# About task
GPR is one of the geophysical research methods. The essence of the problem comes down to the emission of electromagnetic waves into the near-surface space and obtaining reflections of these waves from electrophysical inhomogeneities.
There are different approaches to interpreting the data obtained, but they all have their own shortcomings and inaccuracies. In this competition, it is proposed to test various ML methods for solving a problem to find new solutions for processing georadar data.

# About data
The data was generated using the gprMax package. This is synthetic data that has been processed and scaled. The exact values ​​of the simulation here are most likely not interesting, if this is not so, then write.
The input signal is a square 128x128x1 - a radio signal, the output signal 128x128x2 - these are the parameters of the environment, dielectric constant and electrical conductivity.

# What do we see in the sample?
- The output sample is a meter by meter square with a randomly simulated environment with values ​​similar to real ones, which includes metal cylinders distributed randomly. The figure also has a thin layer of air on top.
- The input signal (x) shows the result of a georadar survey along a profile one meter long (horizontal) and 30 ns deep (vertical).