verahession 0.1.28

Python client library for Vera AI chatbot by Hession Dynamics

verahession — Quick Start Guide

This guide explains how to install and use the verahession library to train an intent classifier and connect it to an LLM API.

---

1. Installation

Install the library using pip:

python3 -m pip install verahession

---

2. Train the Intent Classifier

Create a file named train.py with the following content:

from verahession.assistant import train_model

if __name__ == "__main__":
    INTENTS_PATH = "intents.json"
    MODEL_SAVE_PATH = "model.pth"
    train_model(INTENTS_PATH, MODEL_SAVE_PATH)

Make sure you have an intents.json file containing your training data.

Run the script:

python3 train.py

This will generate a model file named model.pth.

---

3. Use the Classifier for Chat

Create a file named chat.py with the following content:

from verahession.assistant import Classifier
from verahession.assistant import export_to_numpy

export_to_numpy("model.pth", "model.pkl")

clf = Classifier("model.pkl")

while True:
    sentence = input("You: ")
    if sentence == "quit":
        break
    tag, confidence = clf.classify(sentence)
    print(f"Intent: {tag} (Confidence: {confidence:.2f})")

Run the script:

python3 chat.py

Enter sentences to classify intents. Type quit to exit.

---

4. Connect to an LLM API

To send input to the LLM API, use the following code:

from verahession.api import *

vera_llm = vera_interface(API_KEY, AGENTNAME)
response = vera_llm.send(sentence)

Replace API_KEY and AGENTNAME with the appropriate values.

---

If you do not yet have an API key, you need to create an account at https://hessiondynamics.com/login - and then press 'create API key' - do NOT share the key with nanyone.

This completes the basic usage instructions for the verahession library. For additional help, consult the documentation or support channels.