Using TensorFlow backend. WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them. /usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version! RequestsDependencyWarning) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_NVVM=/usr/local/cuda/nvvm/lib64/libnvvm.so. For more information about alternatives visit: ('http://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) /usr/local/lib/python3.6/dist-packages/numba/cuda/envvars.py:17: NumbaWarning: Environment variables with the 'NUMBAPRO' prefix are deprecated and consequently ignored, found use of NUMBAPRO_LIBDEVICE=/usr/local/cuda/nvvm/libdevice/. For more information about alternatives visit: ('http://numba.pydata.org/numba-doc/latest/cuda/overview.html', '#cudatoolkit-lookup') warnings.warn(errors.NumbaWarning(msg)) 2022-07-12 15:27:42,035 [INFO] iva.mask_rcnn.utils.spec_loader: Loading specification from /workspace/tao-experiments/specs/spec_mrcnn_demi_epoch.txt 2022-07-12 15:27:42,079 [INFO] root: Loading weights from /workspace/tao-experiments/mask_rcnn/model.step-403.tlt INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmpjt9xkve_', '_tf_random_seed': 123, '_save_summary_steps': None, '_save_checkpoints_steps': None, '_save_checkpoints_secs': None, '_session_config': gpu_options { allow_growth: true force_gpu_compatible: true } allow_soft_placement: true graph_options { rewrite_options { meta_optimizer_iterations: TWO } } , '_keep_checkpoint_max': 20, '_keep_checkpoint_every_n_hours': None, '_log_step_count_steps': None, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': , '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1} 2022-07-12 15:28:05,022 [INFO] tensorflow: Using config: {'_model_dir': '/tmp/tmpjt9xkve_', '_tf_random_seed': 123, '_save_summary_steps': None, '_save_checkpoints_steps': None, '_save_checkpoints_secs': None, '_session_config': gpu_options { allow_growth: true force_gpu_compatible: true } allow_soft_placement: true graph_options { rewrite_options { meta_optimizer_iterations: TWO } } , '_keep_checkpoint_max': 20, '_keep_checkpoint_every_n_hours': None, '_log_step_count_steps': None, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': , '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1} INFO:tensorflow:Create CheckpointSaverHook. 2022-07-12 15:28:05,023 [INFO] tensorflow: Create CheckpointSaverHook. WARNING:tensorflow:Entity ._prefetch_dataset at 0x7fb7c50522f0> 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 ._prefetch_dataset at 0x7fb7c50522f0>. 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 2022-07-12 15:28:05,161 [WARNING] tensorflow: Entity ._prefetch_dataset at 0x7fb7c50522f0> 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 ._prefetch_dataset at 0x7fb7c50522f0>. 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 : [*] Limiting the amount of sample to: 200 WARNING:tensorflow:From /opt/nvidia/third_party/keras/tensorflow_backend.py:349: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead. 2022-07-12 15:28:05,197 [WARNING] tensorflow: From /opt/nvidia/third_party/keras/tensorflow_backend.py:349: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead. 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. 2022-07-12 15:28:05,226 [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 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 . 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 2022-07-12 15:28:05,288 [WARNING] tensorflow: Entity 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 . 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:The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. 2022-07-12 15:28:05,311 [WARNING] tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. WARNING:tensorflow:The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. 2022-07-12 15:28:05,315 [WARNING] tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. WARNING:tensorflow:The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. 2022-07-12 15:28:05,323 [WARNING] tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. WARNING:tensorflow:The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. 2022-07-12 15:28:05,330 [WARNING] tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical. INFO:tensorflow:Calling model_fn. 2022-07-12 15:28:05,814 [INFO] tensorflow: Calling model_fn. [MaskRCNN] INFO : *********************** [MaskRCNN] INFO : Loading model graph... [MaskRCNN] INFO : *********************** WARNING:tensorflow:Entity > 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 >. 