Add an ipython notebook demo (#2553)

* Fix missing `absolute_import`

* Fix cycle import

* Fix missing export

* Remove some extra blank line and whitespaces

* Rename `create_feeds` -> `create_feed`

* Fix minor issues in tool scripts

* Add ipython notebook demo

* Tweak visualization bbox label style

PaddleCV/object_detection/configs/mask_rcnn_r50_1x.yml

@@ -20,7 +20,6 @@ MaskRCNN:
   mask_assigner: MaskAssigner
   mask_head: MaskHead
 
-
 ResNet:
   norm_type: affine_channel
   norm_decay: true

PaddleCV/object_detection/configs/mask_rcnn_se154_vd_fpn_s1x.yml

@@ -4,11 +4,11 @@ eval_feed: MaskRCNNEvalFeed
 test_feed: MaskRCNNTestFeed
 max_iters: 260000
 snapshot_iter: 10000
-use_gpu: True
+use_gpu: true
 log_smooth_window: 20
 save_dir: output
 pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SE154_vd_pretrained.tar
-weights: output/mask_rcnn_se154_vd_fpn_s1x/model_final/ 
+weights: output/mask_rcnn_se154_vd_fpn_s1x/model_final/
 metric: COCO
 
 MaskRCNN:
@@ -26,7 +26,7 @@ SENet:
   group_width: 4
   groups: 64
   norm_type: affine_channel
-  variant: d 
+  variant: d
 
 FPN:
   max_level: 6
@@ -124,7 +124,7 @@ MaskRCNNTrainFeed:
   - !PadBatch
     pad_to_stride: 32
   dataset:
-    dataset_dir: data/coco 
+    dataset_dir: data/coco
     image_dir: train2017
     annotation: annotations/instances_train2017.json
   num_workers: 2

PaddleCV/object_detection/ppdet/core/__init__.py

@@ -14,4 +14,4 @@
 
 import ppdet.modeling
 import ppdet.optimizer
-import ppdet.data.data_feed
+import ppdet.data

PaddleCV/object_detection/ppdet/data/__init__.py

@@ -33,6 +33,8 @@
 #    * 'MappedDataset' accept a 'xxxSource' and a list of 'xxxOperator'
 #       to build a transformed 'Dataset'
 
+from __future__ import absolute_import
+
 from .dataset import Dataset
 from .reader import Reader
 from .data_feed import create_reader

PaddleCV/object_detection/ppdet/modeling/__init__.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import absolute_import
+
 # XXX for triggering decorators
 from . import anchor_heads
 from . import architectures

PaddleCV/object_detection/ppdet/modeling/anchor_heads/__init__.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import absolute_import
+
 from . import rpn_head
 from . import yolo_head
 from . import retina_head

PaddleCV/object_detection/ppdet/modeling/anchor_heads/rpn_head.py

@@ -22,9 +22,10 @@ from paddle.fluid.initializer import Normal
 from paddle.fluid.regularizer import L2Decay
 
 from ppdet.core.workspace import register
-from ppdet.modeling.ops import AnchorGenerator, RPNTargetAssign, GenerateProposals
+from ppdet.modeling.ops import (AnchorGenerator,
+                                RPNTargetAssign, GenerateProposals)
 
-__all__ = ['RPNTargetAssign', 'GenerateProposals', 'RPNHead']
+__all__ = ['RPNTargetAssign', 'GenerateProposals', 'RPNHead', 'FPNRPNHead']
 
 
 @register

PaddleCV/object_detection/ppdet/modeling/architectures/__init__.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import absolute_import
+
 from . import faster_rcnn
 from . import mask_rcnn
 from . import cascade_rcnn

PaddleCV/object_detection/ppdet/modeling/backbones/__init__.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import absolute_import
+
 from . import resnet
 from . import resnext
 from . import darknet

PaddleCV/object_detection/ppdet/modeling/model_input.py

@@ -20,7 +20,7 @@ from collections import OrderedDict
 
 from paddle import fluid
 
-__all__ = ['create_feeds']
+__all__ = ['create_feed']
 
 # yapf: disable
 feed_var_def = [
@@ -37,7 +37,7 @@ feed_var_def = [
 # yapf: enable
 
 
-def create_feeds(feed, use_pyreader=True):
+def create_feed(feed, use_pyreader=True):
     image_shape = feed.image_shape
     feed_var_map = {var['name']: var for var in feed_var_def}
     feed_var_map['image'] = {

