Update app.py

app.py

@@ -169,7 +169,7 @@ pipe = load_and_prepare_model()
 checkpoint = "ford442/Phi-3.5-mini-instruct-bf16"
 #captioner = pipeline(model="ydshieh/vit-gpt2-coco-en",device='cuda:0', task="image-to-text")
 #captioner_2 = pipeline(model="Salesforce/blip-image-captioning-base",device='cuda', task="image-to-text")
-captioner_2 = pipeline(model="ford442/blip-image-to-text-large-bf16",device='cuda', task="image-to-text")
+#captioner_2 = pipeline(model="ford442/blip-image-to-text-large-bf16",device='cuda', task="image-to-text")
 #model5 = Blip2ForConditionalGeneration.from_pretrained("ford442/blip2-image-to-text-bf16").to('cuda')
 #processor5 = Blip2Processor.from_pretrained("ford442/blip2-image-to-text-bf16", device_map='cuda')
 #txt_tokenizer = AutoTokenizer.from_pretrained(checkpoint, device_map='cuda', add_prefix_space=False)
@@ -408,7 +408,7 @@ def generate_30(
             #sd_image_b.resize((height,width), Image.LANCZOS)
             sd_image_b.resize((224,224), Image.LANCZOS)
             #caption.append(captioner(sd_image_b))
-            caption.append(captioner2(sd_image_b))
+            ##caption.append(captioner2(sd_image_b))
             #caption.append(captioner_3(sd_image_b))
             caption_2.append(captioning(sd_image_b))
         else:
@@ -418,7 +418,7 @@ def generate_30(
             #sd_image_c.resize((height,width), Image.LANCZOS)
             sd_image_c.resize((224,224), Image.LANCZOS)
             #caption.append(captioner(sd_image_c))
-            caption.append(captioner2(sd_image_c))
+            #caption.append(captioner2(sd_image_c))
             #caption.append(captioner_3(sd_image_c))
             caption_2.append(captioning(sd_image_c))
         else:
@@ -428,7 +428,7 @@ def generate_30(
             #sd_image_d.resize((height,width), Image.LANCZOS)
             sd_image_d.resize((224,224), Image.LANCZOS)
             #caption.append(captioner(sd_image_d))
-            caption.append(captioner2(sd_image_d))
+            #caption.append(captioner2(sd_image_d))
             #caption.append(captioner_3(sd_image_d))
             caption_2.append(captioning(sd_image_d))
         else:
@@ -438,7 +438,7 @@ def generate_30(
             #sd_image_e.resize((height,width), Image.LANCZOS)
             sd_image_e.resize((224,224), Image.LANCZOS)
             #caption.append(captioner(sd_image_e))
-            caption.append(captioner2(sd_image_e))
+            #caption.append(captioner2(sd_image_e))
             #caption.append(captioner_3(sd_image_e))
             caption_2.append(captioning(sd_image_e))
         else:
@@ -559,7 +559,7 @@ def generate_60(
             #sd_image_b.resize((height,width), Image.LANCZOS)
             sd_image_b.resize((224,224), Image.LANCZOS)
             #caption.append(captioner(sd_image_b))
-            caption.append(captioner2(sd_image_b))
+            #caption.append(captioner2(sd_image_b))
             #caption.append(captioner_3(sd_image_b))
             caption_2.append(captioning(sd_image_b))
         else:
@@ -711,7 +711,7 @@ def generate_90(
             #sd_image_b.resize((height,width), Image.LANCZOS)
             sd_image_b.resize((224,224), Image.LANCZOS)
             #caption.append(captioner(sd_image_b))
-            caption.append(captioner2(sd_image_b))
+            #caption.append(captioner2(sd_image_b))
             #caption.append(captioner_3(sd_image_b))
             caption_2.append(captioning(sd_image_b))
         else: