用fastai ResNet50训练CIFAR10，85%准确度

版权声明：本文为博主原创文章，欢迎转载，并请注明出处。联系方式：460356155@qq.com

 

Fastai是在pytorch上封装的深度学习框架，效果出众，以下是训练CIFAR10的过程。

导入库

from fastai import *
from fastai.vision import *
from fastai.callbacks import CSVLogger, SaveModelCallback

验证集上训练结果计算和显示

def show_result(learn):
    # 得到验证集上的准确度
    probs, val_labels = learn.get_preds(ds_type=DatasetType.Valid)
    print('Accuracy', accuracy(probs, val_labels)),
    print('Error Rate', error_rate(probs, val_labels))

训练结果混淆矩阵及预测错误最多的类型显示

def show_matrix(learn):
    # 画训练结果的混合矩阵
    interp = ClassificationInterpretation.from_learner(learn)
    interp.confusion_matrix()
    interp.plot_confusion_matrix(dpi=120)

    # 显示判断错误最多的类型，min_val指定错误次数，默认1
    # 打印顺序为actual, predicted, number of occurrences.
    interp.most_confused(min_val=5)

    # 模型预测最困难的9个样本显示
    # 显示顺序为预测值、实际值、损失值、预测对的概率
    interp.plot_top_losses(9, figsize=(10, 10))

下载数据集，因调用linux的tar进行解压，在windows下会出错，可手动解压，解压后的目录：

 

# 下载数据集
untar_data(URLs.CIFAR)

# 训练数据目录
path = Path(r'C:\Users\Administrator\.fastai\data\cifar10')

定义数据及数据在线增强方式

# 数据在线增强方式定义
tfms = get_transforms(do_flip=False)

data = (ImageList.from_folder(path)  # Where to find the data? -> in path and its subfolders
        .split_by_rand_pct()  # How to split in train/valid? -> use the folders
        .label_from_folder()  # How to label? -> depending on the folder of the filenames
        .add_test_folder()  # Optionally add a test set (here default name is test)
        .transform(tfms, size=(32, 32))  # Data augmentation? -> use tfms with a size of 164
        .databunch(bs=128)  # Finally? -> use the defaults for conversion to ImageDataBunch
        .normalize(imagenet_stats))

 查看数据

# 查看数据信息
data.classes, data.c, data

(['airplane',
  'automobile',
  'bird',
  'cat',
  'deer',
  'dog',
  'frog',
  'horse',
  'ship',
  'truck'],
 10,
 ImageDataBunch;
 
 Train: LabelList (39072 items)
 x: ImageList
 Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32)
 y: CategoryList
 airplane,airplane,airplane,airplane,airplane
 Path: C:\Users\Administrator\.fastai\data\cifar10;
 
 Valid: LabelList (9767 items)
 x: ImageList
 Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32)
 y: CategoryList
 airplane,deer,deer,deer,automobile
 Path: C:\Users\Administrator\.fastai\data\cifar10;
 
 Test: LabelList (10000 items)
 x: ImageList
 Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32),Image (3, 32, 32)
 y: EmptyLabelList
 ,,,,
 Path: C:\Users\Administrator\.fastai\data\cifar10)

 创建训练器

# 创建learn
learn = cnn_learner(data, models.resnet50, metrics=[accuracy, error_rate], callback_fns=[ShowGraph, SaveModelCallback])

 第一阶段训练

# 最佳学习率寻找
learn.lr_find(end_lr=1)

LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.

# 画出学习率寻找曲线，给出建议学习率
learn.recorder.plot(suggestion=True)

 

# 根据学习率曲线得到max_lr，开始训练
learn.fit_one_cycle(cyc_len=15, max_lr=1.78e-2)

 

epoch	train_loss	valid_loss	accuracy	error_rate	time
0	1.074162	0.882136	0.709225	0.290775	01:17
1	0.824112	0.766163	0.740453	0.259547	01:16
2	0.811090	0.938345	0.707792	0.292208	01:16
3	0.799450	0.790665	0.733797	0.266203	01:16
4	0.763200	1.364758	0.752636	0.247364	01:18
5	0.693490	0.683559	0.776902	0.223098	01:16
6	0.673621	0.611799	0.800655	0.199345	01:16
7	0.665126	0.630715	0.796150	0.203850	01:16
8	0.612187	0.874567	0.826149	0.173851	01:16
9	0.563634	0.785189	0.820723	0.179277	01:16
10	0.515540	1.286271	0.829835	0.170165	01:21
11	0.485959	0.524455	0.840688	0.159312	01:16
12	0.444417	0.759944	0.842736	0.157264	01:17
13	0.419838	0.830482	0.845500	0.154500	01:17
14	0.421095	0.550606	0.844783	0.155217	01:16

 

 

Better model found at epoch 0 with val_loss value: 0.8821364045143127.
Better model found at epoch 1 with val_loss value: 0.7661632299423218.
Better model found at epoch 5 with val_loss value: 0.6835585832595825.
Better model found at epoch 6 with val_loss value: 0.6117991805076599.
Better model found at epoch 11 with val_loss value: 0.5244545340538025.

