參考自：知乎
import collections
import os
import shutil
import tqdm
import numpy as np
import PIL.Image
import torch
import torchvision
- 一般流程
#外部定義
criterion = nn.NLLLoss().cuda()
optimizer = optim.Adam(net.parameters(),lr=LR)
#訓(xùn)練過程
optimizer.zero_grad() # 梯度清0
out = net(imgData)
out = F.log_softmax(out,dim=1)
loss = criterion(out,imgLabel)
loss.backward() #反向傳播
optimizer.step() #更新梯度信息
————————————————
版權(quán)聲明：本文為CSDN博主「番茄土豆牛肉煲」的原創(chuàng)文章，遵循CC 4.0 BY-SA版權(quán)協(xié)議，轉(zhuǎn)載請附上原文出處鏈接及本聲明。
原文鏈接：https://blog.csdn.net/nienelong3319/article/details/106592371
- 基礎(chǔ)配置
檢查PyTorch版本
torch.__version__ # PyTorch version
torch.version.cuda # Corresponding CUDA version
torch.backends.cudnn.version() # Corresponding cuDNN version
torch.cuda.get_device_name(0) # GPU type
更新PyTorch
PyTorch將被安裝在anaconda3/lib/python3.7/site-packages/torch/目錄下。
conda update pytorch torchvision -c pytorch
固定隨機種子
torch.manual_seed(0)
torch.cuda.manual_seed_all(0)
指定程序運行在特定GPU卡上
在命令行指定環(huán)境變量
CUDA_VISIBLE_DEVICES=0,1 python train.py
或在代碼中指定
os.environ['CUDA_VISIBLE_DEVICES'] = '0,1'
判斷是否有CUDA支持
torch.cuda.is_available()
設(shè)置為cuDNN benchmark模式
Benchmark模式會提升計算速度，但是由于計算中有隨機性，每次網(wǎng)絡(luò)前饋結(jié)果略有差異。
torch.backends.cudnn.benchmark = True
如果想要避免這種結(jié)果波動，設(shè)置
torch.backends.cudnn.deterministic = True
清除GPU存儲
有時Control-C中止運行后GPU存儲沒有及時釋放，需要手動清空。在PyTorch內(nèi)部可以
torch.cuda.empty_cache()
或在命令行可以先使用ps找到程序的PID，再使用kill結(jié)束該進程
ps aux | grep python
kill -9 [pid]
或者直接重置沒有被清空的GPU
nvidia-smi --gpu-reset -i [gpu_id]
2. 張量處理
張量基本信息
tensor.type() # Data type
tensor.size() # Shape of the tensor. It is a subclass of Python tuple
tensor.dim() # Number of dimensions.
數(shù)據(jù)類型轉(zhuǎn)換
# Set default tensor type. Float in PyTorch is much faster than double.
torch.set_default_tensor_type(torch.FloatTensor)
# Type convertions.
tensor = tensor.cuda()
tensor = tensor.cpu()
tensor = tensor.float()
tensor = tensor.long()
torch.Tensor與np.ndarray轉(zhuǎn)換
# torch.Tensor -> np.ndarray.
ndarray = tensor.cpu().numpy()
# np.ndarray -> torch.Tensor.
tensor = torch.from_numpy(ndarray).float()
tensor = torch.from_numpy(ndarray.copy()).float() # If ndarray has negative stride
torch.Tensor與PIL.Image轉(zhuǎn)換
PyTorch中的張量默認(rèn)采用N×D×H×W的順序，并且數(shù)據(jù)范圍在[0, 1]，需要進行轉(zhuǎn)置和規(guī)范化。
# torch.Tensor -> PIL.Image.
image = PIL.Image.fromarray(torch.clamp(tensor * 255, min=0, max=255
).byte().permute(1, 2, 0).cpu().numpy())
image = torchvision.transforms.functional.to_pil_image(tensor) # Equivalently way
