PyTorch Lightning 1.1: research : CIFAR100 (SqueezeNet)
作成 : (株)クラスキャット セールスインフォメーション
作成日時 : 02/24/2021 (1.1.x)
* 本ページは以下の CIFAR10 用リソースを参考に CIFAR100 で遂行した実験結果のレポートです：
- notebooks : PyTorch Lightning CIFAR10 ~94% Baseline Tutorial
- Train CIFAR10 with PyTorch
- Pytorch-cifar100
* ご自由にリンクを張って頂いてかまいませんが、sales-info@classcat.com までご一報いただけると嬉しいです。
★ 無料セミナー実施中 ★ クラスキャット主催 人工知能 & ビジネス Web セミナー
人工知能とビジネスをテーマにウェビナー (WEB セミナー) を定期的に開催しています。スケジュールは弊社 公式 Web サイト でご確認頂けます。
- お住まいの地域に関係なく Web ブラウザからご参加頂けます。事前登録 が必要ですのでご注意ください。
- Windows PC のブラウザからご参加が可能です。スマートデバイスもご利用可能です。
クラスキャットは人工知能・テレワークに関する各種サービスを提供しております :
| 人工知能研究開発支援 | 人工知能研修サービス | テレワーク & オンライン授業を支援 |
| PoC(概念実証)を失敗させないための支援 (本支援はセミナーに参加しアンケートに回答した方を対象としています。) | ||
◆ お問合せ : 本件に関するお問い合わせ先は下記までお願いいたします。
| 株式会社クラスキャット セールス・マーケティング本部 セールス・インフォメーション |
| E-Mail：sales-info@classcat.com ; WebSite: https://www.classcat.com/ |
| Facebook: https://www.facebook.com/ClassCatJP/ |
research: CIFAR100 (SqueezeNet)
仕様
- Total params: 781,156 (781 K)
- Trainable params: 781,156
- Non-trainable params: 0
結果
100 エポック
- {‘test_acc’: 0.6442000269889832, ‘test_loss’: 1.5194581747055054}
- Wall time: 1h 1min 44s
- Tesla T4
- ReduceLROnPlateau
CIFAR 100 DM
from typing import Any, Callable, Optional, Sequence, Union
from pl_bolts.datamodules.vision_datamodule import VisionDataModule
#from pl_bolts.datasets import TrialCIFAR10
#from pl_bolts.transforms.dataset_normalizations import cifar10_normalization
from pl_bolts.utils import _TORCHVISION_AVAILABLE
from pl_bolts.utils.warnings import warn_missing_pkg
if _TORCHVISION_AVAILABLE:
from torchvision import transforms
#from torchvision import transforms as transform_lib
from torchvision.datasets import CIFAR100
else: # pragma: no cover
warn_missing_pkg('torchvision')
CIFAR100 = None
def cifar100_normalization():
if not _TORCHVISION_AVAILABLE: # pragma: no cover
raise ModuleNotFoundError(
'You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.'
)
normalize = transforms.Normalize(
mean=[x / 255.0 for x in [129.3, 124.1, 112.4]],
std=[x / 255.0 for x in [68.2, 65.4, 70.4]],
# cifar10
#mean=[x / 255.0 for x in [125.3, 123.0, 113.9]],
#std=[x / 255.0 for x in [63.0, 62.1, 66.7]],
)
return normalize
class CIFAR100DataModule(VisionDataModule):
"""
.. figure:: https://3qeqpr26caki16dnhd19sv6by6v-wpengine.netdna-ssl.com/wp-content/uploads/2019/01/
Plot-of-a-Subset-of-Images-from-the-CIFAR-10-Dataset.png
:width: 400
:alt: CIFAR-10
Specs:
- 10 classes (1 per class)
- Each image is (3 x 32 x 32)
Standard CIFAR10, train, val, test splits and transforms
Transforms::
mnist_transforms = transform_lib.Compose([
transform_lib.ToTensor(),
transforms.Normalize(
mean=[x / 255.0 for x in [125.3, 123.0, 113.9]],
std=[x / 255.0 for x in [63.0, 62.1, 66.7]]
)
])
Example::
from pl_bolts.datamodules import CIFAR10DataModule
dm = CIFAR10DataModule(PATH)
model = LitModel()
Trainer().fit(model, datamodule=dm)
Or you can set your own transforms
Example::
dm.train_transforms = ...
dm.test_transforms = ...
dm.val_transforms = ...
