Mask_rcnn pruned model inference error

Hi.
I get this error when inferencing the mask_rcnn pruned model:

File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/scripts/inference.py", line 351, in <module>
File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/scripts/inference.py", line 343, in main
File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/scripts/inference.py", line 305, in infer
File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/executer/distributed_executer.py", line 490, in infer
File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/utils/evaluation.py", line 242, in infer
File "/usr/local/lib/python3.6/dist-packages/tensorflow_estimator/python/estimator/estimator.py", line 638, in predict
hooks=all_hooks) as mon_sess:
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1014, in __init__
stop_grace_period_secs=stop_grace_period_secs)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 725, in __init__
self._sess = _RecoverableSession(self._coordinated_creator)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1207, in __init__
_WrappedSession.__init__(self, self._create_session())
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1212, in _create_session
return self._sess_creator.create_session()
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 878, in create_session
self.tf_sess = self._session_creator.create_session()
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 647, in create_session
init_fn=self._scaffold.init_fn)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/session_manager.py", line 290, in prepare_session
config=config)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/session_manager.py", line 204, in _restore_checkpoint
saver.restore(sess, checkpoint_filename_with_path)
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/saver.py", line 1282, in restore
checkpoint_prefix)
ValueError: The passed save_path is not a valid checkpoint: /tmp/tmp70qmt1vn/model.ckpt-90000
Using TensorFlow backend.

Can you share the full command and full log?
How about run inference against the unpruned model?

BTW, you can refer to the official released jupyter notebook.

Full command:

!mask_rcnn inference -i /workspace/tlt/results/not_corrosion \
                             -o /workspace/tlt/results/tlt_mask_rcnn_corrosion1000_resnet50/annotated_images \
                             -e /workspace/tlt/results/tlt_mask_rcnn_corrosion1000_resnet50/pruned_model/model.step-90000.tlt/final_spec.txt \
                             -m /workspace/tlt/results/tlt_mask_rcnn_corrosion1000_resnet50/pruned_model/model.step-90000.tlt/model.tlt \
                             -t 0.6 \
                             -k $KEY \
                             --gpu_index 0 \
                             --include_mask

Full log:

Using TensorFlow backend.
WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
Using TensorFlow backend.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/horovod/tensorflow/__init__.py:117: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/horovod/tensorflow/__init__.py:143: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

Label file does not exist. Skipping...
[MaskRCNN] INFO    : Running inference...
[MaskRCNN] INFO    : Loading weights from /workspace/tlt/results/tlt_mask_rcnn_corrosion1000_resnet50/pruned_model2/model.step-90000.tlt/model.tlt
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/autograph/converters/directives.py:119: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.

WARNING:tensorflow:Entity <function infer.<locals>.infer_input_fn.<locals>.process_path at 0x7f850de6fb70> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <function infer.<locals>.infer_input_fn.<locals>.process_path at 0x7f850de6fb70>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
[MaskRCNN] INFO    : ***********************
[MaskRCNN] INFO    : Loading model graph...
[MaskRCNN] INFO    : ***********************
WARNING:tensorflow:Entity <bound method AnchorLayer.call of <iva.mask_rcnn.layers.anchor_layer.AnchorLayer object at 0x7f83df32a400>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method AnchorLayer.call of <iva.mask_rcnn.layers.anchor_layer.AnchorLayer object at 0x7f83df32a400>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method MultilevelProposal.call of <iva.mask_rcnn.layers.multilevel_proposal_layer.MultilevelProposal object at 0x7f83df32a5f8>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method MultilevelProposal.call of <iva.mask_rcnn.layers.multilevel_proposal_layer.MultilevelProposal object at 0x7f83df32a5f8>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
[MaskRCNN] INFO    : [ROI OPs] Using Batched NMS... Scope: MLP/multilevel_propose_rois/level_2/
[MaskRCNN] INFO    : [ROI OPs] Using Batched NMS... Scope: MLP/multilevel_propose_rois/level_3/
[MaskRCNN] INFO    : [ROI OPs] Using Batched NMS... Scope: MLP/multilevel_propose_rois/level_4/
[MaskRCNN] INFO    : [ROI OPs] Using Batched NMS... Scope: MLP/multilevel_propose_rois/level_5/
[MaskRCNN] INFO    : [ROI OPs] Using Batched NMS... Scope: MLP/multilevel_propose_rois/level_6/
WARNING:tensorflow:Entity <bound method MultilevelCropResize.call of <iva.mask_rcnn.layers.multilevel_crop_resize_layer.MultilevelCropResize object at 0x7f83df32a7f0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method MultilevelCropResize.call of <iva.mask_rcnn.layers.multilevel_crop_resize_layer.MultilevelCropResize object at 0x7f83df32a7f0>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df32ac88>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df32ac88>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df3339e8>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df3339e8>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df333b00>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df333b00>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method GPUDetections.call of <iva.mask_rcnn.layers.gpu_detection_layer.GPUDetections object at 0x7f83df333c18>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method GPUDetections.call of <iva.mask_rcnn.layers.gpu_detection_layer.GPUDetections object at 0x7f83df333c18>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method MultilevelCropResize.call of <iva.mask_rcnn.layers.multilevel_crop_resize_layer.MultilevelCropResize object at 0x7f83df333f28>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method MultilevelCropResize.call of <iva.mask_rcnn.layers.multilevel_crop_resize_layer.MultilevelCropResize object at 0x7f83df333f28>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df33d080>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method ReshapeLayer.call of <iva.mask_rcnn.layers.reshape_layer.ReshapeLayer object at 0x7f83df33d080>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING:tensorflow:Entity <bound method MaskPostprocess.call of <iva.mask_rcnn.layers.mask_postprocess_layer.MaskPostprocess object at 0x7f83df344d30>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method MaskPostprocess.call of <iva.mask_rcnn.layers.mask_postprocess_layer.MaskPostprocess object at 0x7f83df344d30>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
Model: "model"
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to                     
==================================================================================================
image_input (ImageInput)        [(2, 3, 640, 640)]   0                                            
