Searchable Objects¶
When defining custom Python objects such as network architectures, or specialized optimizers, it may be hard to decide what values to set for all of their attributes. AutoGluon provides an API that allows you to instead specify a search space of possible values to consider for such attributes, within which the optimal value will be automatically searched for at runtime. This tutorial demonstrates how easy this is to do, without having to modify your existing code at all!
Example for Constructing a Network¶
This tutorial covers an example of selecting a neural network’s
architecture as a hyperparameter optimization (HPO) task. If you are
interested in efficient neural architecture search (NAS), please refer
to this other tutorial instead: sec_proxyless_ .
CIFAR ResNet in GluonCV¶
GluonCV provides CIFARResNet, which allow user to specify how many layers at each stage. For example, we can construct a CIFAR ResNet with only 1 layer per stage:
from gluoncv.model_zoo.cifarresnet import CIFARResNetV1, CIFARBasicBlockV1
layers = [1, 1, 1]
channels = [16, 16, 32, 64]
net = CIFARResNetV1(CIFARBasicBlockV1, layers, channels)
We can visualize the network:
import autogluon.core as ag
from autogluon.vision.utils import plot_network
plot_network(net, (1, 3, 32, 32))
Searchable Network Architecture Using AutoGluon Object¶
autogluon.obj() enables customized search space to any user
defined class. It can also be used within autogluon.Categorical() if
you have multiple networks to choose from.
@ag.obj(
nstage1=ag.space.Int(2, 4),
nstage2=ag.space.Int(2, 4),
)
class MyCifarResNet(CIFARResNetV1):
def __init__(self, nstage1, nstage2):
nstage3 = 9 - nstage1 - nstage2
layers = [nstage1, nstage2, nstage3]
channels = [16, 16, 32, 64]
super().__init__(CIFARBasicBlockV1, layers=layers, channels=channels)
Create one network instance and print the configuration space:
mynet=MyCifarResNet()
print(mynet.cs)
Configuration space object:
Hyperparameters:
nstage1, Type: UniformInteger, Range: [2, 4], Default: 3
nstage2, Type: UniformInteger, Range: [2, 4], Default: 3
We can also overwrite existing search spaces:
mynet1 = MyCifarResNet(nstage1=1,
nstage2=ag.space.Int(5, 10))
print(mynet1.cs)
Configuration space object:
Hyperparameters:
nstage2, Type: UniformInteger, Range: [5, 10], Default: 8
Decorate Existing Class¶
We can also use autogluon.obj() to easily decorate any existing
classes. For example, if we want to search learning rate and weight
decay for Adam optimizer, we only need to add a decorator:
from mxnet import optimizer as optim
@ag.obj()
class Adam(optim.Adam):
pass
Then we can create an instance:
myoptim = Adam(learning_rate=ag.Real(1e-2, 1e-1, log=True), wd=ag.Real(1e-5, 1e-3, log=True))
print(myoptim.cs)
Configuration space object:
Hyperparameters:
learning_rate, Type: UniformFloat, Range: [0.01, 0.1], Default: 0.0316227766, on log-scale
wd, Type: UniformFloat, Range: [1e-05, 0.001], Default: 0.0001, on log-scale
Launch Experiments Using AutoGluon Object¶
AutoGluon Object is compatible with Fit API in AutoGluon tasks, and also
works with user-defined training scripts using
autogluon.autogluon_register_args(). We can start fitting:
from autogluon.vision import ImagePredictor
classifier = ImagePredictor().fit('cifar10', hyperparameters={'net': mynet, 'optimizer': myoptim, 'epochs': 1}, ngpus_per_trial=1)
INFO:root:time_limit=auto set to time_limit=7200. INFO:gluoncv.auto.tasks.image_classification:Starting fit without HPO INFO:ImageClassificationEstimator:modified configs(<old> != <new>): { INFO:ImageClassificationEstimator:root.valid.num_workers 4 != 8 INFO:ImageClassificationEstimator:root.valid.batch_size 128 != 16 INFO:ImageClassificationEstimator:root.img_cls.model resnet50_v1 != resnet50_v1b INFO:ImageClassificationEstimator:root.train.data_dir ~/.mxnet/datasets/imagenet != auto INFO:ImageClassificationEstimator:root.train.rec_train_idx ~/.mxnet/datasets/imagenet/rec/train.idx != auto INFO:ImageClassificationEstimator:root.train.epochs 10 != 1 INFO:ImageClassificationEstimator:root.train.batch_size 128 != 16 INFO:ImageClassificationEstimator:root.train.rec_val_idx ~/.mxnet/datasets/imagenet/rec/val.idx != auto INFO:ImageClassificationEstimator:root.train.num_training_samples 1281167 != -1 INFO:ImageClassificationEstimator:root.train.rec_val ~/.mxnet/datasets/imagenet/rec/val.rec != auto INFO:ImageClassificationEstimator:root.train.num_workers 4 != 8 INFO:ImageClassificationEstimator:root.train.rec_train ~/.mxnet/datasets/imagenet/rec/train.rec != auto INFO:ImageClassificationEstimator:root.train.lr 0.1 != 0.01 INFO:ImageClassificationEstimator:} INFO:ImageClassificationEstimator:Saved config to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-course-v3/docs/_build/eval/tutorials/course/63704242/.trial_0/config.yaml INFO:ImageClassificationEstimator:Start training from [Epoch 