# PIL.Image -> torch.Tensor.
tensor = torch.from_numpy(np.asarray(PIL.Image.open(path))
).permute(2, 0, 1).float() / 255
tensor = torchvision.transforms.functional.to_tensor(PIL.Image.open(path)) # Equivalently way
np.ndarray與PIL.Image轉(zhuǎn)換
# np.ndarray -> PIL.Image.
image = PIL.Image.fromarray(ndarray.astypde(np.uint8))
# PIL.Image -> np.ndarray.
ndarray = np.asarray(PIL.Image.open(path))
從只包含一個元素的張量中提取值
這在訓(xùn)練時統(tǒng)計loss的變化過程中特別有用。否則這將累積計算圖，使GPU存儲占用量越來越大。
value = tensor.item()
張量形變
張量形變常常需要用于將卷積層特征輸入全連接層的情形。相比torch.view，torch.reshape可以自動處理輸入張量不連續(xù)的情況。
tensor = torch.reshape(tensor, shape)
打亂順序
tensor = tensor[torch.randperm(tensor.size(0))] # Shuffle the first dimension
水平翻轉(zhuǎn)
PyTorch不支持tensor[::-1]這樣的負(fù)步長操作，水平翻轉(zhuǎn)可以用張量索引實現(xiàn)。
# Assume tensor has shape N*D*H*W.
tensor = tensor[:, :, :, torch.arange(tensor.size(3) - 1, -1, -1).long()]
復(fù)制張量
有三種復(fù)制的方式，對應(yīng)不同的需求。
# Operation | New/Shared memory | Still in computation graph |
tensor.clone() # | New | Yes |
tensor.detach() # | Shared | No |
tensor.detach.clone()() # | New | No |
拼接張量
注意torch.cat和torch.stack的區(qū)別在于torch.cat沿著給定的維度拼接，而torch.stack會新增一維。例如當(dāng)參數(shù)是3個10×5的張量，torch.cat的結(jié)果是30×5的張量，而torch.stack的結(jié)果是3×10×5的張量。
tensor = torch.cat(list_of_tensors, dim=0)
tensor = torch.stack(list_of_tensors, dim=0)
將整數(shù)標(biāo)記轉(zhuǎn)換成獨熱（one-hot）編碼
PyTorch中的標(biāo)記默認(rèn)從0開始。
N = tensor.size(0)
one_hot = torch.zeros(N, num_classes).long()
one_hot.scatter_(dim=1, index=torch.unsqueeze(tensor, dim=1), src=torch.ones(N, num_classes).long())
得到非零/零元素
torch.nonzero(tensor) # Index of non-zero elements
torch.nonzero(tensor == 0) # Index of zero elements
torch.nonzero(tensor).size(0) # Number of non-zero elements
torch.nonzero(tensor == 0).size(0) # Number of zero elements
判斷兩個張量相等
torch.allclose(tensor1, tensor2) # float tensor
torch.equal(tensor1, tensor2) # int tensor
張量擴展
# Expand tensor of shape 64*512 to shape 64*512*7*7.
torch.reshape(tensor, (64, 512, 1, 1)).expand(64, 512, 7, 7)
矩陣乘法
# Matrix multiplication: (m*n) * (n*p) -> (m*p).
result = torch.mm(tensor1, tensor2)
# Batch matrix multiplication: (b*m*n) * (b*n*p) -> (b*m*p).
result = torch.bmm(tensor1, tensor2)