PaddleCV/object_detection/ppdet/modeling/roi_extractors/__init__.py

@@ -12,5 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import absolute_import
+
 from . import roi_extractor
 from .roi_extractor import *

PaddleCV/object_detection/ppdet/modeling/roi_heads/__init__.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import absolute_import
+
 from . import bbox_head
 from . import mask_head
 from . import cascade_head

PaddleCV/object_detection/ppdet/modeling/roi_heads/bbox_head.py

@@ -26,7 +26,7 @@ from paddle.fluid.regularizer import L2Decay
 from ppdet.modeling.ops import MultiClassNMS
 from ppdet.core.workspace import register, serializable
 
-__all__ = ['BBoxHead']
+__all__ = ['BBoxHead', 'TwoFCHead']
 
 
 @register

PaddleCV/object_detection/ppdet/modeling/tests/test_architectures.py

@@ -23,7 +23,7 @@ import paddle.fluid as fluid
 
 from ppdet.modeling.tests.decorator_helper import prog_scope
 from ppdet.core.workspace import load_config, merge_config, create
-from ppdet.modeling.model_input import create_feeds
+from ppdet.modeling.model_input import create_feed
 
 
 class TestFasterRCNN(unittest.TestCase):
@@ -39,14 +39,14 @@ class TestFasterRCNN(unittest.TestCase):
     def test_train(self):
         train_feed = create(self.cfg['train_feed'])
         model = create(self.detector_type)
-        _, feed_vars = create_feeds(train_feed)
+        _, feed_vars = create_feed(train_feed)
         train_fetches = model.train(feed_vars)
 
     @prog_scope()
     def test_test(self):
         test_feed = create(self.cfg['eval_feed'])
         model = create(self.detector_type)
-        _, feed_vars = create_feeds(test_feed)
+        _, feed_vars = create_feed(test_feed)
         test_fetches = model.eval(feed_vars)
 
 

PaddleCV/object_detection/ppdet/utils/visualizer.py

@@ -17,7 +17,6 @@ from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
 
-import logging
 import numpy as np
 import pycocotools.mask as mask_util
 from PIL import Image, ImageDraw
@@ -26,8 +25,6 @@ from .colormap import colormap
 
 __all__ = ['visualize_results']
 
-logger = logging.getLogger(__name__)
-
 
 def visualize_results(image,
                       im_id,
@@ -73,7 +70,7 @@ def draw_mask(image, im_id, segms, threshold, alpha=0.7):
     return Image.fromarray(img_array.astype('uint8'))
 
 
-def draw_bbox(image, im_id, catid2name, bboxes, threshold, 
+def draw_bbox(image, im_id, catid2name, bboxes, threshold,
               is_bbox_normalized=False):
     """
     Draw bbox on image
@@ -103,8 +100,11 @@ def draw_bbox(image, im_id, catid2name, bboxes, threshold,
             width=2,
             fill='red')
         if image.mode == 'RGB':
-            draw.text((xmin, ymin), catid2name[catid], (255, 255, 0))
-        logger.debug("\t {:15s} at {:25} score: {:.5f}".format(catid2name[catid],
-                    str(list(map(int, list([xmin, ymin, xmax, ymax])))), score))
+            text = catid2name[catid]
+            tw, th = draw.textsize(text)
+            draw.rectangle([(xmin + 1, ymin + 1),
+                            (xmin + tw + 1, ymin + th + 1)],
+                           fill='red')
+            draw.text((xmin + 1, ymin + 1), text, fill=(255, 255, 255))
 
     return image

PaddleCV/object_detection/tools/eval.py

@@ -24,7 +24,7 @@ import paddle.fluid as fluid
 from ppdet.utils.eval_utils import parse_fetches, eval_run, eval_results
 import ppdet.utils.checkpoint as checkpoint
 from ppdet.utils.cli import ArgsParser
-from ppdet.modeling.model_input import create_feeds
+from ppdet.modeling.model_input import create_feed
 from ppdet.data.data_feed import create_reader
 from ppdet.core.workspace import load_config, merge_config, create
 
