Solving the inverse scattering problem is of utmost importance in various fields such as medical imaging, remote sensing, radar systems, and many more. However, obtaining complete information about the scattered wavefield is often challenging, leading to partial data measurements known as “phase-less” data. In this paper, we propose a novel method for solving the inverse scattering problem using deep learning-based DConvNet. Our approach leverages the phase-less data and learns a mapping between the unknown object and scattered wave field using the convolutional neural network architecture. Since the DConvNet can be trained on a large dataset of scattering field data, it can generalise well to new objects and scenarios. We demonstrate the effectiveness of our proposed method which also works for Lossy scatterers. Our method produces superior reconstruction results when compared with literature.