__________________________________________________________________________________________________
conv1 (Conv2D)                  (2, 64, 320, 320)    9408        image_input[0][0]                
__________________________________________________________________________________________________
bn_conv1 (BatchNormalization)   (2, 64, 320, 320)    256         conv1[0][0]                      
__________________________________________________________________________________________________
activation (Activation)         (2, 64, 320, 320)    0           bn_conv1[0][0]                   
__________________________________________________________________________________________________
max_pooling2d (MaxPooling2D)    (2, 64, 160, 160)    0           activation[0][0]                 
__________________________________________________________________________________________________
block_1a_conv_1 (Conv2D)        (2, 64, 160, 160)    4096        max_pooling2d[0][0]              
__________________________________________________________________________________________________
block_1a_bn_1 (BatchNormalizati (2, 64, 160, 160)    256         block_1a_conv_1[0][0]            
__________________________________________________________________________________________________
block_1a_relu_1 (Activation)    (2, 64, 160, 160)    0           block_1a_bn_1[0][0]              
__________________________________________________________________________________________________
block_1a_conv_2 (Conv2D)        (2, 64, 160, 160)    36864       block_1a_relu_1[0][0]            
__________________________________________________________________________________________________
block_1a_bn_2 (BatchNormalizati (2, 64, 160, 160)    256         block_1a_conv_2[0][0]            
__________________________________________________________________________________________________
block_1a_relu_2 (Activation)    (2, 64, 160, 160)    0           block_1a_bn_2[0][0]              
__________________________________________________________________________________________________
block_1a_conv_3 (Conv2D)        (2, 256, 160, 160)   16384       block_1a_relu_2[0][0]            
__________________________________________________________________________________________________
block_1a_conv_shortcut (Conv2D) (2, 256, 160, 160)   16384       max_pooling2d[0][0]              
__________________________________________________________________________________________________
block_1a_bn_3 (BatchNormalizati (2, 256, 160, 160)   1024        block_1a_conv_3[0][0]            
__________________________________________________________________________________________________
block_1a_bn_shortcut (BatchNorm (2, 256, 160, 160)   1024        block_1a_conv_shortcut[0][0]     
__________________________________________________________________________________________________
add (Add)                       (2, 256, 160, 160)   0           block_1a_bn_3[0][0]              
                                                                 block_1a_bn_shortcut[0][0]       
__________________________________________________________________________________________________
block_1a_relu (Activation)      (2, 256, 160, 160)   0           add[0][0]                        
__________________________________________________________________________________________________
block_1b_conv_1 (Conv2D)        (2, 64, 160, 160)    16384       block_1a_relu[0][0]              
__________________________________________________________________________________________________
block_1b_bn_1 (BatchNormalizati (2, 64, 160, 160)    256         block_1b_conv_1[0][0]            
__________________________________________________________________________________________________
block_1b_relu_1 (Activation)    (2, 64, 160, 160)    0           block_1b_bn_1[0][0]              
__________________________________________________________________________________________________
block_1b_conv_2 (Conv2D)        (2, 64, 160, 160)    36864       block_1b_relu_1[0][0]            
__________________________________________________________________________________________________
block_1b_bn_2 (BatchNormalizati (2, 64, 160, 160)    256         block_1b_conv_2[0][0]            
__________________________________________________________________________________________________
block_1b_relu_2 (Activation)    (2, 64, 160, 160)    0           block_1b_bn_2[0][0]              
__________________________________________________________________________________________________
block_1b_conv_3 (Conv2D)        (2, 256, 160, 160)   16384       block_1b_relu_2[0][0]            
__________________________________________________________________________________________________
block_1b_bn_3 (BatchNormalizati (2, 256, 160, 160)   1024        block_1b_conv_3[0][0]            
__________________________________________________________________________________________________
add_1 (Add)                     (2, 256, 160, 160)   0           block_1b_bn_3[0][0]              
                                                                 block_1a_relu[0][0]              
__________________________________________________________________________________________________
block_1b_relu (Activation)      (2, 256, 160, 160)   0           add_1[0][0]                      
__________________________________________________________________________________________________
block_1c_conv_1 (Conv2D)        (2, 64, 160, 160)    16384       block_1b_relu[0][0]              
__________________________________________________________________________________________________
block_1c_bn_1 (BatchNormalizati (2, 64, 160, 160)    256         block_1c_conv_1[0][0]            
__________________________________________________________________________________________________
block_1c_relu_1 (Activation)    (2, 64, 160, 160)    0           block_1c_bn_1[0][0]              
__________________________________________________________________________________________________
block_1c_conv_2 (Conv2D)        (2, 64, 160, 160)    36864       block_1c_relu_1[0][0]            
__________________________________________________________________________________________________
block_1c_bn_2 (BatchNormalizati (2, 64, 160, 160)    256         block_1c_conv_2[0][0]            
__________________________________________________________________________________________________
block_1c_relu_2 (Activation)    (2, 64, 160, 160)    0           block_1c_bn_2[0][0]              
__________________________________________________________________________________________________
block_1c_conv_3 (Conv2D)        (2, 256, 160, 160)   16384       block_1c_relu_2[0][0]            
__________________________________________________________________________________________________
block_1c_bn_3 (BatchNormalizati (2, 256, 160, 160)   1024        block_1c_conv_3[0][0]            
__________________________________________________________________________________________________
add_2 (Add)                     (2, 256, 160, 160)   0           block_1c_bn_3[0][0]              
                                                                 block_1b_relu[0][0]              
__________________________________________________________________________________________________
block_1c_relu (Activation)      (2, 256, 160, 160)   0           add_2[0][0]                      
