created app.py

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+import pandas as pd
+import numpy as np
+import pickle
+#from sentence_transformers import SentenceTransformer
+from sklearn.metrics.pairwise import cosine_similarity
+
+# Load the dataset
+data = pd.read_csv('data/Recommender_data.csv')
+
+# Load the precomputed embeddings
+with open('data/Model/tool_embeddings.pkl', 'rb') as f:
+    tool_embeddings = pickle.load(f)
+
+# # Load the model from the file using pickle
+with open('data/Model/model.pkl', 'rb') as f:
+    model = pickle.load(f)
+
+
+# def get_sentiment(input_text):
+#   return sentiment(input_text)
+# Define the recommendation function
+def recommend(product_idea, model=model):#, model=model
+    # Load the pre-trained model
+    #model = SentenceTransformer('paraphrase-distilroberta-base-v1')
+
+    # Compute the embedding for the product idea
+    product_embedding = model.encode([product_idea])
+
+    # Compute the cosine similarity between the product idea vector and all tool vectors
+    cosine_sim_product = cosine_similarity(product_embedding, tool_embeddings)
+
+    # Sort the results in descending order and get the indices of the most similar tools
+    indices = np.argsort(cosine_sim_product)[0][::-1]
+
+    # Return the top 5 most similar tools
+    recommendations = {data.iloc[indices[i]]['Tools']: {"Description":data.iloc[indices[i]]['Description'],"URL":data.iloc[indices[i]]['URL'],"Logo":data.iloc[indices[i]]['Logo']} for i in range(5)}
+
+    return recommendations
+
+iface = gr.Interface(fn = recommend, 
+                     inputs = "text", 
+                     outputs = ['text'],
+                     title = 'OHA', 
+                     description="Get OHA Recomandation for the given input")
+                     
+iface.launch(inline = False)