use new images

app.py

@@ -30,6 +30,8 @@ model = model.to(device).half()
 clip_model, preprocess = clip.load("ViT-L/14", device=device)
 gender_learn = load_learner('gender_model.pkl')
 gender_labels = gender_learn.dls.vocab
+beard_learn = load_learner('facial_hair_model.pkl')
+beard_labels = beard_learn.dls.vocab
 
 n_inputs = 5
 
@@ -123,28 +125,39 @@ def run_image_mixer(args):
 def is_female(img):
     pred,pred_idx,probs = gender_learn.predict(img)
     return float(probs[0]) > float(probs[1])
+    
+def has_beard(img):
+    pred,pred_idx,probs = beard_learn.predict(img)
+    return float(probs[0]) > float(probs[1])
 
 
 import gradio
 
 def boutsify(person):
+    portrait_path = "bouts_m1.jpg"
     female_detected = is_female(person)
     
     if female_detected:
-        print("Picture of a female")
+        print("This is a female portrait")
+        portrait_path = "bouts_f1.jpg"
+    else:
+        print("This is a male portrait")
+        if has_beard(person):
+            print("The person has a beard")
+            portrait_path = "bouts_mb1.jpg"
     
     person_image = Image.fromarray(person)
     
     inputs = [
         "Image", "Image", "Image", "Image", "Nothing",
         "","","","","",
-        Image.open("bouts_f1.jpg").convert("RGB") if female_detected else Image.open("bouts_m1.jpg").convert("RGB"),
-        Image.open("bouts_banner.png").convert("RGB"),
-        Image.open("bouts_city.png").convert("RGB"),
+        Image.open(portrait_path).convert("RGB"),
+        Image.open("boutsback_o1.png").convert("RGB"),
+        Image.open("boutsback_o2.png").convert("RGB"),
         person_image,
-        Image.open("bouts_m24.jpg").convert("RGB"),
-        1.2,1,1,1.4,1,
-        3.0, 1, random.randrange(0, 10000), 40,
+        "",
+        1.1,1,1,1.4,1,
+        3.0, 1, random.randrange(0, 10000), 50,
     ]
     
     return run_image_mixer(inputs)