__________________________________________________________________________________________________
block_2a_conv_1 (Conv2D)        (2, 128, 80, 80)     32768       block_1c_relu[0][0]              
__________________________________________________________________________________________________
block_2a_bn_1 (BatchNormalizati (2, 128, 80, 80)     512         block_2a_conv_1[0][0]            
__________________________________________________________________________________________________
block_2a_relu_1 (Activation)    (2, 128, 80, 80)     0           block_2a_bn_1[0][0]              
__________________________________________________________________________________________________
block_2a_conv_2 (Conv2D)        (2, 128, 80, 80)     147456      block_2a_relu_1[0][0]            
__________________________________________________________________________________________________
block_2a_bn_2 (BatchNormalizati (2, 128, 80, 80)     512         block_2a_conv_2[0][0]            
__________________________________________________________________________________________________
block_2a_relu_2 (Activation)    (2, 128, 80, 80)     0           block_2a_bn_2[0][0]              
__________________________________________________________________________________________________
block_2a_conv_3 (Conv2D)        (2, 512, 80, 80)     65536       block_2a_relu_2[0][0]            
__________________________________________________________________________________________________
block_2a_conv_shortcut (Conv2D) (2, 512, 80, 80)     131072      block_1c_relu[0][0]              
__________________________________________________________________________________________________
block_2a_bn_3 (BatchNormalizati (2, 512, 80, 80)     2048        block_2a_conv_3[0][0]            
__________________________________________________________________________________________________
block_2a_bn_shortcut (BatchNorm (2, 512, 80, 80)     2048        block_2a_conv_shortcut[0][0]     
__________________________________________________________________________________________________
add_3 (Add)                     (2, 512, 80, 80)     0           block_2a_bn_3[0][0]              
                                                                 block_2a_bn_shortcut[0][0]       
__________________________________________________________________________________________________
block_2a_relu (Activation)      (2, 512, 80, 80)     0           add_3[0][0]                      
__________________________________________________________________________________________________
block_2b_conv_1 (Conv2D)        (2, 128, 80, 80)     65536       block_2a_relu[0][0]              
__________________________________________________________________________________________________
block_2b_bn_1 (BatchNormalizati (2, 128, 80, 80)     512         block_2b_conv_1[0][0]            
__________________________________________________________________________________________________
block_2b_relu_1 (Activation)    (2, 128, 80, 80)     0           block_2b_bn_1[0][0]              
__________________________________________________________________________________________________
block_2b_conv_2 (Conv2D)        (2, 128, 80, 80)     147456      block_2b_relu_1[0][0]            
__________________________________________________________________________________________________
block_2b_bn_2 (BatchNormalizati (2, 128, 80, 80)     512         block_2b_conv_2[0][0]            
__________________________________________________________________________________________________
block_2b_relu_2 (Activation)    (2, 128, 80, 80)     0           block_2b_bn_2[0][0]              
__________________________________________________________________________________________________
block_2b_conv_3 (Conv2D)        (2, 512, 80, 80)     65536       block_2b_relu_2[0][0]            
__________________________________________________________________________________________________
block_2b_bn_3 (BatchNormalizati (2, 512, 80, 80)     2048        block_2b_conv_3[0][0]            
__________________________________________________________________________________________________
add_4 (Add)                     (2, 512, 80, 80)     0           block_2b_bn_3[0][0]              
                                                                 block_2a_relu[0][0]              
__________________________________________________________________________________________________
block_2b_relu (Activation)      (2, 512, 80, 80)     0           add_4[0][0]                      
__________________________________________________________________________________________________
block_2c_conv_1 (Conv2D)        (2, 128, 80, 80)     65536       block_2b_relu[0][0]              
__________________________________________________________________________________________________
block_2c_bn_1 (BatchNormalizati (2, 128, 80, 80)     512         block_2c_conv_1[0][0]            
__________________________________________________________________________________________________
block_2c_relu_1 (Activation)    (2, 128, 80, 80)     0           block_2c_bn_1[0][0]              
__________________________________________________________________________________________________
block_2c_conv_2 (Conv2D)        (2, 128, 80, 80)     147456      block_2c_relu_1[0][0]            
__________________________________________________________________________________________________
block_2c_bn_2 (BatchNormalizati (2, 128, 80, 80)     512         block_2c_conv_2[0][0]            
__________________________________________________________________________________________________
block_2c_relu_2 (Activation)    (2, 128, 80, 80)     0           block_2c_bn_2[0][0]              
__________________________________________________________________________________________________
block_2c_conv_3 (Conv2D)        (2, 512, 80, 80)     65536       block_2c_relu_2[0][0]            
__________________________________________________________________________________________________
block_2c_bn_3 (BatchNormalizati (2, 512, 80, 80)     2048        block_2c_conv_3[0][0]            
__________________________________________________________________________________________________
add_5 (Add)                     (2, 512, 80, 80)     0           block_2c_bn_3[0][0]              
                                                                 block_2b_relu[0][0]              
__________________________________________________________________________________________________
block_2c_relu (Activation)      (2, 512, 80, 80)     0           add_5[0][0]                      
__________________________________________________________________________________________________
block_2d_conv_1 (Conv2D)        (2, 128, 80, 80)     65536       block_2c_relu[0][0]              
__________________________________________________________________________________________________
block_2d_bn_1 (BatchNormalizati (2, 128, 80, 80)     512         block_2d_conv_1[0][0]            
__________________________________________________________________________________________________
block_2d_relu_1 (Activation)    (2, 128, 80, 80)     0           block_2d_bn_1[0][0]              
__________________________________________________________________________________________________
block_2d_conv_2 (Conv2D)        (2, 128, 80, 80)     147456      block_2d_relu_1[0][0]            
__________________________________________________________________________________________________
block_2d_bn_2 (BatchNormalizati (2, 128, 80, 80)     512         block_2d_conv_2[0][0]            
