Upload app.py

app.py

@@ -0,0 +1,30 @@
+import numpy as np # linear algebra
+import os
+from sklearn.metrics.pairwise import cosine_similarity
+from sentence_transformers import SentenceTransformer
+import pickle
+import gradio as gr
+
+with open('embeddings.pkl', "rb") as fIn:
+    stored_data = pickle.load(fIn)
+    embeddings = stored_data['embeddings']
+
+with open('answers.pkl', "rb") as fIn:
+    stored_data = pickle.load(fIn)
+    answers = stored_data['answers']
+
+model = SentenceTransformer('all-MiniLM-L6-v2')
+
+def predict(ques):
+    embeded_ques = model.encode(ques)
+    em_vec = np.vstack([embeded_ques]*10)
+    scores = cosine_similarity(embeddings, em_vec)
+    idx = np.argmax([np.mean(arr) for arr in scores])
+    return answers[idx]
+
+
+inputs = gr.Textbox(label='query')
+outputs = gr.Textbox(label='query')
+title = "Similar faq"
+description = "Retrive answers of similar queries sing sentence transformers"
+gr.Interface(fn = predict, inputs = inputs, outputs = outputs, title = title, description = description).launch()