0] INFO:ImageClassificationEstimator:Epoch[0] Batch [49] Speed: 96.569802 samples/sec accuracy=0.162500 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [99] Speed: 98.105191 samples/sec accuracy=0.243125 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [149] Speed: 97.129931 samples/sec accuracy=0.276667 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [199] Speed: 96.094901 samples/sec accuracy=0.307812 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [249] Speed: 95.498021 samples/sec accuracy=0.332250 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [299] Speed: 94.937852 samples/sec accuracy=0.351042 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [349] Speed: 94.221995 samples/sec accuracy=0.374464 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [399] Speed: 93.480737 samples/sec accuracy=0.389062 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [449] Speed: 92.981490 samples/sec accuracy=0.407083 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [499] Speed: 92.318216 samples/sec accuracy=0.422250 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [549] Speed: 91.584709 samples/sec accuracy=0.434091 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [599] Speed: 91.110570 samples/sec accuracy=0.446146 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [649] Speed: 90.637199 samples/sec accuracy=0.455865 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [699] Speed: 91.030171 samples/sec accuracy=0.466875 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [749] Speed: 90.672365 samples/sec accuracy=0.476583 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [799] Speed: 90.657254 samples/sec accuracy=0.484609 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [849] Speed: 90.520070 samples/sec accuracy=0.492941 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [899] Speed: 90.167553 samples/sec accuracy=0.500139 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [949] Speed: 90.181223 samples/sec accuracy=0.505855 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [999] Speed: 90.728911 samples/sec accuracy=0.512437 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1049] Speed: 91.028524 samples/sec accuracy=0.517024 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1099] Speed: 91.286133 samples/sec accuracy=0.521705 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1149] Speed: 91.391470 samples/sec accuracy=0.526522 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1199] Speed: 91.288013 samples/sec accuracy=0.530365 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1249] Speed: 91.446811 samples/sec accuracy=0.536400 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1299] Speed: 91.548280 samples/sec accuracy=0.540769 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1349] Speed: 91.431980 samples/sec accuracy=0.544861 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1399] Speed: 91.326463 samples/sec accuracy=0.547813 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1449] Speed: 91.459149 samples/sec accuracy=0.551034 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1499] Speed: 91.382980 samples/sec accuracy=0.554042 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1549] Speed: 91.422904 samples/sec accuracy=0.557258 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1599] Speed: 91.481541 samples/sec accuracy=0.560312 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1649] Speed: 91.396478 samples/sec accuracy=0.563674 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1699] Speed: 91.633117 samples/sec accuracy=0.566875 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1749] Speed: 91.519779 samples/sec accuracy=0.569607 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1799] Speed: 91.586604 samples/sec accuracy=0.572500 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1849] Speed: 91.671287 samples/sec accuracy=0.575372 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1899] Speed: 91.519969 samples/sec accuracy=0.578026 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1949] Speed: 91.635454 samples/sec accuracy=0.580513 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [1999] Speed: 91.826577 samples/sec accuracy=0.583375 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2049] Speed: 91.340094 samples/sec accuracy=0.585976 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2099] Speed: 91.797232 samples/sec accuracy=0.588155 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2149] Speed: 91.669867 samples/sec accuracy=0.590291 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2199] Speed: 91.659828 samples/sec accuracy=0.592472 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2249] Speed: 91.685582 samples/sec accuracy=0.594361 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2299] Speed: 91.891807 samples/sec accuracy=0.596440 