Update app.py

app.py

@@ -4,6 +4,11 @@ import pandas as pd
 
 from functions import *
 
+from qdrant_client import QdrantClient
+from qdrant_client.http.models import VectorParams, Distance, Record, Filter
+from random import uniform
+
+
 
 
 backgroundPattern = """
@@ -107,6 +112,33 @@ resume_combined.reset_index(drop=True, inplace=True)
 
 #QDRANT VECTORIZER
 
+vector_dimension = encoder.model.get_sentence_embedding_dimension()
+qdrant_interface = QdrantInterface(QUADRANT_ENDPOINT, QUADRANT_API_KEY, vector_dimension)
+qdrant_interface.create_collection('jobs', Distance.COSINE)
+qdrant_interface.create_collection('resumes', Distance.COSINE)
+
+# Function to ensure vectors are in list format
+def ensure_list_format(df, vector_col):
+    df[vector_col] = df[vector_col].apply(lambda x: x.tolist() if hasattr(x, 'tolist') else x)
+    return df
+
+# Ensure vectors are in the correct format before uploading
+jobs_combined = ensure_list_format(jobs_combined, 'summarized_description_encoded')
+resume_combined = ensure_list_format(resume_combined, 'summarized_resume_encoded')
+
+given_job_vector = jobs_combined['summarized_description_encoded'].iloc[0]
+
+# Now upload to Qdrant
+qdrant_interface.save_to_qdrant(jobs_combined, 'jobs', 'summarized_description_encoded', ['processed_title', 'processed_description', 'token_count', 'summarized_description'])
+qdrant_interface.save_to_qdrant(resume_combined, 'resumes', 'summarized_resume_encoded', ['processed_resume', 'token_count', 'summarized_resume'])
+
+# Retrieve specific records by IDs from the 'jobs' collection
+specific_jobs_records = qdrant_interface.retrieve_specific_records('jobs', ids=[1])
+
+# Find top 5 matching resumes for the example job
+matched_resumes = qdrant_interface.match_jobs_to_resumes(given_job_vector, top_k=5)
+for resume, score in matched_resumes:
+    print(f"Matched Resume: {resume['summarized_resume']}, Score: {score}")