of a white Sahin car"
 
 
 pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')