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2349] Speed: 91.733282 samples/sec accuracy=0.598298 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2399] Speed: 91.840315 samples/sec accuracy=0.600313 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2449] Speed: 91.947290 samples/sec accuracy=0.601862 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2499] Speed: 91.881085 samples/sec accuracy=0.603675 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2549] Speed: 91.829919 samples/sec accuracy=0.605294 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2599] Speed: 91.974973 samples/sec accuracy=0.607139 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2649] Speed: 91.998508 samples/sec accuracy=0.609080 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2699] Speed: 91.947219 samples/sec accuracy=0.610625 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2749] Speed: 92.050090 samples/sec accuracy=0.612477 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2799] Speed: 92.028780 samples/sec accuracy=0.614196 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2849] Speed: 91.926697 samples/sec accuracy=0.615504 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2899] Speed: 92.031882 samples/sec accuracy=0.617284 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2949] Speed: 91.953907 samples/sec accuracy=0.619174 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [2999] Speed: 91.904648 samples/sec accuracy=0.621000 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3049] Speed: 91.831284 samples/sec accuracy=0.622561 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3099] Speed: 91.692254 samples/sec accuracy=0.623972 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3149] Speed: 91.946574 samples/sec accuracy=0.625397 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3199] Speed: 91.985079 samples/sec accuracy=0.626758 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3249] Speed: 91.857502 samples/sec accuracy=0.628135 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3299] Speed: 91.992755 samples/sec accuracy=0.629261 lr=0.010000 INFO:ImageClassificationEstimator:Epoch[0] Batch [3349] Speed: 91.916898 samples/sec accuracy=0.630784 lr=0.010000 INFO:ImageClassificationEstimator:[Epoch 0] training: accuracy=0.631426 INFO:ImageClassificationEstimator:[Epoch 0] speed: 92 samples/sec time cost: 609.103123 INFO:ImageClassificationEstimator:[Epoch 0] validation: top1=0.902000 top5=0.997500 INFO:ImageClassificationEstimator:[Epoch 0] Current best top-1: 0.902000 vs previous 0.000000, saved to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-course-v3/docs/_build/eval/tutorials/course/63704242/.trial_0/best_checkpoint.pkl INFO:ImageClassificationEstimator:Pickled to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-course-v3/docs/_build/eval/tutorials/course/63704242/.trial_0/best_checkpoint.pkl INFO:gluoncv.auto.tasks.image_classification:Finished, total runtime is 617.84 s INFO:gluoncv.auto.tasks.image_classification:{ 'best_config': { 'batch_size': 16, 'custom_net': MyCifarResNet( (features): HybridSequential( (0): Conv2D(None -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): HybridSequential( (0): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (1): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (2): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) ) (3): HybridSequential( (0): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(16 -> 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) (downsample): HybridSequential( (0): Conv2D(16 -> 32, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (1): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (2): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (3): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (4): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (5): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (6): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) (7): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) ) (4): HybridSequential( (0): CIFARBasicBlockV1( (body): HybridSequential( (0): Conv2D(32 -> 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) (2): Activation(relu) (3): Conv2D(64 -> 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) (downsample): HybridSequential( (0): Conv2D(32 -> 64, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None) ) ) ) (5): GlobalAvgPool2D(size=(1, 1), stride=(1, 1), padding=(0, 0), ceil_mode=True, global_pool=True, pool_type=avg, layout=NCHW) ) (output): Dense(64 -> 10, linear) ), 'custom_optimizer': <__main__.Adam object at 0x7f3bebfcd050>, 'dist_ip_addrs': None, 'epochs': 1, 'estimator': <class 'gluoncv.auto.estimators.image_classification.image_classification.ImageClassificationEstimator'>, 'final_fit': False, 'gpus': [0], 'log_dir': '/var/lib/jenkins/workspace/workspace/autogluon-tutorial-course-v3/docs/_build/eval/tutorials/course/63704242', 'lr': 0.01, 'model': 'resnet50_v1b', 'ngpus_per_trial': 1, 'nthreads_per_trial': 128, 'num_trials': 1, 'num_workers': 8, 'scheduler': 'local', 'search_strategy': 'random', 'searcher': 'random', 'seed': 460, 'time_limits': 7200, 'wall_clock_tick': 1615356727.153655}, 'total_time': 600.31250166893, 'train_acc': 0.631425925925926, 'valid_acc': 0.902}