# Element-wise multiplication.
result = tensor1 * tensor2
計算兩組數(shù)據(jù)之間的兩兩歐式距離
# X1 is of shape m*d, X2 is of shape n*d.
dist = torch.sqrt(torch.sum((X1[:,None,:] - X2) ** 2, dim=2))
3. 模型定義
卷積層
最常用的卷積層配置是
conv = torch.nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1, bias=True)
conv = torch.nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=1, padding=0, bias=True)
如果卷積層配置比較復(fù)雜，不方便計算輸出大小時，可以利用如下可視化工具輔助
Convolution Visualizerezyang.github.io
GAP（Global average pooling）層
gap = torch.nn.AdaptiveAvgPool2d(output_size=1)
雙線性匯合（bilinear pooling）[1]
X = torch.reshape(N, D, H * W) # Assume X has shape N*D*H*W
X = torch.bmm(X, torch.transpose(X, 1, 2)) / (H * W) # Bilinear pooling
assert X.size() == (N, D, D)
X = torch.reshape(X, (N, D * D))
X = torch.sign(X) * torch.sqrt(torch.abs(X) + 1e-5) # Signed-sqrt normalization
X = torch.nn.functional.normalize(X) # L2 normalization
多卡同步BN（Batch normalization）
當(dāng)使用torch.nn.DataParallel將代碼運行在多張GPU卡上時，PyTorch的BN層默認(rèn)操作是各卡上數(shù)據(jù)獨立地計算均值和標(biāo)準(zhǔn)差，同步BN使用所有卡上的數(shù)據(jù)一起計算BN層的均值和標(biāo)準(zhǔn)差，緩解了當(dāng)批量大?。╞atch size）比較小時對均值和標(biāo)準(zhǔn)差估計不準(zhǔn)的情況，是在目標(biāo)檢測等任務(wù)中一個有效的提升性能的技巧。
vacancy/Synchronized-BatchNorm-PyTorchgithub.com
現(xiàn)在PyTorch官方已經(jīng)支持同步BN操作
sync_bn = torch.nn.SyncBatchNorm(num_features, eps=1e-05, momentum=0.1, affine=True,
track_running_stats=True)
將已有網(wǎng)絡(luò)的所有BN層改為同步BN層
def convertBNtoSyncBN(module, process_group=None):
'''Recursively replace all BN layers to SyncBN layer.
Args:
module[torch.nn.Module]. Network
'''
if isinstance(module, torch.nn.modules.batchnorm._BatchNorm):
sync_bn = torch.nn.SyncBatchNorm(module.num_features, module.eps, module.momentum,
module.affine, module.track_running_stats, process_group)
sync_bn.running_mean = module.running_mean
sync_bn.running_var = module.running_var
if module.affine:
sync_bn.weight = module.weight.clone().detach()
sync_bn.bias = module.bias.clone().detach()
return sync_bn
else:
for name, child_module in module.named_children():
setattr(module, name) = convert_syncbn_model(child_module, process_group=process_group))
return module
類似BN滑動平均
如果要實現(xiàn)類似BN滑動平均的操作，在forward函數(shù)中要使用原地（inplace）操作給滑動平均賦值。
class BN(torch.nn.Module)
def __init__(self):
...
self.register_buffer('running_mean', torch.zeros(num_features))
def forward(self, X):
...
self.running_mean += momentum * (current - self.running_mean)
計算模型整體參數(shù)量
num_parameters = sum(torch.numel(parameter) for parameter in model.parameters())
類似Keras的model.summary()輸出模型信息
sksq96/pytorch-summarygithub.com
模型權(quán)值初始化
注意model.modules()和model.children()的區(qū)別：model.modules()會迭代地遍歷模型的所有子層，而model.children()只會遍歷模型下的一層。
# Common practise for initialization.
for layer in model.modules():
if isinstance(layer, torch.nn.Conv2d):
torch.nn.init.kaiming_normal_(layer.weight, mode='fan_out',
nonlinearity='relu')
if layer.bias is not None:
torch.nn.init.constant_(layer.bias, val=0.0)
elif isinstance(layer, torch.nn.BatchNorm2d):
torch.nn.init.constant_(layer.weight, val=1.0)
torch.nn.init.constant_(layer.bias, val=0.0)
elif isinstance(layer, torch.nn.Linear):
torch.nn.init.xavier_normal_(layer.weight)
if layer.bias is not None:
torch.nn.init.constant_(layer.bias, val=0.0)
# Initialization with given tensor.
layer.weight = torch.nn.Parameter(tensor)
部分層使用預(yù)訓(xùn)練模型
注意如果保存的模型是torch.nn.DataParallel，則當(dāng)前的模型也需要是torch.nn.DataParallel。torch.nn.DataParallel(model).module == model。
model.load_state_dict(torch.load('model,pth'), strict=False)
將在GPU保存的模型加載到CPU
model.load_state_dict(torch.load('model,pth', map_location='cpu'))
4. 數(shù)據(jù)準(zhǔn)備、特征提取與微調(diào)
圖像分塊打散（image shuffle）/區(qū)域混淆機制（region confusion mechanism，RCM）[2]
