Model card auto-generated by SimpleTuner

README.md

@@ -15,12 +15,12 @@ inference: true
 widget:
 - text: 'unconditional (blank prompt)'
   parameters:
-    negative_prompt: 'blurry'
+    negative_prompt: 'blurry, floating leaves, multiple plants'
   output:
     url: ./assets/image_0_0.png
-- text: 'A pirate ship heading out to sea'
+- text: 'One canola seedling that is about 11 days old in a bright blue cylindrical cup that has fertilizer granules on a bright blue background'
   parameters:
-    negative_prompt: 'blurry'
+    negative_prompt: 'blurry, floating leaves, multiple plants'
   output:
     url: ./assets/image_1_0.png
 ---
@@ -35,7 +35,7 @@ The main validation prompt used during training was:
 
 
 ```
-A pirate ship heading out to sea
+One canola seedling that is about 11 days old in a bright blue cylindrical cup that has fertilizer granules on a bright blue background
 ```
 
 ## Validation settings
@@ -59,8 +59,8 @@ You may reuse the base model text encoder for inference.
 
 ## Training settings
 
-- Training epochs: 1
-- Training steps: 100
+- Training epochs: 0
+- Training steps: 5
 - Learning rate: 0.00105
 - Max grad norm: 0.01
 - Effective batch size: 1
@@ -81,11 +81,11 @@ You may reuse the base model text encoder for inference.
 
 ## Datasets
 
-### wikiart_s
+### canola-test
 - Repeats: 1
-- Total number of images: 40
-- Total number of aspect buckets: 4
-- Resolution: 1.0 megapixels
+- Total number of images: 10
+- Total number of aspect buckets: 1
+- Resolution: 1024 px
 - Cropped: False
 - Crop style: None
 - Crop aspect: None
@@ -104,8 +104,8 @@ adapter_id = 'rdeinla/test'
 pipeline = DiffusionPipeline.from_pretrained(model_id)
 pipeline.load_lora_weights(adapter_id)
 
-prompt = "A pirate ship heading out to sea"
-negative_prompt = 'blurry'
+prompt = "One canola seedling that is about 11 days old in a bright blue cylindrical cup that has fertilizer granules on a bright blue background"
+negative_prompt = 'blurry, floating leaves, multiple plants'
 pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
 image = pipeline(
     prompt=prompt,