"""
name = "cifar100"
dataset_cls = CIFAR100
dims = (3, 32, 32)
def __init__(
self,
data_dir: Optional[str] = None,
val_split: Union[int, float] = 0.2,
num_workers: int = 16,
normalize: bool = False,
batch_size: int = 32,
seed: int = 42,
shuffle: bool = False,
pin_memory: bool = False,
drop_last: bool = False,
*args: Any,
**kwargs: Any,
) -> None:
"""
Args:
data_dir: Where to save/load the data
val_split: Percent (float) or number (int) of samples to use for the validation split
num_workers: How many workers to use for loading data
normalize: If true applies image normalize
batch_size: How many samples per batch to load
seed: Random seed to be used for train/val/test splits
shuffle: If true shuffles the train data every epoch
pin_memory: If true, the data loader will copy Tensors into CUDA pinned memory before
returning them
drop_last: If true drops the last incomplete batch
"""
super().__init__( # type: ignore[misc]
data_dir=data_dir,
val_split=val_split,
num_workers=num_workers,
normalize=normalize,
batch_size=batch_size,
seed=seed,
shuffle=shuffle,
pin_memory=pin_memory,
drop_last=drop_last,
*args,
**kwargs,
)
@property
def num_samples(self) -> int:
train_len, _ = self._get_splits(len_dataset=50_000)
return train_len
@property
def num_classes(self) -> int:
"""
Return:
10
"""
return 100
def default_transforms(self) -> Callable:
if self.normalize:
cf100_transforms = transforms.Compose([transform_lib.ToTensor(), cifar100_normalization()])
else:
cf100_transforms = transforms.Compose([transform_lib.ToTensor()])
return cf100_transforms
モデル
import torch import torch.nn as nn
class Fire(nn.Module):
def __init__(self, in_channel, out_channel, squzee_channel):
super().__init__()
self.squeeze = nn.Sequential(
nn.Conv2d(in_channel, squzee_channel, 1),
nn.BatchNorm2d(squzee_channel),
nn.ReLU(inplace=True)
)
self.expand_1x1 = nn.Sequential(
nn.Conv2d(squzee_channel, int(out_channel / 2), 1),
nn.BatchNorm2d(int(out_channel / 2)),
nn.ReLU(inplace=True)
)
self.expand_3x3 = nn.Sequential(
nn.Conv2d(squzee_channel, int(out_channel / 2), 3, padding=1),
nn.BatchNorm2d(int(out_channel / 2)),
nn.ReLU(inplace=True)
)
def forward(self, x):
x = self.squeeze(x)
x = torch.cat([
self.expand_1x1(x),
self.expand_3x3(x)
], 1)
return x
class SqueezeNet(nn.Module):
"""mobile net with simple bypass"""
def __init__(self, class_num=100):
super().__init__()
self.stem = nn.Sequential(
nn.Conv2d(3, 96, 3, padding=1),
nn.BatchNorm2d(96),
nn.ReLU(inplace=True),
nn.MaxPool2d(2, 2)
)
self.fire2 = Fire(96, 128, 16)
self.fire3 = Fire(128, 128, 16)
self.fire4 = Fire(128, 256, 32)
self.fire5 = Fire(256, 256, 32)
self.fire6 = Fire(256, 384, 48)
self.fire7 = Fire(384, 384, 48)
self.fire8 = Fire(384, 512, 64)
self.fire9 = Fire(512, 512, 64)
self.conv10 = nn.Conv2d(512, class_num, 1)
self.avg = nn.AdaptiveAvgPool2d(1)
self.maxpool = nn.MaxPool2d(2, 2)
def forward(self, x):
x = self.stem(x)
f2 = self.fire2(x)
f3 = self.fire3(f2) + f2
f4 = self.fire4(f3)
f4 = self.maxpool(f4)
f5 = self.fire5(f4) + f4
f6 = self.fire6(f5)
f7 = self.fire7(f6) + f6