# X is torch.Tensor of size N*D*H*W.
# Shuffle rows
Q = (torch.unsqueeze(torch.arange(num_blocks), dim=1) * torch.ones(1, num_blocks).long()
+ torch.randint(low=-neighbour, high=neighbour, size=(num_blocks, num_blocks)))
Q = torch.argsort(Q, dim=0)
assert Q.size() == (num_blocks, num_blocks)
X = [torch.chunk(row, chunks=num_blocks, dim=2)
for row in torch.chunk(X, chunks=num_blocks, dim=1)]
X = [[X[Q[i, j].item()][j] for j in range(num_blocks)]
for i in range(num_blocks)]
# Shulle columns.
Q = (torch.ones(num_blocks, 1).long() * torch.unsqueeze(torch.arange(num_blocks), dim=0)
+ torch.randint(low=-neighbour, high=neighbour, size=(num_blocks, num_blocks)))
Q = torch.argsort(Q, dim=1)
assert Q.size() == (num_blocks, num_blocks)
X = [[X[i][Q[i, j].item()] for j in range(num_blocks)]
for i in range(num_blocks)]
Y = torch.cat([torch.cat(row, dim=2) for row in X], dim=1)
得到視頻數(shù)據(jù)基本信息
import cv2
video = cv2.VideoCapture(mp4_path)
height = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT))
width = int(video.get(cv2.CAP_PROP_FRAME_WIDTH))
num_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
fps = int(video.get(cv2.CAP_PROP_FPS))
video.release()
TSN每段（segment）采樣一幀視頻[3]
K = self._num_segments
if is_train:
if num_frames > K:
# Random index for each segment.
frame_indices = torch.randint(
high=num_frames // K, size=(K,), dtype=torch.long)
frame_indices += num_frames // K * torch.arange(K)
else:
frame_indices = torch.randint(
high=num_frames, size=(K - num_frames,), dtype=torch.long)
frame_indices = torch.sort(torch.cat((
torch.arange(num_frames), frame_indices)))[0]
else:
if num_frames > K:
# Middle index for each segment.
frame_indices = num_frames / K // 2
frame_indices += num_frames // K * torch.arange(K)
else:
frame_indices = torch.sort(torch.cat((
torch.arange(num_frames), torch.arange(K - num_frames))))[0]
assert frame_indices.size() == (K,)
return [frame_indices[i] for i in range(K)]
提取ImageNet預(yù)訓(xùn)練模型某層的卷積特征
# VGG-16 relu5-3 feature.
model = torchvision.models.vgg16(pretrained=True).features[:-1]
# VGG-16 pool5 feature.
model = torchvision.models.vgg16(pretrained=True).features
# VGG-16 fc7 feature.
model = torchvision.models.vgg16(pretrained=True)
model.classifier = torch.nn.Sequential(*list(model.classifier.children())[:-3])
# ResNet GAP feature.
model = torchvision.models.resnet18(pretrained=True)
model = torch.nn.Sequential(collections.OrderedDict(
list(model.named_children())[:-1]))
with torch.no_grad():
model.eval()
conv_representation = model(image)
提取ImageNet預(yù)訓(xùn)練模型多層的卷積特征
class FeatureExtractor(torch.nn.Module):
"""Helper class to extract several convolution features from the given
pre-trained model.
Attributes:
_model, torch.nn.Module.