__________________________________________________________________________________________________
block_2d_relu_2 (Activation)    (2, 128, 80, 80)     0           block_2d_bn_2[0][0]              
__________________________________________________________________________________________________
block_2d_conv_3 (Conv2D)        (2, 512, 80, 80)     65536       block_2d_relu_2[0][0]            
__________________________________________________________________________________________________
block_2d_bn_3 (BatchNormalizati (2, 512, 80, 80)     2048        block_2d_conv_3[0][0]            
__________________________________________________________________________________________________
add_6 (Add)                     (2, 512, 80, 80)     0           block_2d_bn_3[0][0]              
                                                                 block_2c_relu[0][0]              
__________________________________________________________________________________________________
block_2d_relu (Activation)      (2, 512, 80, 80)     0           add_6[0][0]                      
__________________________________________________________________________________________________
block_3a_conv_1 (Conv2D)        (2, 256, 40, 40)     131072      block_2d_relu[0][0]              
__________________________________________________________________________________________________
block_3a_bn_1 (BatchNormalizati (2, 256, 40, 40)     1024        block_3a_conv_1[0][0]            
__________________________________________________________________________________________________
block_3a_relu_1 (Activation)    (2, 256, 40, 40)     0           block_3a_bn_1[0][0]              
__________________________________________________________________________________________________
block_3a_conv_2 (Conv2D)        (2, 256, 40, 40)     589824      block_3a_relu_1[0][0]            
__________________________________________________________________________________________________
block_3a_bn_2 (BatchNormalizati (2, 256, 40, 40)     1024        block_3a_conv_2[0][0]            
__________________________________________________________________________________________________
block_3a_relu_2 (Activation)    (2, 256, 40, 40)     0           block_3a_bn_2[0][0]              
__________________________________________________________________________________________________
block_3a_conv_3 (Conv2D)        (2, 1024, 40, 40)    262144      block_3a_relu_2[0][0]            
__________________________________________________________________________________________________
block_3a_conv_shortcut (Conv2D) (2, 1024, 40, 40)    524288      block_2d_relu[0][0]              
__________________________________________________________________________________________________
block_3a_bn_3 (BatchNormalizati (2, 1024, 40, 40)    4096        block_3a_conv_3[0][0]            
__________________________________________________________________________________________________
block_3a_bn_shortcut (BatchNorm (2, 1024, 40, 40)    4096        block_3a_conv_shortcut[0][0]     
__________________________________________________________________________________________________
add_7 (Add)                     (2, 1024, 40, 40)    0           block_3a_bn_3[0][0]              
                                                                 block_3a_bn_shortcut[0][0]       
__________________________________________________________________________________________________
block_3a_relu (Activation)      (2, 1024, 40, 40)    0           add_7[0][0]                      
__________________________________________________________________________________________________
block_3b_conv_1 (Conv2D)        (2, 256, 40, 40)     262144      block_3a_relu[0][0]              
__________________________________________________________________________________________________
block_3b_bn_1 (BatchNormalizati (2, 256, 40, 40)     1024        block_3b_conv_1[0][0]            
__________________________________________________________________________________________________
block_3b_relu_1 (Activation)    (2, 256, 40, 40)     0           block_3b_bn_1[0][0]              
__________________________________________________________________________________________________
block_3b_conv_2 (Conv2D)        (2, 256, 40, 40)     589824      block_3b_relu_1[0][0]            
__________________________________________________________________________________________________
block_3b_bn_2 (BatchNormalizati (2, 256, 40, 40)     1024        block_3b_conv_2[0][0]            
__________________________________________________________________________________________________
block_3b_relu_2 (Activation)    (2, 256, 40, 40)     0           block_3b_bn_2[0][0]              
__________________________________________________________________________________________________
block_3b_conv_3 (Conv2D)        (2, 1024, 40, 40)    262144      block_3b_relu_2[0][0]            
__________________________________________________________________________________________________
block_3b_bn_3 (BatchNormalizati (2, 1024, 40, 40)    4096        block_3b_conv_3[0][0]            
__________________________________________________________________________________________________
add_8 (Add)                     (2, 1024, 40, 40)    0           block_3b_bn_3[0][0]              
                                                                 block_3a_relu[0][0]              
__________________________________________________________________________________________________
block_3b_relu (Activation)      (2, 1024, 40, 40)    0           add_8[0][0]                      
__________________________________________________________________________________________________
block_3c_conv_1 (Conv2D)        (2, 256, 40, 40)     262144      block_3b_relu[0][0]              
__________________________________________________________________________________________________
block_3c_bn_1 (BatchNormalizati (2, 256, 40, 40)     1024        block_3c_conv_1[0][0]            
__________________________________________________________________________________________________
block_3c_relu_1 (Activation)    (2, 256, 40, 40)     0           block_3c_bn_1[0][0]              
__________________________________________________________________________________________________
block_3c_conv_2 (Conv2D)        (2, 256, 40, 40)     589824      block_3c_relu_1[0][0]            
__________________________________________________________________________________________________
block_3c_bn_2 (BatchNormalizati (2, 256, 40, 40)     1024        block_3c_conv_2[0][0]            
__________________________________________________________________________________________________
block_3c_relu_2 (Activation)    (2, 256, 40, 40)     0           block_3c_bn_2[0][0]              
__________________________________________________________________________________________________
block_3c_conv_3 (Conv2D)        (2, 1024, 40, 40)    262144      block_3c_relu_2[0][0]            
__________________________________________________________________________________________________
block_3c_bn_3 (BatchNormalizati (2, 1024, 40, 40)    4096        block_3c_conv_3[0][0]            
__________________________________________________________________________________________________
add_9 (Add)                     (2, 1024, 40, 40)    0           block_3c_bn_3[0][0]              