f8 = self.fire8(f7)
f8 = self.maxpool(f8)
f9 = self.fire9(f8)
c10 = self.conv10(f9)
x = self.avg(c10)
x = x.view(x.size(0), -1)
return x
def squeezenet(class_num=100):
return SqueezeNet(class_num=class_num)
net = squeezenet() print(net) y = net(torch.randn(1, 3, 32, 32)) print(y.size())
SqueezeNet(
(stem): Sequential(
(0): Conv2d(3, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(fire2): Fire(
(squeeze): Sequential(
(0): Conv2d(96, 16, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(16, 64, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(16, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire3): Fire(
(squeeze): Sequential(
(0): Conv2d(128, 16, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(16, 64, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(16, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire4): Fire(
(squeeze): Sequential(
(0): Conv2d(128, 32, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(32, 128, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(32, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire5): Fire(
(squeeze): Sequential(
(0): Conv2d(256, 32, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(32, 128, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(32, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire6): Fire(
(squeeze): Sequential(
(0): Conv2d(256, 48, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(48, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire7): Fire(
(squeeze): Sequential(
(0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(48, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire8): Fire(
(squeeze): Sequential(
(0): Conv2d(384, 64, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(64, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(fire9): Fire(
(squeeze): Sequential(
(0): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_1x1): Sequential(
(0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(expand_3x3): Sequential(
(0): Conv2d(64, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
)
(conv10): Conv2d(512, 100, kernel_size=(1, 1), stride=(1, 1))
(avg): AdaptiveAvgPool2d(output_size=1)
(maxpool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
torch.Size([1, 100])
from torchsummary import summary
summary(squeezenet().to('cuda'), (3, 32, 32))
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Conv2d-1 [-1, 96, 32, 32] 2,688
BatchNorm2d-2 [-1, 96, 32, 32] 192
ReLU-3 [-1, 96, 32, 32] 0
MaxPool2d-4 [-1, 96, 16, 16] 0
Conv2d-5 [-1, 16, 16, 16] 1,552
BatchNorm2d-6 [-1, 16, 16, 16] 32
ReLU-7 [-1, 16, 16, 16] 0
Conv2d-8 [-1, 64, 16, 16] 1,088
BatchNorm2d-9 [-1, 64, 16, 16] 128
ReLU-10 [-1, 64, 16, 16] 0
Conv2d-11 [-1, 64, 16, 16] 9,280
BatchNorm2d-12 [-1, 64, 16, 16] 128
ReLU-13 [-1, 64, 16, 16] 0
Fire-14 [-1, 128, 16, 16] 0
Conv2d-15 [-1, 16, 16, 16] 2,064
BatchNorm2d-16 [-1, 16, 16, 16] 32
ReLU-17 [-1, 16, 16, 16] 0
Conv2d-18 [-1, 64, 16, 16] 1,088
BatchNorm2d-19 [-1, 64, 16, 16] 128
ReLU-20 [-1, 64, 16, 16] 0
Conv2d-21 [-1, 64, 16, 16] 9,280
BatchNorm2d-22 [-1, 64, 16, 16] 128