_layers_to_extract, list<str> or set<str>
Example:
>>> model = torchvision.models.resnet152(pretrained=True)
>>> model = torch.nn.Sequential(collections.OrderedDict(
list(model.named_children())[:-1]))
>>> conv_representation = FeatureExtractor(
pretrained_model=model,
layers_to_extract={'layer1', 'layer2', 'layer3', 'layer4'})(image)
"""
def __init__(self, pretrained_model, layers_to_extract):
torch.nn.Module.__init__(self)
self._model = pretrained_model
self._model.eval()
self._layers_to_extract = set(layers_to_extract)
def forward(self, x):
with torch.no_grad():
conv_representation = []
for name, layer in self._model.named_children():
x = layer(x)
if name in self._layers_to_extract:
conv_representation.append(x)
return conv_representation
微調(diào)全連接層
model = torchvision.models.resnet18(pretrained=True)
for param in model.parameters():
param.requires_grad = False
model.fc = nn.Linear(512, 100) # Replace the last fc layer
optimizer = torch.optim.SGD(model.fc.parameters(), lr=1e-2, momentum=0.9, weight_decay=1e-4)
以較大學(xué)習(xí)率微調(diào)全連接層，較小學(xué)習(xí)率微調(diào)卷積層
model = torchvision.models.resnet18(pretrained=True)
finetuned_parameters = list(map(id, model.fc.parameters()))
conv_parameters = (p for p in model.parameters() if id(p) not in finetuned_parameters)
parameters = [{'params': conv_parameters, 'lr': 1e-3},
{'params': model.fc.parameters()}]
optimizer = torch.optim.SGD(parameters, lr=1e-2, momentum=0.9, weight_decay=1e-4)
5. 模型訓(xùn)練
常用訓(xùn)練和驗證數(shù)據(jù)預(yù)處理
其中ToTensor操作會將PIL.Image或形狀為H×W×D，數(shù)值范圍為[0, 255]的np.ndarray轉(zhuǎn)換為形狀為D×H×W，數(shù)值范圍為[0.0, 1.0]的torch.Tensor。
train_transform = torchvision.transforms.Compose([
torchvision.transforms.RandomResizedCrop(size=224,
scale=(0.08, 1.0)),
torchvision.transforms.RandomHorizontalFlip(),
torchvision.transforms.ToTensor(),
torchvision.transforms.Normalize(mean=(0.485, 0.456, 0.406),
std=(0.229, 0.224, 0.225)),
])
val_transform = torchvision.transforms.Compose([
torchvision.transforms.Resize(256),
torchvision.transforms.CenterCrop(224),
torchvision.transforms.ToTensor(),
torchvision.transforms.Normalize(mean=(0.485, 0.456, 0.406),
std=(0.229, 0.224, 0.225)),
])
訓(xùn)練基本代碼框架
for t in epoch(80):
for images, labels in tqdm.tqdm(train_loader, desc='Epoch %3d' % (t + 1)):
images, labels = images.cuda(), labels.cuda()
scores = model(images)
loss = loss_function(scores, labels)
optimizer.zero_grad()
loss.backward()
optimizer.step()
標(biāo)記平滑（label smoothing）[4]
for images, labels in train_loader:
images, labels = images.cuda(), labels.cuda()
N = labels.size(0)
# C is the number of classes.
smoothed_labels = torch.full(size=(N, C), fill_value=0.1 / (C - 1)).cuda()
smoothed_labels.scatter_(dim=1, index=torch.unsqueeze(labels, dim=1), value=0.9)
score = model(images)
log_prob = torch.nn.functional.log_softmax(score, dim=1)
loss = -torch.sum(log_prob * smoothed_labels) / N
optimizer.zero_grad()
loss.backward()
optimizer.step()
Mixup[5]
beta_distribution = torch.distributions.beta.Beta(alpha, alpha)
for images, labels in train_loader:
images, labels = images.cuda(), labels.cuda()
# Mixup images.
lambda_ = beta_distribution.sample([]).item()
index = torch.randperm(images.size(0)).cuda()
mixed_images = lambda_ * images + (1 - lambda_) * images[index, :]
