Upload folder using huggingface_hub

api/index.py

@@ -60,9 +60,20 @@ def image_classifier(moodboard, starter_image, image_strength, prompt):
     pil_image = Image.fromarray(moodboard.astype('uint8'))
     starter_image_pil = Image.fromarray(starter_image.astype('uint8'))
 
-    # Resize the starter image if it's larger than 768x768
+    # Resize the starter image if either dimension is larger than 768 pixels
     if starter_image_pil.size[0] > 768 or starter_image_pil.size[1] > 768:
-        starter_image_pil = starter_image_pil.resize((768, 768), Image.LANCZOS)
+        # Calculate the new size while maintaining the aspect ratio
+        if starter_image_pil.size[0] > starter_image_pil.size[1]:
+            # Width is larger than height
+            new_width = 768
+            new_height = int((768 / starter_image_pil.size[0]) * starter_image_pil.size[1])
+        else:
+            # Height is larger than width
+            new_height = 768
+            new_width = int((768 / starter_image_pil.size[1]) * starter_image_pil.size[0])
+        
+        # Resize the image
+        starter_image_pil = starter_image_pil.resize((new_width, new_height), Image.LANCZOS)
     
     openai_response = call_openai(pil_image)
     openai_response = openai_response.replace('moodboard', '')
@@ -84,7 +95,8 @@ def image_classifier(moodboard, starter_image, image_strength, prompt):
         "image": "data:image/jpeg;base64," + starter_image_base64, 
         "apply_watermark": False,
         "num_inference_steps": 25,
-        "prompt_strength": 1-image_strength
+        "prompt_strength": 1-image_strength,
+        "num_outputs": 3,
     }
     
     output = replicate.run(
@@ -97,9 +109,21 @@ def image_classifier(moodboard, starter_image, image_strength, prompt):
     print(image_url)
     response = requests.get(image_url)
     print(response)
-    img = Image.open(io.BytesIO(response.content))
+    img1 = Image.open(io.BytesIO(response.content))
+
+    image_url = output[1]
+    print(image_url)
+    response = requests.get(image_url)
+    print(response)
+    img2 = Image.open(io.BytesIO(response.content))
+
+    image_url = output[2]
+    print(image_url)
+    response = requests.get(image_url)
+    print(response)
+    img3 = Image.open(io.BytesIO(response.content))
     
-    return img  # Return the image object
+    return [img1, img2, img3]  # Return the image object
 
 
 # app = Flask(__name__)
@@ -108,5 +132,5 @@ def image_classifier(moodboard, starter_image, image_strength, prompt):
 # @app.route("/")
 # def index():
 
-demo = gr.Interface(fn=image_classifier, inputs=["image", "image", gr.Slider(0, 1, step=0.025, value=0.2, label="Image Strength"), "text"], outputs="image")
-demo.launch(share=True)
+demo = gr.Interface(fn=image_classifier, inputs=["image", "image", gr.Slider(0, 1, step=0.025, value=0.2, label="Image Strength"), "text"], outputs=["image", "image", "image"])
+demo.launch(share=False)