ReLU-23 [-1, 64, 16, 16] 0
Fire-24 [-1, 128, 16, 16] 0
Conv2d-25 [-1, 32, 16, 16] 4,128
BatchNorm2d-26 [-1, 32, 16, 16] 64
ReLU-27 [-1, 32, 16, 16] 0
Conv2d-28 [-1, 128, 16, 16] 4,224
BatchNorm2d-29 [-1, 128, 16, 16] 256
ReLU-30 [-1, 128, 16, 16] 0
Conv2d-31 [-1, 128, 16, 16] 36,992
BatchNorm2d-32 [-1, 128, 16, 16] 256
ReLU-33 [-1, 128, 16, 16] 0
Fire-34 [-1, 256, 16, 16] 0
MaxPool2d-35 [-1, 256, 8, 8] 0
Conv2d-36 [-1, 32, 8, 8] 8,224
BatchNorm2d-37 [-1, 32, 8, 8] 64
ReLU-38 [-1, 32, 8, 8] 0
Conv2d-39 [-1, 128, 8, 8] 4,224
BatchNorm2d-40 [-1, 128, 8, 8] 256
ReLU-41 [-1, 128, 8, 8] 0
Conv2d-42 [-1, 128, 8, 8] 36,992
BatchNorm2d-43 [-1, 128, 8, 8] 256
ReLU-44 [-1, 128, 8, 8] 0
Fire-45 [-1, 256, 8, 8] 0
Conv2d-46 [-1, 48, 8, 8] 12,336
BatchNorm2d-47 [-1, 48, 8, 8] 96
ReLU-48 [-1, 48, 8, 8] 0
Conv2d-49 [-1, 192, 8, 8] 9,408
BatchNorm2d-50 [-1, 192, 8, 8] 384
ReLU-51 [-1, 192, 8, 8] 0
Conv2d-52 [-1, 192, 8, 8] 83,136
BatchNorm2d-53 [-1, 192, 8, 8] 384
ReLU-54 [-1, 192, 8, 8] 0
Fire-55 [-1, 384, 8, 8] 0
Conv2d-56 [-1, 48, 8, 8] 18,480
BatchNorm2d-57 [-1, 48, 8, 8] 96
ReLU-58 [-1, 48, 8, 8] 0
Conv2d-59 [-1, 192, 8, 8] 9,408
BatchNorm2d-60 [-1, 192, 8, 8] 384
ReLU-61 [-1, 192, 8, 8] 0
Conv2d-62 [-1, 192, 8, 8] 83,136
BatchNorm2d-63 [-1, 192, 8, 8] 384
ReLU-64 [-1, 192, 8, 8] 0
Fire-65 [-1, 384, 8, 8] 0
Conv2d-66 [-1, 64, 8, 8] 24,640
BatchNorm2d-67 [-1, 64, 8, 8] 128
ReLU-68 [-1, 64, 8, 8] 0
Conv2d-69 [-1, 256, 8, 8] 16,640
BatchNorm2d-70 [-1, 256, 8, 8] 512
ReLU-71 [-1, 256, 8, 8] 0
Conv2d-72 [-1, 256, 8, 8] 147,712
BatchNorm2d-73 [-1, 256, 8, 8] 512
ReLU-74 [-1, 256, 8, 8] 0
Fire-75 [-1, 512, 8, 8] 0
MaxPool2d-76 [-1, 512, 4, 4] 0
Conv2d-77 [-1, 64, 4, 4] 32,832
BatchNorm2d-78 [-1, 64, 4, 4] 128
ReLU-79 [-1, 64, 4, 4] 0
Conv2d-80 [-1, 256, 4, 4] 16,640
BatchNorm2d-81 [-1, 256, 4, 4] 512
ReLU-82 [-1, 256, 4, 4] 0
Conv2d-83 [-1, 256, 4, 4] 147,712
BatchNorm2d-84 [-1, 256, 4, 4] 512
ReLU-85 [-1, 256, 4, 4] 0
Fire-86 [-1, 512, 4, 4] 0
Conv2d-87 [-1, 100, 4, 4] 51,300
AdaptiveAvgPool2d-88 [-1, 100, 1, 1] 0
================================================================
Total params: 781,156
Trainable params: 781,156
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.01
Forward/backward pass size (MB): 10.57
Params size (MB): 2.98
Estimated Total Size (MB): 13.56
----------------------------------------------------------------
Lightning モジュール
import torch import torch.nn as nn import torch.nn.functional as F from torch.optim.lr_scheduler import OneCycleLR, CyclicLR, ExponentialLR, CosineAnnealingLR, ReduceLROnPlateau from torch.optim.swa_utils import AveragedModel, update_bn import torchvision import pytorch_lightning as pl from pytorch_lightning.callbacks import LearningRateMonitor, GPUStatsMonitor, EarlyStopping from pytorch_lightning.metrics.functional import accuracy #from pl_bolts.datamodules import CIFAR10DataModule #from pl_bolts.transforms.dataset_normalizations import cifar10_normalization
pl.seed_everything(7);
batch_size = 50
train_transforms = torchvision.transforms.Compose([
