Update app.py

app.py

@@ -1,18 +1,26 @@
 import os
 import streamlit as st
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from groq import Groq
 
-# Load the IBM Granite model and tokenizer
-model_path = "ibm-granite/granite-8b-code-base"  # Replace with the desired model path
-tokenizer = AutoTokenizer.from_pretrained(model_path)
-model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto")
+# Set the Groq API key
+os.environ["GROQ_API_KEY"] = "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"
 
-# Function to generate text using the IBM Granite model
-def generate_text(prompt):
-    inputs = tokenizer(prompt, return_tensors="pt", truncation=True, padding=True, max_length=512)
-    outputs = model.generate(**inputs, max_new_tokens=150)
-    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return generated_text
+# Initialize Groq client
+client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
+
+# Define the LLaMA model to be used
+MODEL_NAME = "llama3-8b-8192"
+
+# Function to call Groq API
+def call_groq_api(prompt):
+    try:
+        chat_completion = client.chat.completions.create(
+            messages=[{"role": "user", "content": prompt}],
+            model=MODEL_NAME
+        )
+        return chat_completion.choices[0].message.content
+    except Exception as e:
+        return f"Error: {str(e)}"
 
 # Define functions for each tool
 def personalized_learning_assistant(topic):
@@ -22,7 +30,7 @@ def personalized_learning_assistant(topic):
         "Explain machine learning. Example: Machine learning involves algorithms that improve through experience."
     ]
     prompt = f"Here are some examples of explanations:\n\n{examples}\n\nNow, explain the topic: {topic}"
-    return generate_text(prompt)
+    return call_groq_api(prompt)
 
 def ai_coding_mentor(code_snippet):
     examples = [
@@ -30,7 +38,7 @@ def ai_coding_mentor(code_snippet):
         "Review this code:\n\nCode: 'def add(a, b): return a + b'\nSuggestion: Add type hints for better readability."
     ]
     prompt = f"Here are some examples of code reviews:\n\n{examples}\n\nReview the following code snippet:\n{code_snippet}"
-    return generate_text(prompt)
+    return call_groq_api(prompt)
 
 def smart_document_summarizer(document_text):
     examples = [
@@ -38,7 +46,7 @@ def smart_document_summarizer(document_text):
         "Summarize this passage:\n\nText: 'The global climate change crisis necessitates urgent action to reduce carbon emissions.'\nSummary: 'Immediate action is needed to tackle climate change.'"
     ]
     prompt = f"Here are some examples of summaries:\n\n{examples}\n\nSummarize this document:\n{document_text}"
-    return generate_text(prompt)
+    return call_groq_api(prompt)
 
 def interactive_study_planner(exam_schedule):
     examples = [
@@ -46,7 +54,7 @@ def interactive_study_planner(exam_schedule):
         "Generate a study plan for:\n\nSchedule: 'Exams in 2 weeks'\nPlan: 'Focus on subjects with more weight and review daily.'"
     ]
     prompt = f"Here are some examples of study plans:\n\n{examples}\n\nCreate a study plan for the following schedule:\n{exam_schedule}"
-    return generate_text(prompt)
+    return call_groq_api(prompt)
 
 def real_time_qa_support(question):
     examples = [
@@ -54,7 +62,7 @@ def real_time_qa_support(question):
         "Provide an explanation for:\n\nQuestion: 'What is photosynthesis?'\nAnswer: 'Photosynthesis is the process by which plants convert light energy into chemical energy.'"
     ]
     prompt = f"Here are some examples of Q&A responses:\n\n{examples}\n\nAnswer the following question:\n{question}"
-    return generate_text(prompt)
+    return call_groq_api(prompt)
 
 def mental_health_check_in(feelings):
     examples = [
@@ -62,7 +70,7 @@ def mental_health_check_in(feelings):
         "Offer support for:\n\nFeeling: 'Feeling overwhelmed'\nAdvice: 'Consider relaxation techniques and seek support from friends or a counselor.'"
     ]
     prompt = f"Here are some examples of mental health advice:\n\n{examples}\n\nProvide advice for:\n{feelings}"
-    return generate_text(prompt)
+    return call_groq_api(prompt)
 
 # Define Streamlit app
 st.set_page_config(page_title="EduNexus", page_icon=":book:", layout="wide")