Upload folder using huggingface_hub

.env

@@ -0,0 +1,2 @@
+REPLICATE_API_TOKEN=r8_DAzyOBdCwUdt0b26ZMPWLyvyHTh55uh2Lwb3c
+OPENAI_API_KEY=sk-proj-6lTXmIwTYmNo7uUpQwujT3BlbkFJDMVzyH5hzblFbgYLLMCP

index.py

@@ -23,22 +23,42 @@ REPLICATE_API_TOKEN = os.getenv('REPLICATE_API_TOKEN')
 
 client = OpenAI()
 
-def main(img, strength):
+def main(img, strength, prompt):
     mask = img['layers'][0]
+    if is_transparent(mask) == True:
+        mask_img = None
+        mask_base_64 = None
+    else:
+        mask_img = create_mask_image(mask)
+        
 
     # Match prompt strength from .4 to 1 (total destruction)
     prompt_strength = round(-0.6 * strength + 1, 2)
 
     base_image = Image.fromarray(img['background'].astype('uint8'))
-    img_base_64 = img_to_base64(base_image)
 
-    if is_transparent(mask) == True:
-        mask_base_64 = None
-    else:
-        mask_img = create_mask_image(mask)
-        mask_base_64 = img_to_base64(mask_img)
+    # Resize the starter image if either dimension is larger than 768 pixels
+    if base_image.size[0] > 768 or base_image.size[1] > 768:
+        # Calculate the new size while maintaining the aspect ratio
+        if base_image.size[0] > base_image.size[1]:
+            # Width is larger than height
+            new_width = 768
+            new_height = int((768 / base_image.size[0]) * base_image.size[1])
+        else:
+            # Height is larger than width
+            new_height = 768
+            new_width = int((768 / base_image.size[1]) * base_image.size[0])
+        
+        # Resize the image
+        base_image = base_image.resize((new_width, new_height), Image.LANCZOS)
+        if mask_img is not None:
+            mask_img = mask_img.resize((new_width, new_height), Image.LANCZOS)
+
+    img_base_64 = img_to_base64(base_image)
+    mask_base_64 = img_to_base64(mask_img)
 
-    prompt = call_openai(img_base_64)
+    #prompt = call_openai(img_base_64)
+    prompt = prompt
     #prompt = "The image shows a person wearing sleek, over-ear headphones with a matte finish and a cool, light beige color (Pantone 7527 C), captured under soft, diffused natural lighting, emphasizing the smooth and minimalist design of the headphones."
 
     output_urls = generate_image(prompt, img_base_64, mask_base_64, prompt_strength)
@@ -49,7 +69,7 @@ def main(img, strength):
 def generate_image(prompt, img, mask, prompt_strength):
     input_data = {
         "image": img,
-        "prompt": prompt + " expensive",
+        "prompt": prompt,
         "refine": "no_refiner",
         "scheduler": "K_EULER",
         "lora_scale": 0.8,
@@ -160,7 +180,7 @@ black_brush = gr.Brush(colors=["#000000"], default_color="#000000", color_mode="
 # Using the ImageEditor component to enable drawing on the image with limited colors
 demo = gr.Interface(
     fn=main,
-    inputs=[gr.ImageEditor(brush=black_brush), gr.Slider(0, 1, step=0.025, value=0.5, label="Image Strength")],
+    inputs=[gr.ImageEditor(brush=black_brush), gr.Slider(0, 1, step=0.025, value=0.5, label="Image Strength"), gr.Textbox(label="Describe the object in as much detail as possible")],
     #outputs=[gr.Image(type="pil"), gr.Image(type="pil"), gr.Image(type="pil"), gr.Image(type="pil")]
     outputs=["image", "image", "image", "image"]
 )