                                                                 block_3b_relu[0][0]              
__________________________________________________________________________________________________
block_3c_relu (Activation)      (2, 1024, 40, 40)    0           add_9[0][0]                      
__________________________________________________________________________________________________
block_3d_conv_1 (Conv2D)        (2, 256, 40, 40)     262144      block_3c_relu[0][0]              
__________________________________________________________________________________________________
block_3d_bn_1 (BatchNormalizati (2, 256, 40, 40)     1024        block_3d_conv_1[0][0]            
__________________________________________________________________________________________________
block_3d_relu_1 (Activation)    (2, 256, 40, 40)     0           block_3d_bn_1[0][0]              
__________________________________________________________________________________________________
block_3d_conv_2 (Conv2D)        (2, 256, 40, 40)     589824      block_3d_relu_1[0][0]            
__________________________________________________________________________________________________
block_3d_bn_2 (BatchNormalizati (2, 256, 40, 40)     1024        block_3d_conv_2[0][0]            
__________________________________________________________________________________________________
block_3d_relu_2 (Activation)    (2, 256, 40, 40)     0           block_3d_bn_2[0][0]              
__________________________________________________________________________________________________
block_3d_conv_3 (Conv2D)        (2, 1024, 40, 40)    262144      block_3d_relu_2[0][0]            
__________________________________________________________________________________________________
block_3d_bn_3 (BatchNormalizati (2, 1024, 40, 40)    4096        block_3d_conv_3[0][0]            
__________________________________________________________________________________________________
add_10 (Add)                    (2, 1024, 40, 40)    0           block_3d_bn_3[0][0]              
                                                                 block_3c_relu[0][0]              
__________________________________________________________________________________________________
block_3d_relu (Activation)      (2, 1024, 40, 40)    0           add_10[0][0]                     
__________________________________________________________________________________________________
block_3e_conv_1 (Conv2D)        (2, 256, 40, 40)     262144      block_3d_relu[0][0]              
__________________________________________________________________________________________________
block_3e_bn_1 (BatchNormalizati (2, 256, 40, 40)     1024        block_3e_conv_1[0][0]            
__________________________________________________________________________________________________
block_3e_relu_1 (Activation)    (2, 256, 40, 40)     0           block_3e_bn_1[0][0]              
__________________________________________________________________________________________________
block_3e_conv_2 (Conv2D)        (2, 256, 40, 40)     589824      block_3e_relu_1[0][0]            
__________________________________________________________________________________________________
block_3e_bn_2 (BatchNormalizati (2, 256, 40, 40)     1024        block_3e_conv_2[0][0]            
__________________________________________________________________________________________________
block_3e_relu_2 (Activation)    (2, 256, 40, 40)     0           block_3e_bn_2[0][0]              
__________________________________________________________________________________________________
block_3e_conv_3 (Conv2D)        (2, 1024, 40, 40)    262144      block_3e_relu_2[0][0]            
__________________________________________________________________________________________________
block_3e_bn_3 (BatchNormalizati (2, 1024, 40, 40)    4096        block_3e_conv_3[0][0]            
__________________________________________________________________________________________________
add_11 (Add)                    (2, 1024, 40, 40)    0           block_3e_bn_3[0][0]              
                                                                 block_3d_relu[0][0]              
__________________________________________________________________________________________________
block_3e_relu (Activation)      (2, 1024, 40, 40)    0           add_11[0][0]                     
__________________________________________________________________________________________________
block_3f_conv_1 (Conv2D)        (2, 256, 40, 40)     262144      block_3e_relu[0][0]              
__________________________________________________________________________________________________
block_3f_bn_1 (BatchNormalizati (2, 256, 40, 40)     1024        block_3f_conv_1[0][0]            
__________________________________________________________________________________________________
block_3f_relu_1 (Activation)    (2, 256, 40, 40)     0           block_3f_bn_1[0][0]              
__________________________________________________________________________________________________
block_3f_conv_2 (Conv2D)        (2, 256, 40, 40)     589824      block_3f_relu_1[0][0]            
__________________________________________________________________________________________________
block_3f_bn_2 (BatchNormalizati (2, 256, 40, 40)     1024        block_3f_conv_2[0][0]            
__________________________________________________________________________________________________
block_3f_relu_2 (Activation)    (2, 256, 40, 40)     0           block_3f_bn_2[0][0]              
__________________________________________________________________________________________________
block_3f_conv_3 (Conv2D)        (2, 1024, 40, 40)    262144      block_3f_relu_2[0][0]            
__________________________________________________________________________________________________
block_3f_bn_3 (BatchNormalizati (2, 1024, 40, 40)    4096        block_3f_conv_3[0][0]            
__________________________________________________________________________________________________
add_12 (Add)                    (2, 1024, 40, 40)    0           block_3f_bn_3[0][0]              
                                                                 block_3e_relu[0][0]              
__________________________________________________________________________________________________
block_3f_relu (Activation)      (2, 1024, 40, 40)    0           add_12[0][0]                     
__________________________________________________________________________________________________
block_4a_conv_1 (Conv2D)        (2, 512, 20, 20)     524288      block_3f_relu[0][0]              
__________________________________________________________________________________________________
block_4a_bn_1 (BatchNormalizati (2, 512, 20, 20)     2048        block_4a_conv_1[0][0]            
__________________________________________________________________________________________________
block_4a_relu_1 (Activation)    (2, 512, 20, 20)     0           block_4a_bn_1[0][0]              
__________________________________________________________________________________________________
block_4a_conv_2 (Conv2D)        (2, 512, 20, 20)     2359296     block_4a_relu_1[0][0]            
__________________________________________________________________________________________________
block_4a_bn_2 (BatchNormalizati (2, 512, 20, 20)     2048        block_4a_conv_2[0][0]            
