Upload 15 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+vector_db/chroma_v01/chroma.sqlite3 filter=lfs diff=lfs merge=lfs -text

Config/API_KEYS.yml

@@ -0,0 +1,6 @@
+OPEN_AI:
+    Key: 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
+
+TWILIO:
+    account_sid: AC106e12a4dce8d8920630a3addb6f14e2
+    auth_token: 941ad55307ae272ebd077f7c3fd7abdf

Config/config.yml

@@ -0,0 +1,3 @@
+OPEN_AI:
+    Key: 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
+ 

Dockerfile

@@ -0,0 +1,21 @@
+# pull python base image
+FROM python:3.10
+
+ADD requirements.txt requirements.txt
+
+
+# update pip
+RUN pip install --upgrade pip
+
+# install dependencies
+RUN pip install -r requirements.txt
+
+
+# copy application files
+COPY app/. app/.
+
+# expose port for application
+EXPOSE 8001
+
+# start fastapi application
+CMD ["python", "app/main.py"]

app/helper.py

@@ -0,0 +1,31 @@
+### Vectorstores
+from langchain_community.vectorstores import Chroma
+import openai
+
+def retrieve_relevant_context(vectordb, input_text, num_documents=5):
+    # Use the vector store to retrieve the most relevant documents
+    results = vectordb.similarity_search(query=input_text,k=num_documents) # Without MMR
+    # Extract the page content of the retrieved documents
+    context = "\n".join([result.page_content for result in results])
+    return context
+
+def generate_response_with_context(input_text, context):
+    prompt = (
+        f"You're a GenAI powered assistant to users who're using communication platforms such as Whatsapp, Instagram, Facebook Messenger, etc. to interact with other users. "
+        f"You need to generate responses to the incoming texts on these platforms using the Words, their Meanings, and the Usages contained in the context: {context}\n\n"
+        f"Input text: {input_text}\nResponse:"
+    )
+
+    response = openai.completions.create(
+        model="gpt-3.5-turbo-instruct",
+        prompt=prompt,
+        max_tokens=150,
+        n=1,  # Generate one completion per prompt
+        stop=None,
+        temperature=0.5,
+        frequency_penalty=0.5,
+        presence_penalty=0.5,
+        best_of=1,
+    )
+
+    return response.choices[0].text.strip()

app/main.py

@@ -0,0 +1,82 @@
+from twilio.rest import Client
+import yaml
+import json
+import os
+
+import yaml
+import json
+from langchain.embeddings import OpenAIEmbeddings
+from langchain_community.vectorstores import Chroma
+from helper import retrieve_relevant_context, generate_response_with_context
+
+# Load relevant API Keys
+file_path = './Config/API_KEYS.yml'
+
+with open(file_path, 'r') as file:
+    api_keys = yaml.safe_load(file)
+
+
+# Extract openai username and key
+openai_key = api_keys['OPEN_AI']['Key']
+
+os.environ["OPENAI_API_KEY"] = openai_key
+
+
+
+# Extract openai username and key
+account_sid = api_keys['TWILIO']['account_sid']
+auth_token = api_keys['TWILIO']['auth_token']
+
+account_sid = account_sid
+auth_token = auth_token
+
+# Define the persist directory
+persist_directory = './vector_db/chroma_v01'
+
+# Initialize the embeddings model
+embedding_model = OpenAIEmbeddings()
+
+### Vectorstores
+from langchain_community.vectorstores import Chroma
+
+# Load the Chroma vector store
+vectordb = Chroma(persist_directory=persist_directory, embedding_function=embedding_model)
+
+
+#setup Twilio client
+client = Client(account_sid, auth_token)
+
+
+from flask import Flask, request, redirect
+from twilio.twiml.messaging_response import MessagingResponse
+print("flask app is running")
+app = Flask(__name__)
+
+@app.route("/whatsapp", methods=['GET', 'POST'])
+def incoming_sms():
+    """Send a dynamic reply to an incoming text message"""
+    # Get the message the user sent our Twilio number
+    body = request.values.get('Body', None)
+    print("body :",body)
+    
+    ##### Process incoming text #############
+    incoming_msg = body.strip()
+    if not incoming_msg:
+        return str(MessagingResponse())
+
+    # Generate response using the RAG-powered system
+    retrieved_texts = retrieve_relevant_context(vectordb, incoming_msg)
+    context = "\n".join(retrieved_texts)
+    response = generate_response_with_context(incoming_msg, context)
+    print("response :",response)
+    ##### Process incoming text Done #############
+    
+    
+    # Start our TwiML response
+    resp = MessagingResponse()
+    print("TwiML resp :", resp)
+    resp.message(response)
+    return str(resp)
+
+if __name__ == "__main__":
+    app.run(port=5000, debug=True)

requirements.txt

@@ -0,0 +1,9 @@
+chromadb ==0.5.0
+Flask==3.0.3
+gradio==4.31.5
+huggingface-hub==0.23.0
+openai==1.30.1
+langchain==0.2.0
+langchain-community==0.2.0
+fitz
+twilio

vector_db/chroma_v01/86d715ec-534d-41ba-8058-74b3ecf1b19e/data_level0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f18abd8c514282db82706e52b0a33ed659cd534e925a6f149deb7af9ce34bd8e
+size 6284000

vector_db/chroma_v01/86d715ec-534d-41ba-8058-74b3ecf1b19e/header.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:effaa959ce2b30070fdafc2fe82096fc46e4ee7561b75920dd3ce43d09679b21
+size 100

vector_db/chroma_v01/86d715ec-534d-41ba-8058-74b3ecf1b19e/length.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85f45bb73617241cd10561c82fb101bbc3e3777f2b8f53f1cb2332f803a0366e
+size 4000

vector_db/chroma_v01/86d715ec-534d-41ba-8058-74b3ecf1b19e/link_lists.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
+size 0

vector_db/chroma_v01/chroma.sqlite3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc5cc82d9a38112824d5469c1619e39feb34a29704c9a5eefb13882f11f401e1
+size 8568832