mx.io.MNISTIter
¶
Description¶
Iterating on the MNIST dataset.
One can download the dataset from http://yann.lecun.com/exdb/mnist/
Usage¶
mx.io.MNISTIter(...)
Arguments¶
Argument |
Description |
---|---|
|
string, optional, default=’./train-images-idx3-ubyte’. Dataset Param: Mnist image path. |
|
string, optional, default=’./train-labels-idx1-ubyte’. Dataset Param: Mnist label path. |
|
int, optional, default=’128’. Batch Param: Batch Size. |
|
boolean, optional, default=1. Augmentation Param: Whether to shuffle data. |
|
boolean, optional, default=0. Augmentation Param: Whether to flat the data into 1D. |
|
int, optional, default=’0’. Augmentation Param: Random Seed. |
|
boolean, optional, default=0. Auxiliary Param: Whether to print out data info. |
|
int, optional, default=’1’. partition the data into multiple parts |
|
int, optional, default=’0’. the index of the part will read |
|
long (non-negative), optional, default=4. Maximum number of batches to prefetch. |
|
{‘cpu’, ‘gpu’},optional, default=’gpu’. Context data loader optimized for. |
|
{None, ‘float16’, ‘float32’, ‘float64’, ‘int32’, ‘int64’, ‘int8’, ‘uint8’},optional, default=’None’. Output data type. |
Value¶
iter
The result mx.dataiter
Link to Source Code: http://github.com/apache/incubator-mxnet/blob/1.6.0/src/io/iter_mnist.cc#L265