**5. Conclusions**

The technique of Levenberg-Marquardt with backpropagated neural network is used to study the three-dimensional Prandtl nanofluid flow model with a convectively heated surface by the variations of Prandtl fluid number, flexible number, ratio parameter, Prandtl number, Biot number and thermophoresis number. The governing PDEs are transformed to system of ordinary differential equations by applying suitable transformation. The Mathematica command 'ND Solve' is utilized to compute the reference dataset for 101 points, in which there are 81 points for training, 10 point for testing and 10 points for validation. The regression analysis, error histogram and MSE data validates the performance of TLM-BANN.

There are few points drawn from the results are written below:

