def __init__(self, img_ch=3, output_ch=1):
super(SRF_UNet, self).__init__()
filters = [64, 128, 256, 512]
resnet = resnest.resnest50(pretrained=True)
self.firstconv = resnet.conv1
self.firstbn = resnet.bn1
self.firstrelu = resnet.relu
self.firstmaxpool = resnet.maxpool
self.encoder1 = resnet.layer1
self.encoder2 = resnet.layer2
self.encoder3 = resnet.layer3
self.encoder4 = resnet.layer4
#self.dab1 = DAB()
#self.dab2 = DAB()
#self.dab3 = DAB()
#self.dab4 = DAB()
#self.Up5_thick = up_conv(ch_in=2048, ch_out=1024)
#self.Up_conv5_thick = res_conv_block(ch_in=2048, ch_out=1024)
self.Up4_thick = up_conv(ch_in=1024, ch_out=512)
self.Up_conv4_thick = res_conv_block(ch_in=1024, ch_out=512)
self.Up3_thick = up_conv(ch_in=512, ch_out=256)
self.Up_conv3_thick = res_conv_block(ch_in=512, ch_out=256)
self.Up2_thick = up_conv(ch_in=256, ch_out=64)
self.Up_conv2_thick = res_conv_block(ch_in=128, ch_out=64)
self.Up1_thick = up_conv(ch_in=64, ch_out=64)
self.Up_conv1_thick = res_conv_block(ch_in=64, ch_out=32)
self.Conv_1x1_thick = nn.Conv2d(32, output_ch, kernel_size=1)
"""
def __init__(self):
super(ModelResnest50,self).__init__()
base_model = resnest50(pretrained=True)
self.conv = nn.Sequential(*list(base_model.children())[:-2])
self.avg_pool = nn.AdaptiveAvgPool2d((1, 1))
feature = 2048
self.classifier = nn.Linear(in_features=feature,out_features=5,bias=True)
# load image
#image_path = os.path.join("X:/model", "test.jpg")
img = Image.open("X:/dataset/公安部大楼/00000.jpg")
plt.imshow(img)
# [N, C, H, W]
img = data_transform(img)
# expand batch dimension
img = torch.unsqueeze(img, dim=0)
# read class_indict
try:
json_file = open('./class_indices.json', 'r')
class_indict = json.load(json_file)
except Exception as e:
print(e)
exit(-1)
# create model
model = resnest50(num_classes=5)
# load model weights
model_weight_path = "./resNest50.pth"
model.load_state_dict(torch.load(model_weight_path, map_location=device))
model.eval()
with torch.no_grad():
# predict class
output = torch.squeeze(model(img))
predict = torch.softmax(output, dim=0)
predict_cla = torch.argmax(predict).numpy()
print(class_indict[str(predict_cla)], predict[predict_cla].numpy())
plt.show()
def main():
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print("using {} device.".format(device))
data_transform = {
"train":
transforms.Compose([
transforms.RandomResizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
"val":
transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
}
data_root = os.path.abspath(os.path.join(os.getcwd(),
"../..")) # get data root path
image_path = os.path.join("D:", "imagenet") # flower data set path
assert os.path.exists(image_path), "{} path does not exist.".format(
image_path)
train_dataset = datasets.ImageFolder(root=os.path.join(
image_path, "train"),
transform=data_transform["train"])
train_num = len(train_dataset)
# {'daisy':0, 'dandelion':1, 'roses':2, 'sunflower':3, 'tulips':4}
flower_list = train_dataset.class_to_idx
cla_dict = dict((val, key) for key, val in flower_list.items())
# write dict into json file
json_str = json.dumps(cla_dict, indent=4)
with open('class_indices.json', 'w') as json_file:
json_file.write(json_str)
nw = min([os.cpu_count(), batch_size if batch_size > 1 else 0,
8]) # number of workers
print('Using {} dataloader workers every process'.format(nw))
train_loader = torch.utils.data.DataLoader(
train_dataset,
batch_size=batch_size,
shuffle=True,
)