__________________________________________________________________________________________________
block_4a_relu_2 (Activation)    (2, 512, 20, 20)     0           block_4a_bn_2[0][0]              
__________________________________________________________________________________________________
block_4a_conv_3 (Conv2D)        (2, 2048, 20, 20)    1048576     block_4a_relu_2[0][0]            
__________________________________________________________________________________________________
block_4a_conv_shortcut (Conv2D) (2, 2048, 20, 20)    2097152     block_3f_relu[0][0]              
__________________________________________________________________________________________________
block_4a_bn_3 (BatchNormalizati (2, 2048, 20, 20)    8192        block_4a_conv_3[0][0]            
__________________________________________________________________________________________________
block_4a_bn_shortcut (BatchNorm (2, 2048, 20, 20)    8192        block_4a_conv_shortcut[0][0]     
__________________________________________________________________________________________________
add_13 (Add)                    (2, 2048, 20, 20)    0           block_4a_bn_3[0][0]              
                                                                 block_4a_bn_shortcut[0][0]       
__________________________________________________________________________________________________
block_4a_relu (Activation)      (2, 2048, 20, 20)    0           add_13[0][0]                     
__________________________________________________________________________________________________
block_4b_conv_1 (Conv2D)        (2, 512, 20, 20)     1048576     block_4a_relu[0][0]              
__________________________________________________________________________________________________
block_4b_bn_1 (BatchNormalizati (2, 512, 20, 20)     2048        block_4b_conv_1[0][0]            
__________________________________________________________________________________________________
block_4b_relu_1 (Activation)    (2, 512, 20, 20)     0           block_4b_bn_1[0][0]              
__________________________________________________________________________________________________
block_4b_conv_2 (Conv2D)        (2, 512, 20, 20)     2359296     block_4b_relu_1[0][0]            
__________________________________________________________________________________________________
block_4b_bn_2 (BatchNormalizati (2, 512, 20, 20)     2048        block_4b_conv_2[0][0]            
__________________________________________________________________________________________________
block_4b_relu_2 (Activation)    (2, 512, 20, 20)     0           block_4b_bn_2[0][0]              
__________________________________________________________________________________________________
block_4b_conv_3 (Conv2D)        (2, 2048, 20, 20)    1048576     block_4b_relu_2[0][0]            
__________________________________________________________________________________________________
block_4b_bn_3 (BatchNormalizati (2, 2048, 20, 20)    8192        block_4b_conv_3[0][0]            
__________________________________________________________________________________________________
add_14 (Add)                    (2, 2048, 20, 20)    0           block_4b_bn_3[0][0]              
                                                                 block_4a_relu[0][0]              
__________________________________________________________________________________________________
block_4b_relu (Activation)      (2, 2048, 20, 20)    0           add_14[0][0]                     
__________________________________________________________________________________________________
block_4c_conv_1 (Conv2D)        (2, 512, 20, 20)     1048576     block_4b_relu[0][0]              
__________________________________________________________________________________________________
block_4c_bn_1 (BatchNormalizati (2, 512, 20, 20)     2048        block_4c_conv_1[0][0]            
__________________________________________________________________________________________________
block_4c_relu_1 (Activation)    (2, 512, 20, 20)     0           block_4c_bn_1[0][0]              
__________________________________________________________________________________________________
block_4c_conv_2 (Conv2D)        (2, 512, 20, 20)     2359296     block_4c_relu_1[0][0]            
__________________________________________________________________________________________________
block_4c_bn_2 (BatchNormalizati (2, 512, 20, 20)     2048        block_4c_conv_2[0][0]            
__________________________________________________________________________________________________
block_4c_relu_2 (Activation)    (2, 512, 20, 20)     0           block_4c_bn_2[0][0]              
__________________________________________________________________________________________________
block_4c_conv_3 (Conv2D)        (2, 2048, 20, 20)    1048576     block_4c_relu_2[0][0]            
__________________________________________________________________________________________________
block_4c_bn_3 (BatchNormalizati (2, 2048, 20, 20)    8192        block_4c_conv_3[0][0]            
__________________________________________________________________________________________________
add_15 (Add)                    (2, 2048, 20, 20)    0           block_4c_bn_3[0][0]              
                                                                 block_4b_relu[0][0]              
__________________________________________________________________________________________________
block_4c_relu (Activation)      (2, 2048, 20, 20)    0           add_15[0][0]                     
__________________________________________________________________________________________________
l5 (Conv2D)                     (2, 256, 20, 20)     524544      block_4c_relu[0][0]              
__________________________________________________________________________________________________
l4 (Conv2D)                     (2, 256, 40, 40)     262400      block_3f_relu[0][0]              
__________________________________________________________________________________________________
FPN_up_4 (UpSampling2D)         (2, 256, 40, 40)     0           l5[0][0]                         
__________________________________________________________________________________________________
FPN_add_4 (Add)                 (2, 256, 40, 40)     0           l4[0][0]                         
                                                                 FPN_up_4[0][0]                   
__________________________________________________________________________________________________
l3 (Conv2D)                     (2, 256, 80, 80)     131328      block_2d_relu[0][0]              
__________________________________________________________________________________________________
FPN_up_3 (UpSampling2D)         (2, 256, 80, 80)     0           FPN_add_4[0][0]                  
__________________________________________________________________________________________________
FPN_add_3 (Add)                 (2, 256, 80, 80)     0           l3[0][0]                         
                                                                 FPN_up_3[0][0]                   
__________________________________________________________________________________________________
l2 (Conv2D)                     (2, 256, 160, 160)   65792       block_1c_relu[0][0]              
