fastai 3-planet

最新推荐文章于 2021-08-03 15:01:55 发布

Witcher8848

最新推荐文章于 2021-08-03 15:01:55 发布

阅读量252

点赞数

分类专栏： fastai学习

本文链接：https://blog.csdn.net/qq_23341175/article/details/104715759

版权

fastai学习专栏收录该内容

13 篇文章 1 订阅

订阅专栏

多标签分类

导入包，路径

%reload_ext autoreload
%autoreload 2
%matplotlib inline
from fastai.vision import *

path = Config.data_path()/'planet'
path.mkdir(parents=True, exist_ok=True)
path

WindowsPath('C:/Users/Wither8848/.fastai/data/planet')

查看数据集

df = pd.read_csv(path/'train_v2.csv')
df.head()

	image_name	tags
0	train_0	haze primary
1	train_1	agriculture clear primary water
2	train_2	clear primary
3	train_3	clear primary
4	train_4	agriculture clear habitation primary road

卫星图像，允许flip_vert随机翻转；不允许max_warp随机角度观察，因为卫星的角度恒定

tfms = get_transforms(flip_vert=True, max_lighting=0.1, max_zoom=1.05, max_warp=0.)

从csv读取数据，20%验证，标签的分隔符是空格，这里使用ImageList而不是ImageDataBunch因为这个数据集是多标签的

np.random.seed(42)
src = (ImageList.from_csv(path, 'train_v2.csv', folder='train-jpg', suffix='.jpg')
       .split_by_rand_pct(0.2)
       .label_from_df(label_delim=' '))

矩阵变换，生成数据集

data = (src.transform(tfms, size=128)
        .databunch().normalize(imagenet_stats))

查看数据集

data.show_batch(rows=3, figsize=(12,9))

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-Yq7o0jsK-1583566610690)(output_13_0.png)]

载入模型

arch = models.resnet50

定义度量函数

这里使用了偏函数，一个函数但是他的参数需要被一直设定成一个值
thresh是一个门槛值对比于单标签取最大值，只要超过thresh就认作是标签

acc_02 = partial(accuracy_thresh, thresh=0.2)
f_score = partial(fbeta, thresh=0.2)
learn = cnn_learner(data, arch, metrics=[acc_02, f_score])

寻找学习率

learn.lr_find()
learn.recorder.plot()

<div>
    <style>
        /* Turns off some styling */
        progress {
            /* gets rid of default border in Firefox and Opera. */
            border: none;
            /* Needs to be in here for Safari polyfill so background images work as expected. */
            background-size: auto;
        }
        .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {
            background: #F44336;
        }
    </style>
  <progress value='0' class='' max='1', style='width:300px; height:20px; vertical-align: middle;'></progress>
  0.00% [0/1 00:00<00:00]
</div>

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time

<div>
    <style>
        /* Turns off some styling */
        progress {
            /* gets rid of default border in Firefox and Opera. */
            border: none;
            /* Needs to be in here for Safari polyfill so background images work as expected. */
            background-size: auto;
        }
        .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {
            background: #F44336;
        }
    </style>
  <progress value='91' class='' max='506', style='width:300px; height:20px; vertical-align: middle;'></progress>
  17.98% [91/506 00:43<03:19 1.9991]
</div>



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

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-1dpfEWRO-1583566610691)(output_20_2.png)]

lr = 0.01
learn.fit_one_cycle(5, slice(lr))

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time
0	0.137257	0.118548	0.940821	0.897103	02:00
1	0.109874	0.101886	0.951931	0.911088	02:03
2	0.103863	0.092873	0.950892	0.920988	02:03
3	0.094259	0.088274	0.953791	0.923835	01:58
4	0.091735	0.087416	0.956117	0.924584	02:03

保存模型

learn.save('stage-1-rn50')

解冻再训练

learn.unfreeze()

learn.lr_find()
learn.recorder.plot()

<div>
    <style>
        /* Turns off some styling */
        progress {
            /* gets rid of default border in Firefox and Opera. */
            border: none;
            /* Needs to be in here for Safari polyfill so background images work as expected. */
            background-size: auto;
        }
        .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {
            background: #F44336;
        }
    </style>
  <progress value='0' class='' max='1', style='width:300px; height:20px; vertical-align: middle;'></progress>
  0.00% [0/1 00:00<00:00]
</div>

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time

<div>
    <style>
        /* Turns off some styling */
        progress {
            /* gets rid of default border in Firefox and Opera. */
            border: none;
            /* Needs to be in here for Safari polyfill so background images work as expected. */
            background-size: auto;
        }
        .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {
            background: #F44336;
        }
    </style>
  <progress value='85' class='' max='506', style='width:300px; height:20px; vertical-align: middle;'></progress>
  16.80% [85/506 00:40<03:21 0.2928]
</div>



