Part $1$: GPU Programming with Cuda for Lenet CNN $(70+10)$

Initial Model Deployment: Obtain a trained Lenet CNN model and a testing dataset from an external source such as a computer, server, or Google Colab. Please deploy the model onto the Jetson Nano without utilizing any GPU resources. Please measure and record the average time to obtain results from the Jetson Nano as the baseline.

GPU Parallelism: Next, we will transition from sequential execution to parallel processing by defining GPU kernels and applying parallelism to the code. The degree of parallelism applied will impact the project score for part $1.$

Performance Evaluation: Please measure the average parallel execution time and compare it with the initial baseline. Then, please observe the results and discuss the results with the team. If results do not meet expectations, please identify potential problems and come up with improvement methods to optimize performance further.