Update app.py

app.py

@@ -1,32 +1,56 @@
 import gradio as gr
 import requests
 from PIL import Image
-from transformers import BlipProcessor, BlipForConditionalGeneration
+from transformers import BlipProcessor, BlipForConditionalGeneration, pipeline
 import time
 
+# Carregando o modelo BLIP para geração de legendas
 processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
-model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")
+model_blip = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")
+
+# Carregando um modelo de geração de texto (exemplo: GPT-2)
+generator = pipeline('text-generation', model='gpt2')
 
 def caption(img, min_len, max_len):
     raw_image = Image.open(img).convert('RGB')
-  
     inputs = processor(raw_image, return_tensors="pt")
-    
-    out = model.generate(**inputs, min_length=min_len, max_length=max_len)
+    out = model_blip.generate(**inputs, min_length=min_len, max_length=max_len)
     return processor.decode(out[0], skip_special_tokens=True)
 
+def generate_nutritional_info(food_description):
+    # Gerando informações nutricionais com base na descrição do alimento
+    prompt = f"Provide detailed nutritional information about {food_description}."
+    result = generator(prompt, max_length=150, num_return_sequences=1)
+    return result[0]['generated_text']
+
 def greet(img, min_len, max_len):
     start = time.time()
-    result = caption(img, min_len, max_len)
+    
+    # Passo 1: Gerar legenda para a imagem
+    food_description = caption(img, min_len, max_len)
+    
+    # Passo 2: Gerar informações nutricionais com base na legenda
+    nutritional_info = generate_nutritional_info(food_description)
+    
     end = time.time()
     total_time = str(end - start)
-    result = result + '\n' + total_time + ' seconds'
+    
+    # Combinando resultados
+    result = f"Food Description: {food_description}\n\nNutritional Information:\n{nutritional_info}\n\nGenerated in {total_time} seconds."
     return result
 
-iface = gr.Interface(fn=greet, 
-                     title='Blip Image Captioning Large', 
-                     description="[Salesforce/blip-image-captioning-large](https://huggingface.co/Salesforce/blip-image-captioning-large) Runs on CPU",
-                     inputs=[gr.Image(type='filepath', label='Image'), gr.Slider(label='Minimum Length', minimum=1, maximum=1000, value=30), gr.Slider(label='Maximum Length', minimum=1, maximum=1000, value=100)], 
-                     outputs=gr.Textbox(label='Caption'),
-                     theme = gr.themes.Base(primary_hue="teal",secondary_hue="teal",neutral_hue="slate"),)
+# Interface Gradio
+iface = gr.Interface(
+    fn=greet,
+    title='Nutritionist Agent with BLIP and GPT-2',
+    description="Upload an image of food, and the agent will describe it and provide nutritional information.",
+    inputs=[
+        gr.Image(type='filepath', label='Image'),
+        gr.Slider(label='Minimum Length', minimum=1, maximum=1000, value=30),
+        gr.Slider(label='Maximum Length', minimum=1, maximum=1000, value=100)
+    ],
+    outputs=gr.Textbox(label='Result'),
+    theme=gr.themes.Base(primary_hue="teal", secondary_hue="teal", neutral_hue="slate"),
+)
+
 iface.launch()