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

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-973q0RaK-1583566610691)(output_26_2.png)]

学习率的第一个值是最低点缩小10倍，第二个值是第一次的值/5

learn.fit_one_cycle(5, slice(1e-5, lr/5))

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time
0	0.096562	0.090489	0.956800	0.923122	02:24
1	0.099346	0.089586	0.951655	0.923686	02:24
2	0.092557	0.085414	0.957977	0.926809	02:27
3	0.085692	0.082906	0.958071	0.928684	02:28
4	0.079300	0.082472	0.958275	0.928988	02:30

learn.save('stage-2-rn50')

这里使用了小技巧

比赛要求的是256256，之前是使用128*128的，现在载入256的数据，避免了过拟合，可以再进行微调

载入新数据冻结-训练；解冻-再训练

learn.load('stage-2-rn50')

Learner(data=ImageDataBunch;

Train: LabelList (32384 items)
x: ImageList
Image (3, 128, 128),Image (3, 128, 128),Image (3, 128, 128),Image (3, 128, 128),Image (3, 128, 128)
y: MultiCategoryList
haze;primary,clear;primary,clear;primary,haze;primary;water,agriculture;clear;cultivation;primary;water
Path: C:\Users\Wither8848\.fastai\data\planet;

Valid: LabelList (8095 items)
x: ImageList
Image (3, 128, 128),Image (3, 128, 128),Image (3, 128, 128),Image (3, 128, 128),Image (3, 128, 128)
y: MultiCategoryList
clear;primary;road,clear;primary;water,clear;conventional_mine;habitation;primary;road;water,cloudy,agriculture;clear;cultivation;cultivation;habitation;primary;road;water
Path: C:\Users\Wither8848\.fastai\data\planet;

Test: None, model=Sequential(
  (0): Sequential(
    (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU(inplace=True)
    (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
    (4): Sequential(
      (0): Bottleneck(
        (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
        (downsample): Sequential(
          (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Bottleneck(
        (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): Bottleneck(
        (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (5): Sequential(
      (0): Bottleneck(
        (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
        (downsample): Sequential(
          (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
          (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Bottleneck(
        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): Bottleneck(
        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): Bottleneck(
        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (6): Sequential(
      (0): Bottleneck(
        (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
        (downsample): Sequential(
          (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
          (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Bottleneck(
        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): Bottleneck(
        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): Bottleneck(
        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (4): Bottleneck(
        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (5): Bottleneck(
        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (7): Sequential(
      (0): Bottleneck(
        (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
        (downsample): Sequential(
          (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
          (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Bottleneck(
        (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): Bottleneck(
        (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (1): Sequential(
    (0): AdaptiveConcatPool2d(
      (ap): AdaptiveAvgPool2d(output_size=1)
      (mp): AdaptiveMaxPool2d(output_size=1)
    )
    (1): Flatten()
    (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (3): Dropout(p=0.25, inplace=False)
    (4): Linear(in_features=4096, out_features=512, bias=True)
    (5): ReLU(inplace=True)
    (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (7): Dropout(p=0.5, inplace=False)
    (8): Linear(in_features=512, out_features=17, bias=True)
  )
), opt_func=functools.partial(<class 'torch.optim.adam.Adam'>, betas=(0.9, 0.99)), loss_func=FlattenedLoss of BCEWithLogitsLoss(), metrics=[functools.partial(<function accuracy_thresh at 0x000001F9943B0C18>, thresh=0.2), functools.partial(<function fbeta at 0x000001F98E441048>, thresh=0.2)], true_wd=True, bn_wd=True, wd=0.01, train_bn=True, path=WindowsPath('C:/Users/Wither8848/.fastai/data/planet'), model_dir='models', callback_fns=[functools.partial(<class 'fastai.basic_train.Recorder'>, add_time=True, silent=False)], callbacks=[], layer_groups=[Sequential(
  (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
  (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (2): ReLU(inplace=True)
  (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
  (4): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (5): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (6): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (7): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (8): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (9): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (10): ReLU(inplace=True)
  (11): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (12): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (13): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (14): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (15): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (16): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (17): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (18): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (19): ReLU(inplace=True)
  (20): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (21): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (22): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (23): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (24): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (25): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (26): ReLU(inplace=True)
  (27): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (28): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (29): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
  (30): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (31): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (32): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (33): ReLU(inplace=True)
  (34): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
  (35): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (36): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (37): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (38): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (39): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (40): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (41): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (42): ReLU(inplace=True)
  (43): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (44): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (45): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (46): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (47): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (48): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (49): ReLU(inplace=True)
  (50): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (51): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (52): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (53): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (54): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (55): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (56): ReLU(inplace=True)
), Sequential(
  (0): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
  (3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (4): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (5): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (6): ReLU(inplace=True)
  (7): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
  (8): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (9): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (10): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (11): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (12): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (13): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (14): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (15): ReLU(inplace=True)
  (16): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (17): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (18): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (19): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (20): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (21): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (22): ReLU(inplace=True)
  (23): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (24): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (25): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (26): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (27): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (28): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (29): ReLU(inplace=True)
  (30): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (31): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (32): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (33): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (34): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (35): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (36): ReLU(inplace=True)
  (37): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (38): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (39): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (40): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (41): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (42): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (43): ReLU(inplace=True)
  (44): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (45): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (46): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
  (47): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (48): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (49): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (50): ReLU(inplace=True)
  (51): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
  (52): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (53): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (54): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (55): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (56): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (57): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (58): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (59): ReLU(inplace=True)
  (60): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (61): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (62): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (63): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (64): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (65): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (66): ReLU(inplace=True)
), Sequential(
  (0): AdaptiveAvgPool2d(output_size=1)
  (1): AdaptiveMaxPool2d(output_size=1)
  (2): Flatten()
  (3): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (4): Dropout(p=0.25, inplace=False)
  (5): Linear(in_features=4096, out_features=512, bias=True)
  (6): ReLU(inplace=True)
  (7): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (8): Dropout(p=0.5, inplace=False)
  (9): Linear(in_features=512, out_features=17, bias=True)
)], add_time=True, silent=False)