# Mixup loss.
scores = model(mixed_images)
loss = (lambda_ * loss_function(scores, labels)
+ (1 - lambda_) * loss_function(scores, labels[index]))
optimizer.zero_grad()
loss.backward()
optimizer.step()
L1正則化
l1_regularization = torch.nn.L1Loss(reduction='sum')
loss = ... # Standard cross-entropy loss
for param in model.parameters():
loss += lambda_ * torch.sum(torch.abs(param))
loss.backward()
不對偏置項進行L2正則化/權(quán)值衰減（weight decay）
bias_list = (param for name, param in model.named_parameters() if name[-4:] == 'bias')
others_list = (param for name, param in model.named_parameters() if name[-4:] != 'bias')
parameters = [{'parameters': bias_list, 'weight_decay': 0},
{'parameters': others_list}]
optimizer = torch.optim.SGD(parameters, lr=1e-2, momentum=0.9, weight_decay=1e-4)
梯度裁剪（gradient clipping）
torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=20)
計算Softmax輸出的準(zhǔn)確率
score = model(images)
prediction = torch.argmax(score, dim=1)
num_correct = torch.sum(prediction == labels).item()
accuruacy = num_correct / labels.size(0)
可視化模型前饋的計算圖
szagoruyko/pytorchvizgithub.com
可視化學(xué)習(xí)曲線
有Facebook自己開發(fā)的Visdom和Tensorboard（仍處于實驗階段）兩個選擇。
facebookresearch/visdomgithub.com torch.utils.tensorboard - PyTorch master documentationpytorch.org
# Example using Visdom.
vis = visdom.Visdom(env='Learning curve', use_incoming_socket=False)
assert self._visdom.check_connection()
self._visdom.close()
options = collections.namedtuple('Options', ['loss', 'acc', 'lr'])(
loss={'xlabel': 'Epoch', 'ylabel': 'Loss', 'showlegend': True},
acc={'xlabel': 'Epoch', 'ylabel': 'Accuracy', 'showlegend': True},
lr={'xlabel': 'Epoch', 'ylabel': 'Learning rate', 'showlegend': True})
for t in epoch(80):
tran(...)
val(...)
vis.line(X=torch.Tensor([t + 1]), Y=torch.Tensor([train_loss]),
name='train', win='Loss', update='append', opts=options.loss)
vis.line(X=torch.Tensor([t + 1]), Y=torch.Tensor([val_loss]),
name='val', win='Loss', update='append', opts=options.loss)
vis.line(X=torch.Tensor([t + 1]), Y=torch.Tensor([train_acc]),
name='train', win='Accuracy', update='append', opts=options.acc)
vis.line(X=torch.Tensor([t + 1]), Y=torch.Tensor([val_acc]),
name='val', win='Accuracy', update='append', opts=options.acc)
vis.line(X=torch.Tensor([t + 1]), Y=torch.Tensor([lr]),
win='Learning rate', update='append', opts=options.lr)
得到當(dāng)前學(xué)習(xí)率
# If there is one global learning rate (which is the common case).
lr = next(iter(optimizer.param_groups))['lr']
# If there are multiple learning rates for different layers.
all_lr = []
for param_group in optimizer.param_groups:
all_lr.append(param_group['lr'])
學(xué)習(xí)率衰減
# Reduce learning rate when validation accuarcy plateau.
scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='max', patience=5, verbose=True)
for t in range(0, 80):
train(...); val(...)
scheduler.step(val_acc)
# Cosine annealing learning rate.
scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=80)
# Reduce learning rate by 10 at given epochs.
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=[50, 70], gamma=0.1)
for t in range(0, 80):
scheduler.step()
train(...); val(...)
# Learning rate warmup by 10 epochs.
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda t: t / 10)
for t in range(0, 10):
scheduler.step()
train(...); val(...)
