Res2Net / app.py
Chujinze's picture
Create app.py
415635a
import gradio as gr
import torch.nn as nn
import math
import torch.utils.model_zoo as model_zoo
import torch
import torch.nn.functional as F
__all__ = ['Res2Net', 'res2net50_v1b', 'res2net101_v1b']
model_urls = {
'res2net50_v1b_26w_4s': 'https://shanghuagao.oss-cn-beijing.aliyuncs.com/res2net/res2net50_v1b_26w_4s-3cf99910.pth',
'res2net101_v1b_26w_4s': 'https://shanghuagao.oss-cn-beijing.aliyuncs.com/res2net/res2net101_v1b_26w_4s-0812c246.pth',
}
class Bottle2neck(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None, baseWidth=26, scale=4, stype='normal'):
""" Constructor
Args:
inplanes: input channel dimensionality
planes: output channel dimensionality
stride: conv stride. Replaces pooling layer.
downsample: None when stride = 1
baseWidth: basic width of conv3x3
scale: number of scale.
type: 'normal': normal set. 'stage': first block of a new stage.
"""
super(Bottle2neck, self).__init__()
width = int(math.floor(planes * (baseWidth / 64.0)))
self.conv1 = nn.Conv2d(inplanes, width * scale, kernel_size=1, bias=False)
self.bn1 = nn.BatchNorm2d(width * scale)
if scale == 1:
self.nums = 1
else:
self.nums = scale - 1
if stype == 'stage':
self.pool = nn.AvgPool2d(kernel_size=3, stride=stride, padding=1)
convs = []
bns = []
for i in range(self.nums):
convs.append(nn.Conv2d(width, width, kernel_size=3, stride=stride, padding=1, bias=False))
bns.append(nn.BatchNorm2d(width))
self.convs = nn.ModuleList(convs)
self.bns = nn.ModuleList(bns)
self.conv3 = nn.Conv2d(width * scale, planes * self.expansion, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * self.expansion)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stype = stype
self.scale = scale
self.width = width
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
spx = torch.split(out, self.width, 1)
for i in range(self.nums):
if i == 0 or self.stype == 'stage':
sp = spx[i]
else:
sp = sp + spx[i]
sp = self.convs[i](sp)
sp = self.relu(self.bns[i](sp))
if i == 0:
out = sp
else:
out = torch.cat((out, sp), 1)
if self.scale != 1 and self.stype == 'normal':
out = torch.cat((out, spx[self.nums]), 1)
elif self.scale != 1 and self.stype == 'stage':
out = torch.cat((out, self.pool(spx[self.nums])), 1)
out = self.conv3(out)
out = self.bn3(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual
out = self.relu(out)
return out
class Res2Net(nn.Module):
def __init__(self, block, layers, baseWidth=26, scale=4, num_classes=1000):
self.inplanes = 64
super(Res2Net, self).__init__()
self.baseWidth = baseWidth
self.scale = scale
self.conv1 = nn.Sequential(
nn.Conv2d(3, 32, 3, 2, 1, bias=False),
nn.BatchNorm2d(32),
nn.ReLU(inplace=True),
nn.Conv2d(32, 32, 3, 1, 1, bias=False),
nn.BatchNorm2d(32),
nn.ReLU(inplace=True),
nn.Conv2d(32, 64, 3, 1, 1, bias=False)
)
self.bn1 = nn.BatchNorm2d(64)
self.relu = nn.ReLU()
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.layer1 = self._make_layer(block, 64, layers[0])
self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
self.avgpool = nn.AdaptiveAvgPool2d(1)
self.fc = nn.Linear(512 * block.expansion, num_classes)
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
elif isinstance(m, nn.BatchNorm2d):
nn.init.constant_(m.weight, 1)
nn.init.constant_(m.bias, 0)
def _make_layer(self, block, planes, blocks, stride=1):
downsample = None
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
nn.AvgPool2d(kernel_size=stride, stride=stride,
ceil_mode=True, count_include_pad=False),
nn.Conv2d(self.inplanes, planes * block.expansion,
kernel_size=1, stride=1, bias=False),
nn.BatchNorm2d(planes * block.expansion),
)
layers = []
layers.append(block(self.inplanes, planes, stride, downsample=downsample,
stype='stage', baseWidth=self.baseWidth, scale=self.scale))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes, baseWidth=self.baseWidth, scale=self.scale))
return nn.Sequential(*layers)
def forward(self, x):
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.maxpool(x)
x = self.layer1(x)
x = self.layer2(x)
x = self.layer3(x)
x = self.layer4(x)
x = self.avgpool(x)
x = x.view(x.size(0), -1)
x = self.fc(x)
return x
def res2net50_v1b(pretrained=False, **kwargs):
"""Constructs a Res2Net-50_v1b model.
Res2Net-50 refers to the Res2Net-50_v1b_26w_4s.
Args:
pretrained (bool): If True, returns a model pre-trained on ImageNet
"""
model = Res2Net(Bottle2neck, [3, 4, 6, 3], baseWidth=26, scale=4, **kwargs)
if pretrained:
model.load_state_dict(model_zoo.load_url(model_urls['res2net50_v1b_26w_4s']))
return model
def res2net101_v1b(pretrained=False, **kwargs):
"""Constructs a Res2Net-50_v1b_26w_4s model.
