update modelzoo and fix configs (#2789)

* fix resnet for resnet34

* fix ResNet18 and ResNet34

* update modelzoo and fix configs

* fix configs

* fix resnet bn

configs/faster_rcnn_x101_vd_64x4d_fpn_1x.yml

@@ -4,7 +4,7 @@ eval_feed: FasterRCNNEvalFeed
 test_feed: FasterRCNNTestFeed
 max_iters: 180000
 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/ResNeXt101_vd_64x4d_pretrained.tar
@@ -118,7 +118,7 @@ FasterRCNNTrainFeed:
   - !PadBatch
     pad_to_stride: 32
   num_workers: 2
-  shuffle: True
+  shuffle: true
 
 FasterRCNNEvalFeed:
   batch_size: 1
@@ -139,4 +139,4 @@ FasterRCNNTestFeed:
   - !PadBatch
     pad_to_stride: 32
   num_workers: 2
-  shuffle: False
+  shuffle: false

configs/faster_rcnn_x101_vd_64x4d_fpn_2x.yml

+architecture: FasterRCNN
+train_feed: FasterRCNNTrainFeed
+eval_feed: FasterRCNNEvalFeed
+test_feed: FasterRCNNTestFeed
+max_iters: 360000
+snapshot_iter: 10000
+use_gpu: true
+log_smooth_window: 20
+save_dir: output
+pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNeXt101_vd_64x4d_pretrained.tar
+weights: output/faster_rcnn_x101_vd_64x4d_fpn_1x/model_final
+metric: COCO
+
+FasterRCNN:
+  backbone: ResNeXt
+  fpn: FPN
+  rpn_head: FPNRPNHead
+  roi_extractor: FPNRoIAlign
+  bbox_head: BBoxHead
+  bbox_assigner: BBoxAssigner
+
+ResNeXt:
+  depth: 101
+  feature_maps: [2, 3, 4, 5]
+  freeze_at: 2
+  group_width: 4
+  groups: 64
+  norm_type: affine_channel
+  variant: d
+
+FPN:
+  max_level: 6
+  min_level: 2
+  num_chan: 256
+  spatial_scale: [0.03125, 0.0625, 0.125, 0.25]
+
+FPNRPNHead:
+  anchor_generator:
+    anchor_sizes: [32, 64, 128, 256, 512]
+    aspect_ratios: [0.5, 1.0, 2.0]
+    stride: [16.0, 16.0]
+    variance: [1.0, 1.0, 1.0, 1.0]
+  anchor_start_size: 32
+  max_level: 6
+  min_level: 2
+  num_chan: 256
+  rpn_target_assign:
+    rpn_batch_size_per_im: 256
+    rpn_fg_fraction: 0.5
+    rpn_negative_overlap: 0.3
+    rpn_positive_overlap: 0.7
+    rpn_straddle_thresh: 0.0
+  train_proposal:
+    min_size: 0.0
+    nms_thresh: 0.7
+    post_nms_top_n: 2000
+    pre_nms_top_n: 2000
+  test_proposal:
+    min_size: 0.0
+    nms_thresh: 0.7
+    post_nms_top_n: 1000
+    pre_nms_top_n: 1000
+
+FPNRoIAlign:
+  canconical_level: 4
+  canonical_size: 224
+  max_level: 5
+  min_level: 2
+  box_resolution: 7
+  sampling_ratio: 2
+
+BBoxAssigner:
+  batch_size_per_im: 512
+  bbox_reg_weights: [0.1, 0.1, 0.2, 0.2]
+  bg_thresh_hi: 0.5
+  bg_thresh_lo: 0.0
+  fg_fraction: 0.25
+  fg_thresh: 0.5
+  num_classes: 81
+
+BBoxHead:
+  head: TwoFCHead
+  nms:
+    keep_top_k: 100
+    nms_threshold: 0.5
+    score_threshold: 0.05
+  num_classes: 81
+
+TwoFCHead:
+  num_chan: 1024
+
+LearningRate:
+  base_lr: 0.01
+  schedulers:
+  - !PiecewiseDecay
+    gamma: 0.1
+    milestones: [240000, 320000]
+  - !LinearWarmup
+    start_factor: 0.1
+    steps: 1000
+
+OptimizerBuilder:
+  optimizer:
+    momentum: 0.9
+    type: Momentum
+  regularizer:
+    factor: 0.0001
+    type: L2
+
+FasterRCNNTrainFeed:
+  # batch size per device
+  batch_size: 1
+  dataset:
+    dataset_dir: dataset/coco
+    image_dir: train2017
+    annotation: annotations/instances_train2017.json
+  batch_transforms:
+  - !PadBatch
+    pad_to_stride: 32
+  num_workers: 2
+  shuffle: true
+
+FasterRCNNEvalFeed:
+  batch_size: 1
+  dataset:
+    dataset_dir: dataset/coco
+    annotation: annotations/instances_val2017.json
+    image_dir: val2017
+  batch_transforms:
+  - !PadBatch
+    pad_to_stride: 32
+  num_workers: 2
+
+FasterRCNNTestFeed:
+  batch_size: 1
+  dataset:
+    annotation: annotations/instances_val2017.json
+  batch_transforms:
+  - !PadBatch
+    pad_to_stride: 32
+  num_workers: 2
+  shuffle: false