  训练结果

# 计算和显示训练结果
show_result(learn)

 Accuracy tensor(0.8407)

Error Rate tensor(0.1593)

 

# 保存训练模型
learn.save('stg1')

 

 

 第二阶段训练

 

learn.load('stg1')
learn.unfreeze()
learn.lr_find(end_lr=1)

 LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.

learn.recorder.plot(suggestion=True)

learn.fit_one_cycle(15, slice(1e-6, 5e-5))

epoch	train_loss	valid_loss	accuracy	error_rate	time
0	0.444569	0.521828	0.840381	0.159619	01:26
1	0.427062	0.513434	0.840483	0.159517	01:27
2	0.430344	0.514867	0.846524	0.153476	01:23
3	0.421480	0.550527	0.845295	0.154705	01:23
4	0.410170	0.506949	0.847855	0.152145	01:23
5	0.402150	0.542091	0.849186	0.150814	01:26
6	0.387639	0.491120	0.850927	0.149073	01:27
7	0.373022	0.511580	0.852155	0.147845	01:28
8	0.375497	0.505493	0.854101	0.145899	01:28
9	0.355466	0.585425	0.852462	0.147538	01:28
10	0.355327	0.506402	0.855534	0.144466	01:28
11	0.341208	0.498502	0.855944	0.144057	01:29
12	0.347057	0.549146	0.851746	0.148254	01:28
13	0.345185	0.533962	0.852155	0.147845	01:28
14	0.334336	0.504231	0.855432	0.144568	01:29

 

 

Better model found at epoch 0 with val_loss value: 0.5218283534049988.
Better model found at epoch 1 with val_loss value: 0.5134344696998596.
Better model found at epoch 4 with val_loss value: 0.5069490671157837.
Better model found at epoch 6 with val_loss value: 0.491120308637619.

训练结果

# 计算和显示训练结果
show_result(learn)

Accuracy tensor(0.8509)
Error Rate tensor(0.1491)

保存模型

learn.save('stg2')

　　

# 画训练结果的混合矩阵
interp = ClassificationInterpretation.from_learner(learn)
interp.confusion_matrix()
interp.plot_confusion_matrix(dpi=120)

 

 

显示预测错误次数最多的类型，错误次数大于5，输出顺序为actual, predicted, number of occurrences.

interp.most_confused(5)

[('bird', 'frog', 86),
 ('truck', 'automobile', 71),
 ('deer', 'frog', 66),
 ('dog', 'bird', 59),
 ('airplane', 'ship', 57),
 ('bird', 'airplane', 54),
 ('dog', 'frog', 54),
 ('bird', 'deer', 53),
 ('dog', 'deer', 50),
 ('cat', 'dog', 47),
 ('deer', 'bird', 47),
 ('automobile', 'truck', 45),
 ('ship', 'airplane', 45),
 ('cat', 'frog', 44),
 ('bird', 'dog', 37),
 ('ship', 'automobile', 34),
 ('ship', 'truck', 32),
 ('airplane', 'bird', 31),
 ('deer', 'dog', 26),
 ('frog', 'bird', 25),
 ('dog', 'cat', 24),
 ('dog', 'horse', 24),
 ('airplane', 'automobile', 23),
 ('horse', 'deer', 23),
 ('airplane', 'truck', 22),
 ('airplane', 'deer', 20),
 ('frog', 'deer', 17),
 ('cat', 'deer', 16),
 ('horse', 'dog', 14),
 ('automobile', 'ship', 13),
 ('deer', 'horse', 13),
 ('truck', 'ship', 13),
 ('bird', 'ship', 12),
 ('cat', 'bird', 12),
 ('deer', 'airplane', 12),
 ('dog', 'truck', 12),
 ('truck', 'airplane', 12),
 ('frog', 'dog', 11),
 ('airplane', 'frog', 10),
 ('deer', 'ship', 10),
 ('dog', 'airplane', 9),
 ('frog', 'automobile', 8),
 ('horse', 'frog', 8),
 ('ship', 'bird', 8),
 ('cat', 'truck', 7),
 ('horse', 'airplane', 7),
 ('horse', 'bird', 7),
 ('ship', 'deer', 7),
 ('dog', 'automobile', 6),
 ('truck', 'frog', 6),
 ('automobile', 'frog', 5),
 ('bird', 'cat', 5),
 ('bird', 'truck', 5),
 ('cat', 'ship', 5),
 ('dog', 'ship', 5),
 ('frog', 'airplane', 5)]

预测最困难的9个样本

# 模型预测最困难的9个样本显示
# 显示顺序为预测值、实际值、损失值、预测对的概率
interp.plot_top_losses(9, figsize=(10, 10))

 

转载于:https://www.cnblogs.com/zhengbiqing/p/10923342.html