保存與加載斷點
# 保存整個網(wǎng)絡(luò)
torch.save(net, PATH)
# 保存網(wǎng)絡(luò)中的參數(shù), 速度快，占空間少
torch.save(net.state_dict(),PATH)
#--------------------------------------------------
#針對上面一般的保存方法，加載的方法分別是：
model_dict=torch.load(PATH)
model_dict=model.load_state_dict(torch.load(PATH))
注意為了能夠恢復(fù)訓(xùn)練，我們需要同時保存模型和優(yōu)化器的狀態(tài)，以及當(dāng)前的訓(xùn)練輪數(shù)。
# Save checkpoint.
is_best = current_acc > best_acc
best_acc = max(best_acc, current_acc)
checkpoint = {
'best_acc': best_acc,
'epoch': t + 1,
'model': model.state_dict(),
'optimizer': optimizer.state_dict(),
}
model_path = os.path.join('model', 'checkpoint.pth.tar')
torch.save(checkpoint, model_path)
if is_best:
shutil.copy('checkpoint.pth.tar', model_path)
# Load checkpoint.
if resume:
model_path = os.path.join('model', 'checkpoint.pth.tar')
assert os.path.isfile(model_path)
checkpoint = torch.load(model_path)
best_acc = checkpoint['best_acc']
start_epoch = checkpoint['epoch']
model.load_state_dict(checkpoint['model'])
optimizer.load_state_dict(checkpoint['optimizer'])
print('Load checkpoint at epoch %d.' % start_epoch)
計算準(zhǔn)確率、查準(zhǔn)率（precision）、查全率（recall）
# data['label'] and data['prediction'] are groundtruth label and prediction
# for each image, respectively.
accuracy = np.mean(data['label'] == data['prediction']) * 100
# Compute recision and recall for each class.
for c in range(len(num_classes)):
tp = np.dot((data['label'] == c).astype(int),
(data['prediction'] == c).astype(int))
tp_fp = np.sum(data['prediction'] == c)
tp_fn = np.sum(data['label'] == c)
precision = tp / tp_fp * 100
recall = tp / tp_fn * 100
6. 模型測試
計算每個類別的查準(zhǔn)率（precision）、查全率（recall）、F1和總體指標(biāo)
import sklearn.metrics
all_label = []
all_prediction = []
for images, labels in tqdm.tqdm(data_loader):
# Data.
images, labels = images.cuda(), labels.cuda()
# Forward pass.
score = model(images)
# Save label and predictions.
prediction = torch.argmax(score, dim=1)
all_label.append(labels.cpu().numpy())
all_prediction.append(prediction.cpu().numpy())
# Compute RP and confusion matrix.
all_label = np.concatenate(all_label)
assert len(all_label.shape) == 1
all_prediction = np.concatenate(all_prediction)
assert all_label.shape == all_prediction.shape
micro_p, micro_r, micro_f1, _ = sklearn.metrics.precision_recall_fscore_support(
all_label, all_prediction, average='micro', labels=range(num_classes))
class_p, class_r, class_f1, class_occurence = sklearn.metrics.precision_recall_fscore_support(
all_label, all_prediction, average=None, labels=range(num_classes))
# Ci,j = #{y=i and hat_y=j}
confusion_mat = sklearn.metrics.confusion_matrix(
all_label, all_prediction, labels=range(num_classes))
assert confusion_mat.shape == (num_classes, num_classes)
