The invention discloses an energy efficient fog computing migration method based on deep learning. Firstly, a fog computing migration optimization problem of task completion time minimization is constructed, a fog computing migration decision algorithm based on deep learning is provided for solving the optimization problem, the algorithm has relatively fast convergence performance, and the task completion time in a complex network scene can be reduced to the greatest extent; secondly, in order to further optimize the energy consumption of fog computing migration, a terminal user energy consumption minimization fog computing migration optimization problem is constructed, an optimal transmission power distribution solving algorithm is provided for solving the optimization problem on the basis of an optimal migration decision solved by the migration decision algorithm, and the solving algorithm dynamically distributes transmission power, so that the optimal transmission power, namely theminimum energy consumption, is obtained; finally, the specific implementation of the method verifies the advantages of the fog computing migration method in reducing task completion time and user energy consumption.