Ada>> digit-recognizer-using-MNIST-dataset>> 返回
项目作者： riyag283

项目描述 ：
Sequential Convolutional Neural Network for handwritten digits recognition trained on MNIST dataset using keras API
高级语言： Jupyter Notebook
项目地址: git://github.com/riyag283/digit-recognizer-using-MNIST-dataset.git


digit-recognizer-using-MNIST-dataset

Sequential Convolutional Neural Network for handwritten digits recognition trained on MNIST dataset using keras API.

The input data was divided into 90% training set and 10% validation set and the neural network was optimized using Adam Optimizer.
tf.keras.losses.
SparseCategoricalCrossentropy was used to compute the crossentropy loss between the labels and predictions.
The model achieved 97.43% accurary.