將各類結(jié)果寫入電子表格
import csv
# Write results onto disk.
with open(os.path.join(path, filename), 'wt', encoding='utf-8') as f:
f = csv.writer(f)
f.writerow(['Class', 'Label', '# occurence', 'Precision', 'Recall', 'F1',
'Confused class 1', 'Confused class 2', 'Confused class 3',
'Confused 4', 'Confused class 5'])
for c in range(num_classes):
index = np.argsort(confusion_mat[:, c])[::-1][:5]
f.writerow([
label2class[c], c, class_occurence[c], '%4.3f' % class_p[c],
'%4.3f' % class_r[c], '%4.3f' % class_f1[c],
'%s:%d' % (label2class[index[0]], confusion_mat[index[0], c]),
'%s:%d' % (label2class[index[1]], confusion_mat[index[1], c]),
'%s:%d' % (label2class[index[2]], confusion_mat[index[2], c]),
'%s:%d' % (label2class[index[3]], confusion_mat[index[3], c]),
'%s:%d' % (label2class[index[4]], confusion_mat[index[4], c])])
f.writerow(['All', '', np.sum(class_occurence), micro_p, micro_r, micro_f1,
'', '', '', '', ''])
7. PyTorch其他注意事項
模型定義
- 建議有參數(shù)的層和匯合（pooling）層使用torch.nn模塊定義，激活函數(shù)直接使用torch.nn.functional。torch.nn模塊和torch.nn.functional的區(qū)別在于，torch.nn模塊在計算時底層調(diào)用了torch.nn.functional，但torch.nn模塊包括該層參數(shù)，還可以應(yīng)對訓(xùn)練和測試兩種網(wǎng)絡(luò)狀態(tài)。使用torch.nn.functional時要注意網(wǎng)絡(luò)狀態(tài)，如
def forward(self, x):
...
x = torch.nn.functional.dropout(x, p=0.5, training=self.training)
- model(x)前用model.train()和model.eval()切換網(wǎng)絡(luò)狀態(tài)。
- 不需要計算梯度的代碼塊用with torch.no_grad()包含起來。model.eval()和torch.no_grad()的區(qū)別在于，model.eval()是將網(wǎng)絡(luò)切換為測試狀態(tài)，例如BN和隨機失活（dropout）在訓(xùn)練和測試階段使用不同的計算方法。torch.no_grad()是關(guān)閉PyTorch張量的自動求導(dǎo)機制，以減少存儲使用和加速計算，得到的結(jié)果無法進行l(wèi)oss.backward()。
- torch.nn.CrossEntropyLoss的輸入不需要經(jīng)過Softmax。torch.nn.CrossEntropyLoss等價于torch.nn.functional.log_softmax + torch.nn.NLLLoss。
- loss.backward()前用optimizer.zero_grad()清除累積梯度。optimizer.zero_grad()和model.zero_grad()效果一樣。
PyTorch性能與調(diào)試
- torch.utils.data.DataLoader中盡量設(shè)置pin_memory=True，對特別小的數(shù)據(jù)集如MNIST設(shè)置pin_memory=False反而更快一些。num_workers的設(shè)置需要在實驗中找到最快的取值。
- 用del及時刪除不用的中間變量，節(jié)約GPU存儲。
- 使用inplace操作可節(jié)約GPU存儲，如
x = torch.nn.functional.relu(x, inplace=True)
此外，還可以通過torch.utils.checkpoint前向傳播時只保留一部分中間結(jié)果來節(jié)約GPU存儲使用，在反向傳播時需要的內(nèi)容從最近中間結(jié)果中計算得到。
- 減少CPU和GPU之間的數(shù)據(jù)傳輸。例如如果你想知道一個epoch中每個mini-batch的loss和準(zhǔn)確率，先將它們累積在GPU中等一個epoch結(jié)束之后一起傳輸回CPU會比每個mini-batch都進行一次GPU到CPU的傳輸更快。
- 使用半精度浮點數(shù)half()會有一定的速度提升，具體效率依賴于GPU型號。需要小心數(shù)值精度過低帶來的穩(wěn)定性問題。
- 時常使用assert tensor.size() == (N, D, H, W)作為調(diào)試手段，確保張量維度和你設(shè)想中一致。
- 除了標(biāo)記y外，盡量少使用一維張量，使用n*1的二維張量代替，可以避免一些意想不到的一維張量計算結(jié)果。
- 統(tǒng)計代碼各部分耗時
with torch.autograd.profiler.profile(enabled=True, use_cuda=False) as profile:
...
print(profile)
或者在命令行運行
python -m torch.utils.bottleneck main.py
7. 查看可訓(xùn)練參數(shù)
for name, param in model.named_parameters():
if param.requires_grad:
print(name)
