PyTorch_Forecasting_with_InfluxDB

Introduction

InfluxDB is a database that specializes in the collection, storage, and processing of time series data. In this Python project, you will see:

1. How to establish connection between InfluxDB and InfluxDB client,
2. How to push data to InfluxDB using a dataframe,
3. How to fetch dataframe from InfluxDB to Python client, and
4. How to use this data to perform forecasting using PyTorch.

First, we install influxdb and influxdb-client that will help establish connection between InfluxDB Cloud and Python client. We use PyTorch forecasting to make predctions on this data. For this, we install torch and pytorch-forecasting. We use the Stallion time series data present in PyTorch. We predict the beverage sales (in dollars) for 6 months based on 21,000 rows of historical data.

Stallion data contains the below datasets:

• pricesalespromotion.csv: Holds the price, sales & promotion in dollars.
• historicalvolume.csv: contains sales data.
• weather.csv: the average maximum temperature at Agency monthly
• Industrysodasales.csv: Holds industry level soda sales
• eventcalendar.csv: Holds event details (sports, carnivals, etc.)
• industry_volume.csv: industry actual beer volume
• demographics.csv: demographic details