Implemented a bit of the RAG (will add more tomorrow)

app.py

@@ -3,6 +3,36 @@ from huggingface_hub import InferenceClient
 
 client = InferenceClient("Qwen/Qwen2.5-7B-Instruct")
 
+with open("knowledge.txt", "r", encoding="utf=8") as file:
+    knowledge_text = file.read()
+print(knowledge_text)
+
+def preprocess_text(text):
+    cleaned_text = text.strip()
+
+    chunks = cleaned_text.split("\n")
+
+    cleaned_chunks = []
+
+    for chunk in chunks:
+        stripped_chunk = chunk.strip()
+        if len(stripped_chunk) > 0:
+            cleaned_chunks.append(stripped_chunk)
+    print(cleaned_chunks)
+    print(len(cleaned_chunks))
+    return cleaned_chunks
+cleaned_chunks = preprocess_text(knowledge_text)
+
+
+model = SentenceTransformer('all-MiniLM-L6-v2')
+
+def create_embeddings(text_chunks):
+    chunk_embeddings = model.encode(text_chunks, convert_to_tensor=True)
+    print(chunk_embeddings)
+    print(chunk_embeddings.shape)
+    return chunk_embeddings
+chunk_embeddings = create_embeddings(cleaned_chunks)
+
 def respond(message, history):
     messages = [
         {"role": "system", "content": "You are a friendly, supportive, and mental wellness chatbot"}