data = (src.transform(tfms, size=256)
        .databunch().normalize(imagenet_stats))

learn.data = data
data.train_ds[0][0].shape

torch.Size([3, 256, 256])

我的gpu运行内存只有8g，内存溢出，所以调小batch

data.batch_size=32

learn.data = data
data.train_ds[0][0].shape

torch.Size([3, 256, 256])

learn.data.batch_size

开始训练

learn.freeze()

learn.lr_find()
learn.recorder.plot()

<div>
    <style>
        /* Turns off some styling */
        progress {
            /* gets rid of default border in Firefox and Opera. */
            border: none;
            /* Needs to be in here for Safari polyfill so background images work as expected. */
            background-size: auto;
        }
        .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {
            background: #F44336;
        }
    </style>
  <progress value='0' class='' max='1', style='width:300px; height:20px; vertical-align: middle;'></progress>
  0.00% [0/1 00:00<00:00]
</div>

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time

<div>
    <style>
        /* Turns off some styling */
        progress {
            /* gets rid of default border in Firefox and Opera. */
            border: none;
            /* Needs to be in here for Safari polyfill so background images work as expected. */
            background-size: auto;
        }
        .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {
            background: #F44336;
        }
    </style>
  <progress value='85' class='' max='1012', style='width:300px; height:20px; vertical-align: middle;'></progress>
  8.40% [85/1012 00:49<08:55 0.2795]
</div>



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

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-ePZN6RYy-1583566610691)(output_40_2.png)]

lr=1e-2/2

learn.fit_one_cycle(5, slice(lr))

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time
0	0.091085	0.086496	0.956938	0.925235	04:35
1	0.089153	0.085715	0.957999	0.926484	04:30
2	0.088167	0.084767	0.957367	0.928577	04:27
3	0.088339	0.083139	0.961015	0.930497	04:26
4	0.083278	0.082782	0.961429	0.929606	04:26

learn.save('stage-1-256-rn50')

learn.unfreeze()

learn.fit_one_cycle(5, slice(1e-5, lr/5))

epoch	train_loss	valid_loss	accuracy_thresh	fbeta	time
0	0.093252	0.086286	0.956073	0.925537	05:44
1	0.089903	0.086518	0.959111	0.926257	05:43
2	0.088383	0.083158	0.957730	0.929617	05:43
3	0.082047	0.082876	0.959539	0.928490	05:42
4	0.074914	0.081280	0.959823	0.930220	05:43

learn.recorder.plot_losses()

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-hsDUp1HF-1583566610692)(output_46_0.png)]

learn.save('stage-2-256-rn50')

learn.export()

Witcher8848

关注

0
点赞
踩
0

收藏

觉得还不错? 一键收藏
0
评论
fastai 3-planet

多标签分类导入包，路径%reload_ext autoreload%autoreload 2%matplotlib inlinefrom fastai.vision import *path = Config.data_path()/'planet'path.mkdir(parents=True, exist_ok=True)pathWindowsPath('C:/Users/W...
复制链接

扫一扫