validate_dataset = datasets.ImageFolder(root=os.path.join(
image_path, "val"),
transform=data_transform["val"])
val_num = len(validate_dataset)
validate_loader = torch.utils.data.DataLoader(
validate_dataset,
batch_size=batch_size,
shuffle=False,
)
print("using {} images for training, {} images for validation.".format(
train_num, val_num))
net = resnest50(num_classes=10)
net_input = torch.randn(batch_size, 3, 224, 224)
flops, params = profile(net, inputs=(net_input, ))
print("网络计算量GFlops:{},网络参数量#P：{}".format(flops, params))
'''
# load pretrain weights
# download url: https://download.pytorch.org/models/resnet34-333f7ec4.pth
#assert os.path.exists(model_weight_path), "file {} does not exist.".format(model_weight_path)
#####读取训练好的参数###############
model_weight_path = "./ResNeSkt.pth"
missing_keys, unexpected_keys = net.load_state_dict(torch.load(model_weight_path), strict=False)
for param in net.parameters():
param.requires_grad = False
'''
# change fc layer structure
in_channel = net.fc.in_features
net.fc = nn.Linear(in_channel, 10)
net.to(device)
loss_function = nn.CrossEntropyLoss()
optimizer = optim.Adam(net.parameters(), lr=lr)
best_acc = 0.0
save_path = './ResNeSkt.pth'
list_loss = list()
list_acc = list()
for epoch in range(epoch_all):
# train
net.train()
running_loss = 0.0
for step, data in enumerate(train_loader, start=0):
images, labels = data
optimizer.zero_grad()
logits = net(images.to(device))
loss = loss_function(logits, labels.to(device))
loss.backward()
optimizer.step()
# print statistics
running_loss += loss.item()
# print train process
rate = (step + 1) / len(train_loader)
a = "*" * int(rate * 50)
b = "." * int((1 - rate) * 50)
print("\rtrain loss: {:^3.0f}%[{}->{}]{:.4f}".format(
int(rate * 100), a, b, loss),
end="")
# validate
net.eval()
acc = 0.0 # accumulate accurate number / epoch
with torch.no_grad():
for val_data in validate_loader:
val_images, val_labels = val_data
outputs = net(val_images.to(
device)) # eval model only have last output layer
# loss = loss_function(outputs, test_labels)
predict_y = torch.max(outputs, dim=1)[1]
acc += (predict_y == val_labels.to(device)).sum().item()
val_accurate = acc / val_num
if val_accurate > best_acc:
best_acc = val_accurate
torch.save(net.state_dict(), save_path)
print('[epoch %d] train_loss: %.3f test_accuracy: %.3f' %
(epoch + 1, running_loss / step, val_accurate))
list_loss.append(running_loss / step)
list_acc.append(val_accurate)
if epoch % 20 == 0:
#保存loss and acc
file_loss_path = "./ResNeSkt_loss.txt"
with open(file_loss_path, "w") as file_object:
json.dump(list_loss, file_object)
file_acc_path = "./ResNeSkt_acc.txt"
with open(file_acc_path, "w") as file_object:
json.dump(list_acc, file_object)
print('Finished Training')
plt.plot(new_est[:, 0],
new_est[:, 1],
color='blue',
linewidth=1,
label='EST')
plt.show()
return
if __name__ == '__main__':
path = 'data/test/EvaluationFramework_ISMIR2014/DATASET'
f_path = 'data/test/Process_data/FEAT'
onset = [0.1, 0.256, 0.279, 0.336, 0.469, 0.53]
offset = [0.01, 0.23, 0.266, 0.267, 0.39, 0.4]
# model = get_Resnet(channel=hparam.FEAT_channel).to(device)
# model.load_state_dict(torch.load("checkpoint/1227_748HP.pth"))
from resnest import resnest50
model = resnest50(channel=9, num_classes=6).to(device)
model.load_state_dict(torch.load("checkpoint/3690.pth"))
model.eval()
print("load OK")
testset_evaluation(path,
f_path,
model=model,
timestep=0,
channel=hparam.FEAT_channel)