Most state-of-the-art approaches for weather and climate modelling are based on physics informed numerical models of the atmosphere. These approaches aim to model the non-linear dynamics and complex interactions between multiple variables, which are challenging to approximate. Additionally, many such numerical models are computationally intensive, especially when modelling the atmospheric phenomenon at a fine-grained spatial and temporal resolution. Recent data-driven approaches based on machine learning instead aim to directly solve a downstream forecasting or projection task by learning a data-driven functional mapping using deep neural networks. However, these networks are trained using curated and homogeneous climate datasets for specific spatiotemporal tasks, and thus lack the generality of numerical models. In this research, the data available on Google for research objectives was used in two different algorithms through neural networks (ANN) and by controlling the design of these networks and training them hard, with the aim of obtaining results with a high degree of accuracy for weather forecasting for a period of up to 365 days in the first proposed model of neural networks and in case of the second algorithm the weather forecasted period is of up to four years (1460 days). All the facilities available in the fifty most used and downloaded weather forecasting applications from Google have been added to the proposed applications so that they are also available for usability through the two proposed models. The degree of accuracy obtained is high for the proposed two algorithms.