Args:
pretrained (bool): If True, returns a model pre-trained on ImageNet
"""
model = Res2Net(Bottle2neck, [3, 4, 23, 3], baseWidth=26, scale=4, **kwargs)
if pretrained:
model.load_state_dict(model_zoo.load_url(model_urls['res2net101_v1b_26w_4s']))
return model
def res2net50_v1b_26w_4s(pretrained=False, **kwargs):
"""Constructs a Res2Net-50_v1b_26w_4s model.
Args:
pretrained (bool): If True, returns a model pre-trained on ImageNet
"""
model = Res2Net(Bottle2neck, [3, 4, 6, 3], baseWidth=26, scale=4, **kwargs)
if pretrained:
model.load_state_dict(torch.load(pthfile, map_location='cpu')) # load model
return model
def res2net101_v1b_26w_4s(pretrained=False, **kwargs):
"""Constructs a Res2Net-50_v1b_26w_4s model.
Args:
pretrained (bool): If True, returns a model pre-trained on ImageNet
"""
model = Res2Net(Bottle2neck, [3, 4, 23, 3], baseWidth=26, scale=4, **kwargs)
if pretrained:
model.load_state_dict(model_zoo.load_url(model_urls['res2net101_v1b_26w_4s']))
return model
def res2net152_v1b_26w_4s(pretrained=False, **kwargs):
"""Constructs a Res2Net-50_v1b_26w_4s model.
Args:
pretrained (bool): If True, returns a model pre-trained on ImageNet
"""
model = Res2Net(Bottle2neck, [3, 8, 36, 3], baseWidth=26, scale=4, **kwargs)
if pretrained:
model.load_state_dict(model_zoo.load_url(model_urls['res2net152_v1b_26w_4s']))
return model
class mutil_model(nn.Module):
def __init__(self, category_num=10):
super(mutil_model, self).__init__()
self.model1 = res2net50_v1b_26w_4s(pretrained=False)
self.model1.fc = nn.Sequential(
nn.Linear(in_features=2048, out_features=category_num, bias=True),
)
self.model2 = torch.load('./enet_b2_8' + '.pt', map_location=torch.device('cpu'))
self.model2.classifier = nn.Sequential(
nn.Linear(in_features=1408, out_features=category_num, bias=True),
)
self.fc = nn.Linear(in_features=category_num * 2, out_features=category_num, bias=True)
def forward(self, x):
x1 = self.model1(x)
x2 = self.model2(x)
x = torch.cat((x1, x2), 1)
x = self.fc(x)
return x
pth_path = './res2net_pretrain_model_999.pt'
category_num = 9
# "cuda" only when GPUs are available.
#device = "cuda" if torch.cuda.is_available() else "cpu"
device = "cpu"
#Initialize a model, and put it on the device specified.
# 导入res2net预训练模型
# pthfile = './res2net50_v1b.pth'
model = res2net50_v1b_26w_4s(pretrained=False)
# 修改全连接层,输出维度为预测 分类
num_ftrs = model.fc.in_features
model.fc = nn.Sequential(
nn.Linear(in_features=2048, out_features=1000, bias=True),
nn.Dropout(0.5),
nn.Linear(1000, out_features=category_num)
)
model.fc = nn.Sequential(
nn.Linear(in_features=2048, out_features=category_num, bias=True),
)
model = model.to(device)
model.device = device
model.load_state_dict(torch.load(pth_path,torch.device('cpu')))
model.eval()
# 增加人脸识别模型
#model = mutil_model(category_num=7)
#model_state = torch.load('./add_face_emotion_model_7.pt', map_location=torch.device('cpu')).state_dict()
#model.load_state_dict(model_state) # 加载模型参数
#model.eval()
labels = ['中国风', '古典', '电子', '摇滚', '乡村', '说唱', '民谣', '动漫', '现代']
import requests
import torch
import gradio as gr
import torchvision.transforms as transforms
# import cv2
# from PIL import Image
# PIL
# from PIL import Image
# inception_net = tf.keras.applications.MobileNetV2() # load the model
# Download human-readable labels for ImageNet.
# response = requests.get("https://git.io/JJkYN")
# labels = response.text.split("\n")
print(len(labels))
def classify_image(inp):
# inp = inp.convert('RGB')
# inp = Image.fromarray(inp.astype('uint8'), 'RGB')
transform_test = transforms.Compose([
# transforms.ToPILImage(),
transforms.Resize((256, 256)),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225)),
])
inp = transform_test(inp)
print(inp)
with torch.no_grad():
prediction = model(torch.unsqueeze(inp, 0)).flatten()
print(prediction)
prediction = torch.nn.Softmax(dim=0)(prediction)
print(prediction)
return {labels[i]: float(prediction[i].item()) for i in range(len(labels))}
# print(classify_image("/jj.jpg"))
# image = gr.inputs.Image(shape=(256, 256))
# image = gr.inputs.Image()
# print(image)
# label = gr.outputs.Label(num_top_classes=6)
gr.Interface(
classify_image,
# gr.inputs.Image(),
gr.inputs.Image(type='pil'),
outputs='label'
# inputs='image',
# outputs='label',
# examples=[["images/cheetah1.jpg"], ["images/lion.jpg"]],
).launch(share=True)
# share=True