Update app.py

app.py

@@ -3,49 +3,57 @@ from fastai.vision.all import *
 import openai
 import os
 
-
 openai.api_key = os.getenv("OPENAI_API_KEY")
 
-# Load your trained model (you should replace 'model.pkl' with the path to your model file)
+# Load the model
 learn = load_learner('model.pkl')
 
-# Define the labels for the output
+# Define the labels
 labels = learn.dls.vocab
 
+# Define a function for generating text
+def generate_text(prompt):
+  response = openai.Completion.create(
+      engine="davinci",
+      prompt=prompt,
+      max_tokens=1024,
+      n=1,
+      stop=None,
+      temperature=0.7,
+  )
+  return response.choices[0].text.strip()
+
+# Define a function to handle user queries
+def handle_query(query, chat_history):
+  response = openai.ChatCompletion.create(
+      model="gpt-3.5-turbo",
+      messages=[{"role": "system", "content": "You are a helpful assistant you kow about plant Disease."},
+               {"role": "user", "content": query}] + chat_history
+  )
+  return response.choices[0].message['content']
+
 # Define the prediction function
 def predict(img):
-    img = PILImage.create(img)
-    pred, pred_idx, probs = learn.predict(img)
-    prediction = {labels[i]: float(probs[i]) for i in range(len(labels))}
-
-    # Now generate a chat/text response based on the model's prediction.
-    chat_prompt = f"The image likely depicts the following: {pred}. What can I help you with next?"
-
-    # Ensure that you have set the OPENAI_API_KEY environment variable,
-    # as we will use it to interact with OpenAI's GPT-3 model.
-    response = openai.Completion.create(
-        engine="text-davinci-003",  # Adjust the engine as needed for your use-case
-        prompt=chat_prompt,
-        max_tokens=1024,
-        n=1,
-        stop=None,
-        temperature=0.7,
-    )
-    text_response = response.choices[0].text.strip()
-
-    return prediction, text_response
-
-# Create examples list by specifying the paths to the example images
-examples = ["path/to/example1.jpg", "path/to/example2.jpg"]  # replace with actual image paths
-
-# Define the Gradio interface
-iface = gr.Interface(
-    fn=predict,
-    inputs=gr.Image(shape=(512, 512)),
-    outputs=[gr.Label(num_top_classes=3), gr.Textbox(label="GPT-3 Response")],
-    examples=examples,
-    enable_queue=True  # This is optional and only necessary if you're hosting under heavy traffic
-)
-
-# Launch the Gradio app
-iface.launch()
+  img = PILImage.create(img)
+  pred,pred_idx,probs = learn.predict(img)
+  prediction = {labels[i]: float(probs[i]) for i in range(len(labels))}
+  chat_prompt = f"The model predicted {prediction}."
+  chat_response = generate_text(chat_prompt)
+  return {**prediction, 'chat_response': chat_response}
+
+# Define the chat function
+def chat(query, chat_history):
+  chat_response = handle_query(query, chat_history)
+  return chat_response
+
+# Define the examples
+examples = ['image.jpg']
+
+# Define the interpretation
+interpretation='default'
+
+# Define the enable_queue
+enable_queue=True
+
+# Launch the interface
+gr.Interface(fn=predict,inputs=gr.inputs.Image(shape=(512, 512)),outputs=gr.outputs.Label(num_top_classes=3),examples=examples,interpretation=interpretation,enable_queue=enable_queue).launch()