__________________________________________________________________________________________________
FPN_up_2 (UpSampling2D)         (2, 256, 160, 160)   0           FPN_add_3[0][0]                  
__________________________________________________________________________________________________
FPN_add_2 (Add)                 (2, 256, 160, 160)   0           l2[0][0]                         
                                                                 FPN_up_2[0][0]                   
__________________________________________________________________________________________________
post_hoc_d5 (Conv2D)            (2, 256, 20, 20)     590080      l5[0][0]                         
__________________________________________________________________________________________________
post_hoc_d2 (Conv2D)            (2, 256, 160, 160)   590080      FPN_add_2[0][0]                  
__________________________________________________________________________________________________
post_hoc_d3 (Conv2D)            (2, 256, 80, 80)     590080      FPN_add_3[0][0]                  
__________________________________________________________________________________________________
post_hoc_d4 (Conv2D)            (2, 256, 40, 40)     590080      FPN_add_4[0][0]                  
__________________________________________________________________________________________________
p6 (MaxPooling2D)               (2, 256, 10, 10)     0           post_hoc_d5[0][0]                
__________________________________________________________________________________________________
rpn (Conv2D)                    multiple             590080      post_hoc_d2[0][0]                
                                                                 post_hoc_d3[0][0]                
                                                                 post_hoc_d4[0][0]                
                                                                 post_hoc_d5[0][0]                
                                                                 p6[0][0]                         
__________________________________________________________________________________________________
rpn-class (Conv2D)              multiple             771         rpn[0][0]                        
                                                                 rpn[1][0]                        
                                                                 rpn[2][0]                        
                                                                 rpn[3][0]                        
                                                                 rpn[4][0]                        
__________________________________________________________________________________________________
rpn-box (Conv2D)                multiple             3084        rpn[0][0]                        
                                                                 rpn[1][0]                        
                                                                 rpn[2][0]                        
                                                                 rpn[3][0]                        
                                                                 rpn[4][0]                        
__________________________________________________________________________________________________
permute (Permute)               (2, 160, 160, 3)     0           rpn-class[0][0]                  
__________________________________________________________________________________________________
permute_2 (Permute)             (2, 80, 80, 3)       0           rpn-class[1][0]                  
__________________________________________________________________________________________________
permute_4 (Permute)             (2, 40, 40, 3)       0           rpn-class[2][0]                  
__________________________________________________________________________________________________
permute_6 (Permute)             (2, 20, 20, 3)       0           rpn-class[3][0]                  
__________________________________________________________________________________________________
permute_8 (Permute)             (2, 10, 10, 3)       0           rpn-class[4][0]                  
__________________________________________________________________________________________________
permute_1 (Permute)             (2, 160, 160, 12)    0           rpn-box[0][0]                    
__________________________________________________________________________________________________
permute_3 (Permute)             (2, 80, 80, 12)      0           rpn-box[1][0]                    
__________________________________________________________________________________________________
permute_5 (Permute)             (2, 40, 40, 12)      0           rpn-box[2][0]                    
__________________________________________________________________________________________________
permute_7 (Permute)             (2, 20, 20, 12)      0           rpn-box[3][0]                    
__________________________________________________________________________________________________
permute_9 (Permute)             (2, 10, 10, 12)      0           rpn-box[4][0]                    
__________________________________________________________________________________________________
anchor_layer (AnchorLayer)      OrderedDict([(2, (16 0           image_input[0][0]                
__________________________________________________________________________________________________
info_input (InfoInput)          [(2, 5)]             0                                            
__________________________________________________________________________________________________
MLP (MultilevelProposal)        ((2, 1000), (2, 1000 0           permute[0][0]                    
                                                                 permute_2[0][0]                  
                                                                 permute_4[0][0]                  
                                                                 permute_6[0][0]                  
                                                                 permute_8[0][0]                  
                                                                 permute_1[0][0]                  
                                                                 permute_3[0][0]                  
                                                                 permute_5[0][0]                  
                                                                 permute_7[0][0]                  
                                                                 permute_9[0][0]                  
                                                                 anchor_layer[0][0]               
                                                                 anchor_layer[0][1]               
                                                                 anchor_layer[0][2]               
                                                                 anchor_layer[0][3]               
                                                                 anchor_layer[0][4]               
                                                                 info_input[0][0]                 
__________________________________________________________________________________________________
multilevel_crop_resize (Multile (2, 1000, 256, 7, 7) 0           post_hoc_d2[0][0]                
