def demand_forecasting():
request_data = request.args
ip_address = request_data.get('ipAddress')
df = DemandForecast()
test_x = df.convert_data_by_ip_address(ip_address)
if len(test_x) > 0:
test_x = np.unique(test_x, axis=0)
model_saved = keras.models.load_model('forecast_demand_model.h5')
pred = model_saved.predict(test_x.reshape(len(test_x), len(test_x[0])),
batch_size=1)
return jsonify({"forecastResults": '{}'.format(pred)})
else:
return 'no data'
def generate_chat_model():
chat_model = ChatBotModel()
data_dump_model = DataBaseDump()
data_dump_model.create_data_dump()
data_dump_model.create_data_dump_ip_address()
res = chat_model.generatechatmodel()
demand_forecast = DemandForecast()
demand_forecast.forecast_demand_model()
demand_forecast.forecast_item_category_demand_model()
demand_forecast.forecast_item_category_demand_model_without_user_id()
demand_forecast.forecast_item_price_demand_model()
demand_forecast.forecast_item_discount_demand_model()
demand_forecast.forecast_order_quantity_demand_model()
demand_forecast.forecast_order_total_amount_demand_model()
demand_forecast.forecast_order_status_demand_model()
return res
def item_category_demand_forecasting_by_userid_chat():
request_data = request.args
user_id = request_data.get('userId')
df = DemandForecast()
test_x = df.convert_data_by_user_id_chat(user_id)
if len(test_x) > 0:
test_x = np.unique(test_x, axis=0)
model_saved = keras.models.load_model(
'forecast_item_category_demand_model.h5')
pred = model_saved.predict(test_x.reshape(len(test_x), len(test_x[0])),
batch_size=1)
print(pred)
return jsonify({"forecastResults": '{}'.format(pred)})
else:
return 'no data'
def get_recommend_cart_items():
query_parameters = request.args
user_id = query_parameters.get('userId')
df = DemandForecast()
test_x = df.convert_data_by_user_id(user_id)
if len(test_x) > 0:
test_x = np.unique(test_x, axis=0)
model_saved = keras.models.load_model('forecast_model.h5')
prediction = model_saved.predict(test_x.reshape(
len(test_x), len(test_x[0])),
batch_size=1)
prediction = json.dumps(prediction.tolist())
print(prediction)
return jsonify({"forecastResults": '{}'.format(prediction)})
else:
return 'no data'
from DemandForecasting import DemandForecast df = DemandForecast() df.forecast_item_category_demand_model()