torchvision.transforms.RandomCrop(32, padding=4),
torchvision.transforms.RandomHorizontalFlip(),
torchvision.transforms.ToTensor(),
cifar100_normalization(),
])
test_transforms = torchvision.transforms.Compose([
torchvision.transforms.ToTensor(),
cifar100_normalization(),
])
cifar100_dm = CIFAR100DataModule(
batch_size=batch_size,
num_workers=8,
train_transforms=train_transforms,
test_transforms=test_transforms,
val_transforms=test_transforms,
)
class LitCifar100(pl.LightningModule):
def __init__(self, lr=0.05, factor=0.8):
super().__init__()
self.save_hyperparameters()
self.model = squeezenet()
def forward(self, x):
out = self.model(x)
return F.log_softmax(out, dim=1)
def training_step(self, batch, batch_idx):
x, y = batch
logits = F.log_softmax(self.model(x), dim=1)
loss = F.nll_loss(logits, y)
self.log('train_loss', loss)
return loss
def evaluate(self, batch, stage=None):
x, y = batch
logits = self(x)
loss = F.nll_loss(logits, y)
preds = torch.argmax(logits, dim=1)
acc = accuracy(preds, y)
if stage:
self.log(f'{stage}_loss', loss, prog_bar=True)
self.log(f'{stage}_acc', acc, prog_bar=True)
def validation_step(self, batch, batch_idx):
self.evaluate(batch, 'val')
def test_step(self, batch, batch_idx):
self.evaluate(batch, 'test')
def configure_optimizers(self):
optimizer = torch.optim.SGD(self.parameters(), lr=self.hparams.lr, momentum=0.9, weight_decay=5e-4)
return {
'optimizer': optimizer,
'lr_scheduler': ReduceLROnPlateau(optimizer, 'max', patience=5, factor=self.hparams.factor, verbose=True, threshold=0.0001, threshold_mode='abs', cooldown=1, min_lr=1e-5),
'monitor': 'val_acc'
}
訓練 / 評価
%%time
model = LitCifar100(lr=0.05, factor=0.5)
model.datamodule = cifar100_dm
trainer = pl.Trainer(
gpus=1,
max_epochs=100,
progress_bar_refresh_rate=100,
logger=pl.loggers.TensorBoardLogger('tblogs/', name='squeezenet'),
callbacks=[LearningRateMonitor(logging_interval='step')],
)
trainer.fit(model, cifar100_dm)
trainer.test(model, datamodule=cifar100_dm);
| Name | Type | Params
-------------------------------------
0 | model | SqueezeNet | 781 K
-------------------------------------
781 K Trainable params
0 Non-trainable params
781 K Total params
3.125 Total estimated model params size (MB)
(...)
Epoch 23: reducing learning rate of group 0 to 2.5000e-02.
Epoch 34: reducing learning rate of group 0 to 1.2500e-02.
Epoch 42: reducing learning rate of group 0 to 6.2500e-03.
Epoch 49: reducing learning rate of group 0 to 3.1250e-03.
Epoch 59: reducing learning rate of group 0 to 1.5625e-03.
Epoch 72: reducing learning rate of group 0 to 7.8125e-04.
Epoch 84: reducing learning rate of group 0 to 3.9063e-04.
Epoch 99: reducing learning rate of group 0 to 1.9531e-04.
(...)
--------------------------------------------------------------------------------
DATALOADER:0 TEST RESULTS
{'test_acc': 0.6442000269889832, 'test_loss': 1.5194581747055054}
--------------------------------------------------------------------------------
CPU times: user 53min 49s, sys: 1min 42s, total: 55min 31s
Wall time: 1h 1min 44s
以上