def work():
train_trans = transforms.Compose(
[transforms.RandomResizedCrop(224),
transforms.ToTensor()])
# https://gluon-cv.mxnet.io/build/examples_datasets/imagenet.html
# You need to specify ``root`` for ImageNet if you extracted the images into
# a different folder, not use rec but general images, such as *.jpg
train_data = DataLoader(ImageNet(
train=True, root="/home1/ImageNet_ILSVRC2012/ILSVRC2012_img_train/").
transform_first(train_trans),
batch_size=128,
shuffle=True)
for x, y in train_data:
print(x.shape, y.shape)
break
from gluoncv.utils import viz
val_dataset = ImageNet(train=False)
viz.plot_image(val_dataset[1234][0]) # index 0 is image, 1 is label
viz.plot_image(val_dataset[4567][0])
from tf2cv.model_provider import get_model as tf2cv_get_model
import tensorflow as tf
from gluoncv.data import ImageNet
import numpy as np
from NN.img2tensor import convert_image
net = tf2cv_get_model("AlexNet", pretrained=True, data_format="channels_last")
x = convert_image("img_23.jpg")
x = ((x - np.array((0.485, 0.456, 0.406))) / np.array(
(0.229, 0.224, 0.225))).reshape(1, 224, 224, 3)
y = net(x)
probs = tf.nn.softmax(y)
top_k = 5
probs_np = probs.numpy().squeeze(axis=0)
top_k_inds = probs_np.argsort()[::-1][:top_k]
classes = ImageNet().classes
print("The input picture is classified to be:")
for k in range(top_k):
print("{idx}: [{class_name}], with probability {prob:.3f}.".format(
idx=(k + 1),
class_name=classes[top_k_inds[k]],
prob=probs_np[top_k_inds[k]]))
"""
from gluoncv.data import ImageNet
from mxnet.gluon.data import DataLoader
from mxnet.gluon.data.vision import transforms
train_trans = transforms.Compose([
transforms.RandomResizedCrop(224),
transforms.ToTensor()
])
# You need to specify ``root`` for ImageNet if you extracted the images into
# a different folder
train_data = DataLoader(
ImageNet(train=True).transform_first(train_trans),
batch_size=128, shuffle=True)
#########################################################################
for x, y in train_data:
print(x.shape, y.shape)
break
#########################################################################
# Plot some validation images
from gluoncv.utils import viz
val_dataset = ImageNet(train=False)
viz.plot_image(val_dataset[1234][0]) # index 0 is image, 1 is label
viz.plot_image(val_dataset[4567][0])
from gluoncv.data import ImageNet
from mxnet.gluon.data import DataLoader
from mxnet.gluon.data.vision import transforms
test_trans = transforms.Compose(
[transforms.RandomResizedCrop(224),
transforms.ToTensor()])
test_data = DataLoader(ImageNet(root='.',
train=False).transform_first(test_trans),
batch_size=128,
shuffle=True)
def get_data_loader(data_dir):
train_dataset = ImageNet(data_dir, train=True)
val_dataset = ImageNet(data_dir, train=False)
return train_dataset, val_dataset