print(classifier.fit_summary())
{'train_acc': 0.631425925925926, 'valid_acc': 0.902, 'total_time': 600.31250166893, 'best_config': {'model': 'resnet50_v1b', 'lr': 0.01, 'num_trials': 1, 'epochs': 1, 'batch_size': 16, 'nthreads_per_trial': 128, 'ngpus_per_trial': 1, 'time_limits': 7200, 'search_strategy': 'random', 'dist_ip_addrs': None, 'log_dir': '/var/lib/jenkins/workspace/workspace/autogluon-tutorial-course-v3/docs/_build/eval/tutorials/course/63704242', 'searcher': 'random', 'scheduler': 'local', 'custom_net': MyCifarResNet(
(features): HybridSequential(
(0): Conv2D(None -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): HybridSequential(
(0): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(1): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(2): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
)
(3): HybridSequential(
(0): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
(downsample): HybridSequential(
(0): Conv2D(16 -> 32, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(1): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(2): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(3): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(4): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(5): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(6): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(7): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
)
(4): HybridSequential(
(0): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(64 -> 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
(downsample): HybridSequential(
(0): Conv2D(32 -> 64, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
)
(5): GlobalAvgPool2D(size=(1, 1), stride=(1, 1), padding=(0, 0), ceil_mode=True, global_pool=True, pool_type=avg, layout=NCHW)
)
(output): Dense(64 -> 10, linear)
), 'custom_optimizer': <__main__.Adam object at 0x7f3bebfcd050>, 'num_workers': 8, 'gpus': [0], 'seed': 460, 'final_fit': False, 'estimator': <class 'gluoncv.auto.estimators.image_classification.image_classification.ImageClassificationEstimator'>, 'wall_clock_tick': 1615356727.153655}, 'fit_history': {'train_acc': 0.631425925925926, 'valid_acc': 0.902, 'total_time': 600.31250166893, 'best_config': {'model': 'resnet50_v1b', 'lr': 0.01, 'num_trials': 1, 'epochs': 1, 'batch_size': 16, 'nthreads_per_trial': 128, 'ngpus_per_trial': 1, 'time_limits': 7200, 'search_strategy': 'random', 'dist_ip_addrs': None, 'log_dir': '/var/lib/jenkins/workspace/workspace/autogluon-tutorial-course-v3/docs/_build/eval/tutorials/course/63704242', 'searcher': 'random', 'scheduler': 'local', 'custom_net': MyCifarResNet(
(features): HybridSequential(
(0): Conv2D(None -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): HybridSequential(
(0): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(1): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(2): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(16 -> 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
)
(3): HybridSequential(
(0): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(16 -> 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
(downsample): HybridSequential(
(0): Conv2D(16 -> 32, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(1): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(2): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(3): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(4): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(5): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(6): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
(7): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
)
(4): HybridSequential(
(0): CIFARBasicBlockV1(
(body): HybridSequential(
(0): Conv2D(32 -> 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
(2): Activation(relu)
(3): Conv2D(64 -> 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
(downsample): HybridSequential(
(0): Conv2D(32 -> 64, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=None)
)
)
)
(5): GlobalAvgPool2D(size=(1, 1), stride=(1, 1), padding=(0, 0), ceil_mode=True, global_pool=True, pool_type=avg, layout=NCHW)
)
(output): Dense(64 -> 10, linear)
), 'custom_optimizer': <__main__.Adam object at 0x7f3bebfcd050>, 'num_workers': 8, 'gpus': [0], 'seed': 460, 'final_fit': False, 'estimator': <class 'gluoncv.auto.estimators.image_classification.image_classification.ImageClassificationEstimator'>, 'wall_clock_tick': 1615356727.153655}}}