                                                                 post_hoc_d3[0][0]                
                                                                 post_hoc_d4[0][0]                
                                                                 post_hoc_d5[0][0]                
                                                                 p6[0][0]                         
                                                                 MLP[0][1]                        
__________________________________________________________________________________________________
box_head_reshape1 (ReshapeLayer (2000, 12544)        0           multilevel_crop_resize[0][0]     
__________________________________________________________________________________________________
fc6 (Dense)                     (2000, 1024)         12846080    box_head_reshape1[0][0]          
__________________________________________________________________________________________________
fc7 (Dense)                     (2000, 1024)         1049600     fc6[0][0]                        
__________________________________________________________________________________________________
class-predict (Dense)           (2000, 2)            2050        fc7[0][0]                        
__________________________________________________________________________________________________
box-predict (Dense)             (2000, 8)            8200        fc7[0][0]                        
__________________________________________________________________________________________________
box_head_reshape2 (ReshapeLayer (2, 1000, 2)         0           class-predict[0][0]              
__________________________________________________________________________________________________
box_head_reshape3 (ReshapeLayer (2, 1000, 8)         0           box-predict[0][0]                
__________________________________________________________________________________________________
gpu_detections (GPUDetections)  ((2,), (2, 100, 4),  0           box_head_reshape2[0][0]          
                                                                 box_head_reshape3[0][0]          
                                                                 MLP[0][1]                        
                                                                 info_input[0][0]                 
__________________________________________________________________________________________________
multilevel_crop_resize_1 (Multi (2, 100, 256, 14, 14 0           post_hoc_d2[0][0]                
                                                                 post_hoc_d3[0][0]                
                                                                 post_hoc_d4[0][0]                
                                                                 post_hoc_d5[0][0]                
                                                                 p6[0][0]                         
                                                                 gpu_detections[0][1]             
__________________________________________________________________________________________________
mask_head_reshape_1 (ReshapeLay (200, 256, 14, 14)   0           multilevel_crop_resize_1[0][0]   
__________________________________________________________________________________________________
mask-conv-l0 (Conv2D)           (200, 256, 14, 14)   590080      mask_head_reshape_1[0][0]        
__________________________________________________________________________________________________
mask-conv-l1 (Conv2D)           (200, 256, 14, 14)   590080      mask-conv-l0[0][0]               
__________________________________________________________________________________________________
mask-conv-l2 (Conv2D)           (200, 256, 14, 14)   590080      mask-conv-l1[0][0]               
__________________________________________________________________________________________________
mask-conv-l3 (Conv2D)           (200, 256, 14, 14)   590080      mask-conv-l2[0][0]               
__________________________________________________________________________________________________
conv5-mask (Conv2DTranspose)    (200, 256, 28, 28)   262400      mask-conv-l3[0][0]               
__________________________________________________________________________________________________
mask_fcn_logits (Conv2D)        (200, 2, 28, 28)     514         conv5-mask[0][0]                 
__________________________________________________________________________________________________
mask_postprocess (MaskPostproce (2, 100, 28, 28)     0           mask_fcn_logits[0][0]            
                                                                 gpu_detections[0][2]             
__________________________________________________________________________________________________
mask_sigmoid (Activation)       (2, 100, 28, 28)     0           mask_postprocess[0][0]           
==================================================================================================
Total params: 44,028,635
Trainable params: 23,508,032
Non-trainable params: 20,520,603
__________________________________________________________________________________________________
Traceback (most recent call last):
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/scripts/inference.py", line 351, in <module>
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/scripts/inference.py", line 343, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/scripts/inference.py", line 305, in infer
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/executer/distributed_executer.py", line 490, in infer
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/mask_rcnn/utils/evaluation.py", line 242, in infer
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_estimator/python/estimator/estimator.py", line 638, in predict
    hooks=all_hooks) as mon_sess:
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1014, in __init__
    stop_grace_period_secs=stop_grace_period_secs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 725, in __init__
    self._sess = _RecoverableSession(self._coordinated_creator)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1207, in __init__
    _WrappedSession.__init__(self, self._create_session())
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1212, in _create_session
    return self._sess_creator.create_session()
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 878, in create_session
    self.tf_sess = self._session_creator.create_session()
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 647, in create_session
    init_fn=self._scaffold.init_fn)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/session_manager.py", line 290, in prepare_session
    config=config)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/session_manager.py", line 204, in _restore_checkpoint
    saver.restore(sess, checkpoint_filename_with_path)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/saver.py", line 1282, in restore
    checkpoint_prefix)
ValueError: The passed save_path is not a valid checkpoint: /tmp/tmpc0iiwecg/model.ckpt-90000

After pruning, did you run retraining?
More, after retraining, did you rename the model’s name? If yes, please do not change it and rerun inference.

No, I did not retrain the model after pruning. Is it necessary?

Yes, it is needed. Refer to Mask Rcnn Inference Error

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.