def get_model_alexnet(classes_num, ctx):
pretrained_net = models.alexnet(pretrained=True)
# print(pretrained_net)
finetune_net = models.alexnet(classes=classes_num) # 输出为classes_num个类
finetune_net.features = pretrained_net.features # 特征设置为inceptionv3的特征
finetune_net.output.initialize(init.Xavier(), ctx=ctx) # 对输出层做初始化
finetune_net.collect_params().reset_ctx(ctx) # 设置CPU或GPU
finetune_net.hybridize() # gluon特征，运算转成符号运算，提高运行速度
return finetune_net
return train_transform, val_transform
def get_imagenet_iterator(root, batch_size, num_workers, data_shape=224, dtype='float32'):
"""Dataset loader with preprocessing."""
train_dir = os.path.join(root, 'train')
train_transform, val_transform = get_imagenet_transforms(data_shape, dtype)
logging.info("Loading image folder %s, this may take a bit long...", train_dir)
train_dataset = ImageFolderDataset(train_dir, transform=train_transform)
train_data = DataLoader(train_dataset, batch_size, shuffle=True,
last_batch='discard', num_workers=num_workers)
val_dir = os.path.join(root, 'val')
if not os.path.isdir(os.path.expanduser(os.path.join(root, 'val', 'n01440764'))):
user_warning = 'Make sure validation images are stored in one subdir per category, a helper script is available at https://git.io/vNQv1'
raise ValueError(user_warning)
logging.info("Loading image folder %s, this may take a bit long...", val_dir)
val_dataset = ImageFolderDataset(val_dir, transform=val_transform)
val_data = DataLoader(val_dataset, batch_size, last_batch='keep', num_workers=num_workers)
return DataLoaderIter(train_data, dtype), DataLoaderIter(val_data, dtype)
resnet18 = vision.resnet18_v1(pretrained=True)
alexnet = vision.alexnet(pretrained=True)
inception = vision.inception_v3(pretrained=True)
#squeezenet = vision.squeezenet1_0()
#densenet = vision.densenet_161()
get_imagenet_iterator("c:\\data\\images", batch_size, num_workers, data_shape=224, dtype='float32'):
import mxnet as mx import time import gluoncv from mxnet import nd, autograd from mxnet import gluon from mxnet.gluon import nn inputShape = (1, 3, 224, 224) from mxnet.gluon.model_zoo import vision alexnet = vision.alexnet() inception = vision.inception_v3() resnet18v1 = vision.resnet18_v1() resnet18v2 = vision.resnet18_v2() squeezenet = vision.squeezenet1_0() densenet = vision.densenet121() mobilenet = vision.mobilenet0_5() ############### 그래프 ############### import gluoncv gluoncv.utils.viz.plot_network(resnet18v1, shape=inputShape) #####################################
transform=transformer),
batch_size=batch_size,
shuffle=False,
last_batch='discard')
for d, l in train_data:
break
print(d.shape, l.shape)
### graph
from gluoncv.utils import viz
viz.plot_image(d[63][2]) # index 0 is image, 1 is label
from mxnet.gluon.model_zoo import vision
net = vision.alexnet(classes=10, pretrained=False, ctx=ctx)
net = gluon.nn.Sequential()
# 은닉층1 (채널=96, 커널=11, 패딩=1, 스트라이드=4, 활성화함수=relu)
# maxpooling(사이즈=3, 스트라이드2)
# 입력사이즈 (224, 224), 출력사이즈 (27, 27)
net.add(
gluon.nn.Conv2D(96,
kernel_size=11,
padding=1,
strides=4,
activation='relu'))
net.add(gluon.nn.MaxPool2D(pool_size=3, strides=2))
# 은닉층2 (채널=256, 커널=5, 패딩=2, 스트라이드=1, 활성화함수=relu)
# maxpooling(사이즈=3, 스트라이드=2)
def alexnetmxnetload():
net = vision.alexnet(pretrained=True)
net.hybridize()
return net
from mxnet.gluon.model_zoo import vision as models
from mxnet import image
from mxnet import nd
def transform(data):
data = data.transpose((2, 0, 1)).expand_dims(axis=0)
rgb_mean = nd.array([0.485, 0.456, 0.406]).reshape((1, 3, 1, 1))
rgb_std = nd.array([0.229, 0.224, 0.225]).reshape((1, 3, 1, 1))
return (data.astype('float32') / 255 - rgb_mean) / rgb_std
with open('./label.txt', 'r') as f:
text_labels = [''.join(l.split("'")[1]) for l in f]
net = models.alexnet(pretrained=True)
image_list = os.listdir('./samples')
for image_name in image_list:
image_path = os.path.join('./samples', image_name)
print(image_path)
x = image.imread(image_path)
x = image.resize_short(x, 256)
x, _ = image.center_crop(x, (224, 224))
prob = net(transform(x)).softmax()
idx = prob.topk(k=1)[0]
for i in idx:
i = int(i.asscalar())
print('With prob = %.5f, it contains %s' %
(prob[0, i].asscalar(), text_labels[i]))
from mxnet import gluon
from mxnet import ndarray as nd
from tqdm import tqdm
from mxnet import autograd
from mxnet.gluon.model_zoo import vision as models
import time
import h5py
import os
# In[2]:
ctx = mx.gpu()
# In[3]:
models.alexnet(pretrained=True, ctx=ctx)
net_functions = [
models.alexnet, models.densenet121, models.densenet161, models.densenet169,
models.densenet201, models.inception_v3, models.resnet18_v2,
models.resnet34_v2, models.resnet50_v1, models.vgg11_bn, models.vgg13_bn,
models.vgg16_bn, models.vgg19_bn
]
# In[4]:
net_dict = {}
for net_function in net_functions:
pretrained_net = net_function(pretrained=True, ctx=ctx)
finetune_net = net_function(classes=120, ctx=ctx)
finetune_net.features = pretrained_net.features
finetune_net.classifier.initialize(init.Xavier())