This repository contains Tensorflow implementation of Handwritten Digit Recognition on MNIST dataset with test accuracy of 0.9923.
The output generated on running this code:
Extracting MNIST_data/train-images-idx3-ubyte.gz Extracting MNIST_data/train-labels-idx1-ubyte.gz Extracting MNIST_data/t10k-images-idx3-ubyte.gz Extracting MNIST_data/t10k-labels-idx1-ubyte.gz 2018-12-04 00:56:30.373794: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1030] Found device 0 with properties: name: GeForce GTX 1080 major: 6 minor: 1 memoryClockRate(GHz): 1.7335 pciBusID: 0000:03:00.0 totalMemory: 7.92GiB freeMemory: 7.81GiB 2018-12-04 00:56:30.374497: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1080, pci bus id: 0000:03:00.0, compute capability: 6.1) 0.9218 2018-12-04 00:56:54.470549: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1080, pci bus id: 0000:03:00.0, compute capability: 6.1) test accuracy 0.9923
Please feel free to use this code for learning purposes and reach out to me in case of doubts/issues.