Skip to content

sacred config for faster rcnn #1358

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 25 commits into from
Jul 8, 2020
Merged

sacred config for faster rcnn #1358

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
25 commits
Select commit Hold shift + click to select a range
da5c626
move rcnn forward backward task to model zoo
Jerryzcn May 6, 2020
f71a984
revert https://github.com/dmlc/gluon-cv/pull/1249
Jerryzcn May 6, 2020
79ac948
fix
Jerryzcn May 6, 2020
f05fd36
fix
Jerryzcn May 6, 2020
3c2cac8
docstring
Jerryzcn May 6, 2020
d97c070
fix style
Jerryzcn May 6, 2020
8207c93
add docs
Jerryzcn May 6, 2020
51b3172
faster rcnn estimator
Jerryzcn Jun 9, 2020
adc3ad4
Merge remote-tracking branch 'upstream/master' into pr-rcnn
Jerryzcn Jun 9, 2020
5ba0fee
refactor
Jerryzcn Jun 10, 2020
c5ea693
move dataset to init
Jerryzcn Jun 16, 2020
dc25e53
lint
Jerryzcn Jun 16, 2020
3bc9ca2
Merge remote-tracking branch 'refs/remotes/upstream/master' into pr-rcnn
Jerryzcn Jun 16, 2020
72e0550
Merge remote-tracking branch 'upstream/estimator' into pr-rcnn
Jerryzcn Jun 16, 2020
e73985a
merge
Jerryzcn Jun 16, 2020
f3aaaba
disable sacred config for now
Jerryzcn Jun 17, 2020
988ce54
Merge branch 'estimator' of https://github.com/dmlc/gluon-cv into pr-…
Jerryzcn Jun 17, 2020
f56bd79
logger fix
Jerryzcn Jun 17, 2020
d46f892
fix fit
Jerryzcn Jun 17, 2020
304ce5b
autogluon integration
Jerryzcn Jun 30, 2020
df66311
Merge remote-tracking branch 'upstream/estimator' into pr-rcnn
Jerryzcn Jun 30, 2020
1c35190
fix small bug. training working
Jerryzcn Jun 30, 2020
c85b928
lint
Jerryzcn Jun 30, 2020
3bc987f
sacred config for faster rcnn
Jerryzcn Jul 8, 2020
b2cd33c
Merge remote-tracking branch 'upstream/estimator' into pr-rcnn
Jerryzcn Jul 8, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
fix small bug. training working
  • Loading branch information
@Jerryzcn
Jerryzcn committed Jun 30, 2020
commit 1c351906abc8d80c493b0956f8fcffd80e083e6a
2 changes: 0 additions & 2 deletions gluoncv/pipelines/estimators/rcnn/faster_rcnn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -250,7 +250,6 @@ def __init__(self, config, logger=None, reporter=None):
self.net.collect_params('.*batchnorm.*').setattr('dtype', 'float32')
self.net.collect_params('.*normalizedperclassboxcenterencoder.*').setattr('dtype',
'float32')
self._cfg.save_prefix += net_name
if self._cfg.resume.strip():
self.net.load_parameters(self._cfg.resume.strip())
else:
Expand Down Expand Up @@ -471,7 +470,6 @@ def evaluate(self, dataset):
dataloader = _get_testloader(self.net, dataset, len(self.ctx), self._cfg)
return self._validate(dataloader, self.ctx, self.eval_metric)


def predict(self, x):
"""Predict an individual example.

Expand Down
1 change: 0 additions & 1 deletion gluoncv/pipelines/tasks/object_detection.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,6 @@ def _train_object_detection(args, reporter):
vars(args).update(kwargs)
else:
raise NotImplementedError(args.meta_arch, 'is not implemented.')
args.save_prefix += net_name

if args.meta_arch == 'faster_rcnn':
estimator = FasterRCNNEstimator(args, reporter=reporter)
Expand Down
4 changes: 2 additions & 2 deletions scripts/detection/autodetection/train_autodetection.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -160,14 +160,14 @@ def get_dataset(args):
'lr_decay_epoch': ag.Categorical('24,28', '35', '50,55', '40', '45', '55',
'30, 35', '20'),
'warmup_iters': ag.Int(5, 500),
'wd': ag.Categorical(1e-4, 5e-4, 2.5e-4), 'syncbn': True,
'wd': ag.Categorical(1e-4, 5e-4, 2.5e-4), 'syncbn': ag.Bool(),
'label_smooth': False, 'time_limits': time_limits, 'dist_ip_addrs': []}
else:
raise NotImplementedError('%s is not implemented.', args.meta_arch)
default_args = {'dataset': 'voc', 'net': Categorical('mobilenet1.0'), 'meta_arch': 'yolo3',
'lr': Categorical(5e-4, 1e-4), 'loss': gluon.loss.SoftmaxCrossEntropyLoss(),
'split_ratio': 0.8, 'batch_size': 16, 'epochs': 50, 'num_trials': 2,
'nthreads_per_trial': 12, 'num_workers': 32, 'ngpus_per_trial': 1,
'nthreads_per_trial': 12, 'num_workers': 16, 'ngpus_per_trial': 1,
'hybridize': True, 'search_strategy': 'random', 'search_options': {},
'time_limits': None, 'verbose': False, 'transfer': 'coco', 'resume': '',
'checkpoint': 'checkpoint/exp1.ag', 'visualizer': 'none', 'dist_ip_addrs': [],
Expand Down