@@ -68,7 +68,7 @@ def main():
     eval_prog = fluid.Program()
     with fluid.program_guard(eval_prog, startup_prog):
         with fluid.unique_name.guard():
-            pyreader, feed_vars = create_feeds(eval_feed)
+            pyreader, feed_vars = create_feed(eval_feed)
             fetches = model.eval(feed_vars)
     eval_prog = eval_prog.clone(True)
 
@@ -100,7 +100,7 @@ def main():
     results = eval_run(exe, compile_program, pyreader, keys, values, cls)
     # Evaluation
     eval_results(results, eval_feed, cfg.metric,
-                 cfg.MaskHead.resolution, FLAGS.output_file)
+                 model.mask_head.resolution, FLAGS.output_file)
 
 
 if __name__ == '__main__':

PaddleCV/object_detection/tools/infer.py

@@ -25,7 +25,7 @@ from PIL import Image
 from paddle import fluid
 
 from ppdet.core.workspace import load_config, merge_config, create
-from ppdet.modeling.model_input import create_feeds
+from ppdet.modeling.model_input import create_feed
 from ppdet.data.data_feed import create_reader
 
 from ppdet.utils.eval_utils import parse_fetches
@@ -55,7 +55,7 @@ def get_test_images(infer_dir, infer_img):
     Get image path list in TEST mode
     """
     assert infer_img is not None or infer_dir is not None, \
-        "--infer-img or --infer-dir should be set"
+        "--infer_img or --infer_dir should be set"
     images = []
 
     # infer_img has a higher priority
@@ -104,7 +104,7 @@ def main():
     infer_prog = fluid.Program()
     with fluid.program_guard(infer_prog, startup_prog):
         with fluid.unique_name.guard():
-            _, feed_vars = create_feeds(test_feed, use_pyreader=False)
+            _, feed_vars = create_feed(test_feed, use_pyreader=False)
             test_fetches = model.test(feed_vars)
     infer_prog = infer_prog.clone(True)
 
@@ -151,11 +151,11 @@ def main():
         mask_results = None
         is_bbox_normalized = True if cfg.metric == 'VOC' else False
         if 'bbox' in res:
-            bbox_results = bbox2out([res], clsid2catid, 
+            bbox_results = bbox2out([res], clsid2catid,
                                     is_bbox_normalized)
         if 'mask' in res:
             mask_results = mask2out([res], clsid2catid,
-                                    cfg.MaskHead['resolution'])
+                                    model.mask_head.resolution)
 
         # visualize result
         im_ids = res['im_id'][0]

PaddleCV/object_detection/tools/train.py

@@ -31,7 +31,7 @@ from ppdet.utils.eval_utils import parse_fetches, eval_run, eval_results
 from ppdet.utils.stats import TrainingStats
 from ppdet.utils.cli import ArgsParser
 import ppdet.utils.checkpoint as checkpoint
-from ppdet.modeling.model_input import create_feeds
+from ppdet.modeling.model_input import create_feed
 
 import logging
 FORMAT = '%(asctime)s-%(levelname)s: %(message)s'
@@ -77,7 +77,7 @@ def main():
     train_prog = fluid.Program()
     with fluid.program_guard(train_prog, startup_prog):
         with fluid.unique_name.guard():
-            train_pyreader, feed_vars = create_feeds(train_feed)
+            train_pyreader, feed_vars = create_feed(train_feed)
             train_fetches = model.train(feed_vars)
             loss = train_fetches['loss']
             lr = lr_builder()
@@ -95,7 +95,7 @@ def main():
         eval_prog = fluid.Program()
         with fluid.program_guard(eval_prog, startup_prog):
             with fluid.unique_name.guard():
-                eval_pyreader, feed_vars = create_feeds(eval_feed)
+                eval_pyreader, feed_vars = create_feed(eval_feed)
                 fetches = model.eval(feed_vars)
         eval_prog = eval_prog.clone(True)
 
@@ -156,7 +156,7 @@ def main():
                                    eval_keys, eval_values, eval_cls)
                 # Evaluation
                 eval_results(results, eval_feed, cfg.metric,
-                             cfg.MaskHead.resolution, FLAGS.output_file)
+                             model.mask_head.resolution, FLAGS.output_file)
 
     checkpoint.save(exe, train_prog, os.path.join(save_dir, "model_final"))
     train_pyreader.reset()