Update app.py

app.py

@@ -1,21 +1,28 @@
 import streamlit as st
+import json
+import os
+import pandas as pd
+import requests
 from transformers import AutoTokenizer, pipeline
 from peft import PeftModel, PeftConfig
 from transformers import AutoModelForSeq2SeqLM
-import pandas as pd
-import kaggle
-import requests
-import json
-import os
-
-# Set up Kaggle API credentials
-KAGGLE_USERNAME = "tahirsherbhatti"  # Replace with your Kaggle username
-KAGGLE_KEY = "9b50dd1e468214e9a6e514fb32afd6d5"          # Replace with your Kaggle API key
-
-# Write Kaggle credentials to file for authentication
-os.environ['KAGGLE_CONFIG_DIR'] = './'
-with open('kaggle.json', 'w') as f:
-    json.dump({'username': KAGGLE_USERNAME, 'key': KAGGLE_KEY}, f)
+from kaggle.api.kaggle_api_extended import KaggleApi
+
+# Function to load Kaggle API credentials
+def load_kaggle_credentials():
+    # Load kaggle.json file
+    with open('kaggle.json') as f:
+        kaggle_credentials = json.load(f)
+    return kaggle_credentials['username'], kaggle_credentials['key']
+
+# Set up the Kaggle API client
+def setup_kaggle_api():
+    username, key = load_kaggle_credentials()
+    os.environ['KAGGLE_USERNAME'] = username
+    os.environ['KAGGLE_KEY'] = key
+    api = KaggleApi()
+    api.authenticate()
+    return api
 
 # Load the LoRA configuration and model
 try:
@@ -30,38 +37,11 @@ except Exception as e:
 # Set up the Q&A pipeline using the LoRA fine-tuned model
 qa_pipeline = pipeline("text2text-generation", model=model, tokenizer=tokenizer)
 
-# Function to load Kaggle datasets directly into pandas DataFrames
-def load_kaggle_dataset(dataset_path, file_name):
-    kaggle.api.authenticate()  # Authenticate using your Kaggle credentials
-    kaggle.api.dataset_download_files(dataset_path, path=".", unzip=True)
-    return pd.read_csv(file_name)
-
-# Load relevant Kaggle datasets
-try:
-    # CareerBuilder Job Listings Dataset
-    job_listings_df = load_kaggle_dataset("promptcloud/careerbuilder-job-listing-2020", "job_listings.csv")
-    st.write("CareerBuilder Job Listings Dataset loaded successfully!")  # Debugging message
-    
-    # Career Prediction Dataset
-    career_df = load_kaggle_dataset("utkarshshrivastav07/career-prediction-dataset", "career_prediction.csv")
-    st.write("Career Prediction Dataset loaded successfully!")  # Debugging message
-except Exception as e:
-    st.error(f"Error loading Kaggle dataset: {e}")
-
-# Load SQuAD dataset for intelligent Q&A from GitHub
-try:
-    url = "https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v2.0.json"
-    response = requests.get(url)
-    squad_data = response.json()
-    st.write("SQuAD Dataset loaded successfully!")  # Debugging message
-except Exception as e:
-    st.error(f"Error loading SQuAD dataset: {e}")
-
 # Streamlit App Structure
 def main():
     st.title("AI-Powered Career Counseling App with Advanced Q&A")
     st.sidebar.title("Navigation")
-    option = st.sidebar.selectbox("Choose an Option", ["Profile Setup", "Career Q&A", "Career Recommendations", "Resource Library"])
+    option = st.sidebar.selectbox("Choose an Option", ["Profile Setup", "Career Q&A", "Career Recommendations", "Resource Library", "Kaggle Integration"])
 
     if option == "Profile Setup":
         profile_setup()
@@ -71,6 +51,8 @@ def main():
         career_recommendations()
     elif option == "Resource Library":
         resource_library()
+    elif option == "Kaggle Integration":
+        kaggle_integration()
 
 # Global variable to store user responses
 user_responses = []
@@ -195,5 +177,25 @@ def resource_library():
     for keyword in keywords:
         st.write(f"- {keyword}")
 
+# Kaggle Integration Section
+def kaggle_integration():
+    st.header("Kaggle Integration")
+    if st.button("Setup Kaggle API"):
+        try:
+            api = setup_kaggle_api()
+            st.success("Kaggle API successfully set up!")
+        except Exception as e:
+            st.error(f"Failed to set up Kaggle API: {e}")
+
+    if st.button("List Kaggle Datasets"):
+        try:
+            api = setup_kaggle_api()  # Ensure API is set up
+            datasets = api.dataset_list()
+            st.write("### Available Datasets:")
+            for dataset in datasets:
+                st.write(dataset.ref)
+        except Exception as e:
+            st.error(f"Failed to retrieve datasets: {e}")
+
 if __name__ == "__main__":
     main()