Spaces:
Sleeping
Sleeping
File size: 2,823 Bytes
6f053b0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 |
import os
import sys
import pandas as pd
from sklearn.pipeline import Pipeline
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.ensemble import RandomForestClassifier
# Ensure infrastructure is importable if it's a sibling package in src
# Assuming src is in PYTHONPATH or we add it relative to this file
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
from infrastructure.rules import crear_dataset_rules
from infrastructure.client_requests import (
consult_insurance_policy,
report_emergency,
consult_payments,
schedule_inspection,
manage_claims,
quote_new_insurance,
consult_bank_channel
)
class InsuranceChatbot:
def __init__(self):
self.pipeline = None
self.intents = {
"consultar_poliza": consult_insurance_policy,
"reportar_emergencia": report_emergency,
"pagos": consult_payments,
# "inspeccion": schedule_inspection, # Not in original rules dictionary but in imports
# "reclamos": manage_claims, # Not in original rules dictionary but in imports
"cotizar": quote_new_insurance,
# "banco": consult_bank_channel # Not in original rules dictionary but in imports
}
# Mapping intent labels to functions
# The rules.py dataset generator produces specific labels, need to match them.
# Labels from rules.py: consultar_poliza, reportar_emergencia, pagos, cotizar
# We can extend the dataset or mapping if needed. For now, matching the notebook logic.
def train_model(self):
"""Generates dataset and trains the classification pipeline."""
print("Generating dataset...")
df = crear_dataset_rules(n_pos_clas=1000)
print(f"Training model on {len(df)} samples...")
self.pipeline = Pipeline([
('tfidf', TfidfVectorizer()),
('clf', RandomForestClassifier(random_state=42))
])
X = df['text']
y = df['label']
self.pipeline.fit(X, y)
print("Model trained successfully.")
def predict_intent(self, text):
"""Predicts the intent of the given text."""
if not self.pipeline:
raise ValueError("Model not trained. Call train_model() first.")
prediction = self.pipeline.predict([text])[0]
return prediction
def handle_message(self, text):
"""Processes a message and executes the corresponding action."""
intent = self.predict_intent(text)
print(f"Detected intent: {intent}")
action = self.intents.get(intent)
if action:
return action()
else:
return "Lo siento, no entendí tu solicitud o no tengo una acción para ese intento."
|