docs/MODEL_ZOO.md

@@ -49,8 +49,9 @@ The backbone models pretrained on ImageNet are available. All backbone models ar
 | ResNet101-FPN        | Faster         |    1    |   2x    |  39.1  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_r101_fpn_2x.tar) |
 | ResNet101-FPN        | Mask           |    1    |   1x    |  39.5  |  35.2   | [model](https://paddlemodels.bj.bcebos.com/object_detection/mask_rcnn_r101_fpn_1x.tar) |
 | ResNet101-vd-FPN     | Faster         |    1    |   1x    |  40.5  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_r101_vd_fpn_1x.tar) |
-| ResNet101-vd-FPN     | Faster         |    1    |   2x    |  40.6  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_r101_fpn_2x.tar) |
+| ResNet101-vd-FPN     | Faster         |    1    |   2x    |  40.8  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_r101_vd_fpn_2x.tar) |
 | ResNeXt101-vd-FPN    | Faster         |    1    |   1x    |  42.2  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_x101_vd_64x4d_fpn_1x.tar) |
+| ResNeXt101-vd-FPN    | Faster         |    1    |   2x    |  41.7  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_x101_vd_64x4d_fpn_2x.tar) |
 | SENet154-vd-FPN      | Faster         |    1    |  1.44x  |  42.9  |    -    | [model](https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_se154_vd_fpn_s1x.tar) |
 | SENet154-vd-FPN      | Mask           |    1    |  1.44x  |  44.0  |  38.7   | [model](https://paddlemodels.bj.bcebos.com/object_detection/mask_rcnn_se154_vd_fpn_s1x.tar) |
 

ppdet/modeling/backbones/resnet.py

@@ -119,6 +119,7 @@ class ResNet(object):
             regularizer=L2Decay(norm_decay))
 
         if self.norm_type in ['bn', 'sync_bn']:
+            global_stats = True if self.freeze_norm else False
             out = fluid.layers.batch_norm(
                 input=conv,
                 act=act,
@@ -126,7 +127,8 @@ class ResNet(object):
                 param_attr=pattr,
                 bias_attr=battr,
                 moving_mean_name=bn_name + '_mean',
-                moving_variance_name=bn_name + '_variance', )
+                moving_variance_name=bn_name + '_variance',
+                use_global_stats=global_stats)
             scale = fluid.framework._get_var(pattr.name)
             bias = fluid.framework._get_var(battr.name)
         elif self.norm_type == 'affine_channel':

tools/train.py

@@ -141,12 +141,12 @@ def main():
 
     exe.run(startup_prog)
 
-    freeze_bn = getattr(model.backbone, 'freeze_norm', False)
+    fuse_bn = getattr(model.backbone, 'norm_type', None) == 'affine_channel'
     start_iter = 0
     if FLAGS.resume_checkpoint:
         checkpoint.load_checkpoint(exe, train_prog, FLAGS.resume_checkpoint)
         start_iter = checkpoint.global_step()
-    elif cfg.pretrain_weights and freeze_bn:
+    elif cfg.pretrain_weights and fuse_bn:
         checkpoint.load_and_fusebn(exe, train_prog, cfg.pretrain_weights)
     elif cfg.pretrain_weights:
         checkpoint.load_pretrain(exe, train_prog, cfg.pretrain_weights)