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 2022-07-12 15:28:07,521 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:07,566 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:07,905 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:08,171 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:08,224 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:08,229 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:08,234 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:08,292 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:08,719 [WARNING] tensorflow: Entity > 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 >. 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 > 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 >. 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 2022-07-12 15:28:24,852 [WARNING] tensorflow: Entity > 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 >. 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) [(1, 3, 832, 1344)] 0 __________________________________________________________________________________________________ conv1 (Conv2D) (1, 64, 416, 672) 9408 image_input[0][0] __________________________________________________________________________________________________ bn_conv1 (BatchNormalization) (1, 64, 416, 672) 256 conv1[0][0] __________________________________________________________________________________________________ activation (Activation) (1, 64, 416, 672) 0 bn_conv1[0][0] __________________________________________________________________________________________________ max_pooling2d (MaxPooling2D) (1, 64, 208, 336) 0 activation[0][0] __________________________________________________________________________________________________ block_1a_conv_1 (Conv2D) (1, 64, 208, 336) 4096 max_pooling2d[0][0] __________________________________________________________________________________________________ block_1a_bn_1 (BatchNormalizati (1, 64, 208, 336) 256 block_1a_conv_1[0][0] __________________________________________________________________________________________________ block_1a_relu_1 (Activation) (1, 64, 208, 336) 0 block_1a_bn_1[0][0] __________________________________________________________________________________________________ block_1a_conv_2 (Conv2D) (1, 64, 208, 336) 36864 block_1a_relu_1[0][0] __________________________________________________________________________________________________ block_1a_bn_2 (BatchNormalizati (1, 64, 208, 336) 256 block_1a_conv_2[0][0] __________________________________________________________________________________________________ block_1a_relu_2 (Activation) (1, 64, 208, 336) 0 block_1a_bn_2[0][0] __________________________________________________________________________________________________ block_1a_conv_3 (Conv2D) (1, 256, 208, 336) 16384 block_1a_relu_2[0][0] __________________________________________________________________________________________________ block_1a_conv_shortcut (Conv2D) (1, 256, 208, 336) 16384 max_pooling2d[0][0] __________________________________________________________________________________________________ block_1a_bn_3 (BatchNormalizati (1, 256, 208, 336) 1024 block_1a_conv_3[0][0] __________________________________________________________________________________________________ block_1a_bn_shortcut (BatchNorm (1, 256, 208, 336) 1024 block_1a_conv_shortcut[0][0] __________________________________________________________________________________________________ add (Add) (1, 256, 208, 336) 0 block_1a_bn_3[0][0] block_1a_bn_shortcut[0][0] __________________________________________________________________________________________________ block_1a_relu (Activation) (1, 256, 208, 336) 0 add[0][0] __________________________________________________________________________________________________ block_1b_conv_1 (Conv2D) (1, 64, 208, 336) 16384 block_1a_relu[0][0] __________________________________________________________________________________________________ block_1b_bn_1 (BatchNormalizati (1, 64, 208, 336) 256 block_1b_conv_1[0][0] __________________________________________________________________________________________________ block_1b_relu_1 (Activation) (1, 64, 208, 336) 0 block_1b_bn_1[0][0] __________________________________________________________________________________________________ block_1b_conv_2 (Conv2D) (1, 64, 208, 336) 36864 block_1b_relu_1[0][0] __________________________________________________________________________________________________ block_1b_bn_2 (BatchNormalizati (1, 64, 208, 336) 256 block_1b_conv_2[0][0] __________________________________________________________________________________________________ block_1b_relu_2 (Activation) (1, 64, 208, 336) 0 block_1b_bn_2[0][0] __________________________________________________________________________________________________ block_1b_conv_3 (Conv2D) (1, 256, 208, 336) 16384 block_1b_relu_2[0][0] __________________________________________________________________________________________________ block_1b_bn_3 (BatchNormalizati (1, 256, 208, 336) 1024 block_1b_conv_3[0][0] __________________________________________________________________________________________________ add_1 (Add) (1, 256, 208, 336) 0 block_1b_bn_3[0][0] block_1a_relu[0][0] __________________________________________________________________________________________________ block_1b_relu (Activation) (1, 256, 208, 336) 0 add_1[0][0] __________________________________________________________________________________________________ block_1c_conv_1 (Conv2D) (1, 64, 208, 336) 16384 block_1b_relu[0][0] __________________________________________________________________________________________________ block_1c_bn_1 (BatchNormalizati (1, 64, 208, 336) 256 block_1c_conv_1[0][0] __________________________________________________________________________________________________ block_1c_relu_1 (Activation) (1, 64, 208, 336) 0 block_1c_bn_1[0][0] __________________________________________________________________________________________________ block_1c_conv_2 (Conv2D) (1, 64, 208, 336) 36864 block_1c_relu_1[0][0] __________________________________________________________________________________________________ block_1c_bn_2 (BatchNormalizati (1, 64, 208, 336) 256 block_1c_conv_2[0][0] __________________________________________________________________________________________________ block_1c_relu_2 (Activation) (1, 64, 208, 336) 0 block_1c_bn_2[0][0] __________________________________________________________________________________________________ block_1c_conv_3 (Conv2D) (1, 256, 208, 336) 16384 block_1c_relu_2[0][0] __________________________________________________________________________________________________ block_1c_bn_3 (BatchNormalizati (1, 256, 208, 336) 1024 block_1c_conv_3[0][0] __________________________________________________________________________________________________ add_2 (Add) (1, 256, 208, 336) 0 block_1c_bn_3[0][0] block_1b_relu[0][0] __________________________________________________________________________________________________ block_1c_relu (Activation) (1, 256, 208, 336) 0 add_2[0][0] __________________________________________________________________________________________________ block_2a_conv_1 (Conv2D) (1, 128, 104, 168) 32768 block_1c_relu[0][0] __________________________________________________________________________________________________ block_2a_bn_1 (BatchNormalizati (1, 128, 104, 168) 512 block_2a_conv_1[0][0] __________________________________________________________________________________________________ block_2a_relu_1 (Activation) (1, 128, 104, 168) 0 block_2a_bn_1[0][0] __________________________________________________________________________________________________ block_2a_conv_2 (Conv2D) (1, 128, 104, 168) 147456 block_2a_relu_1[0][0] __________________________________________________________________________________________________ block_2a_bn_2 (BatchNormalizati (1, 128, 104, 168) 512 block_2a_conv_2[0][0] __________________________________________________________________________________________________ block_2a_relu_2 (Activation) (1, 128, 104, 168) 0 block_2a_bn_2[0][0] __________________________________________________________________________________________________ block_2a_conv_3 (Conv2D) (1, 512, 104, 168) 65536 block_2a_relu_2[0][0] __________________________________________________________________________________________________ block_2a_conv_shortcut (Conv2D) (1, 512, 104, 168) 131072 block_1c_relu[0][0] __________________________________________________________________________________________________ block_2a_bn_3 (BatchNormalizati (1, 512, 104, 168) 2048 block_2a_conv_3[0][0] __________________________________________________________________________________________________ block_2a_bn_shortcut (BatchNorm (1, 512, 104, 168) 2048 block_2a_conv_shortcut[0][0] __________________________________________________________________________________________________ add_3 (Add) (1, 512, 104, 168) 0 block_2a_bn_3[0][0] block_2a_bn_shortcut[0][0] __________________________________________________________________________________________________ block_2a_relu (Activation) (1, 512, 104, 168) 0 add_3[0][0] __________________________________________________________________________________________________ block_2b_conv_1 (Conv2D) (1, 128, 104, 168) 65536 block_2a_relu[0][0] __________________________________________________________________________________________________ block_2b_bn_1 (BatchNormalizati (1, 128, 104, 168) 512 block_2b_conv_1[0][0] __________________________________________________________________________________________________ block_2b_relu_1 (Activation) (1, 128, 104, 168) 0 block_2b_bn_1[0][0] __________________________________________________________________________________________________ block_2b_conv_2 (Conv2D) (1, 128, 104, 168) 147456 block_2b_relu_1[0][0] __________________________________________________________________________________________________ block_2b_bn_2 (BatchNormalizati (1, 128, 104, 168) 512 block_2b_conv_2[0][0] __________________________________________________________________________________________________ block_2b_relu_2 (Activation) (1, 128, 104, 168) 0 block_2b_bn_2[0][0] __________________________________________________________________________________________________ block_2b_conv_3 (Conv2D) (1, 512, 104, 168) 65536 block_2b_relu_2[0][0] __________________________________________________________________________________________________ block_2b_bn_3 (BatchNormalizati (1, 512, 104, 168) 2048 block_2b_conv_3[0][0] __________________________________________________________________________________________________ add_4 (Add) (1, 512, 104, 168) 0 block_2b_bn_3[0][0] block_2a_relu[0][0] __________________________________________________________________________________________________ block_2b_relu (Activation) (1, 512, 104, 168) 0 add_4[0][0] __________________________________________________________________________________________________ block_2c_conv_1 (Conv2D) (1, 128, 104, 168) 65536 block_2b_relu[0][0] __________________________________________________________________________________________________ block_2c_bn_1 (BatchNormalizati (1, 128, 104, 168) 512 block_2c_conv_1[0][0] __________________________________________________________________________________________________ block_2c_relu_1 (Activation) (1, 128, 104, 168) 0 block_2c_bn_1[0][0] __________________________________________________________________________________________________ block_2c_conv_2 (Conv2D) (1, 128, 104, 168) 147456 block_2c_relu_1[0][0] __________________________________________________________________________________________________ block_2c_bn_2 (BatchNormalizati (1, 128, 104, 168) 512 block_2c_conv_2[0][0] __________________________________________________________________________________________________ block_2c_relu_2 (Activation) (1, 128, 104, 168) 0 block_2c_bn_2[0][0] __________________________________________________________________________________________________ block_2c_conv_3 (Conv2D) (1, 512, 104, 168) 65536 block_2c_relu_2[0][0] __________________________________________________________________________________________________ block_2c_bn_3 (BatchNormalizati (1, 512, 104, 168) 2048 block_2c_conv_3[0][0] __________________________________________________________________________________________________ add_5 (Add) (1, 512, 104, 168) 0 block_2c_bn_3[0][0] block_2b_relu[0][0] __________________________________________________________________________________________________ block_2c_relu (Activation) (1, 512, 104, 168) 0 add_5[0][0] __________________________________________________________________________________________________ block_2d_conv_1 (Conv2D) (1, 128, 104, 168) 65536 block_2c_relu[0][0] __________________________________________________________________________________________________ block_2d_bn_1 (BatchNormalizati (1, 128, 104, 168) 512 block_2d_conv_1[0][0] __________________________________________________________________________________________________ block_2d_relu_1 (Activation) (1, 128, 104, 168) 0 block_2d_bn_1[0][0] __________________________________________________________________________________________________ block_2d_conv_2 (Conv2D) (1, 128, 104, 168) 147456 block_2d_relu_1[0][0] __________________________________________________________________________________________________ block_2d_bn_2 (BatchNormalizati (1, 128, 104, 168) 512 block_2d_conv_2[0][0] __________________________________________________________________________________________________ block_2d_relu_2 (Activation) (1, 128, 104, 168) 0 block_2d_bn_2[0][0] __________________________________________________________________________________________________ block_2d_conv_3 (Conv2D) (1, 512, 104, 168) 65536 block_2d_relu_2[0][0] __________________________________________________________________________________________________ block_2d_bn_3 (BatchNormalizati (1, 512, 104, 168) 2048 block_2d_conv_3[0][0] __________________________________________________________________________________________________ add_6 (Add) (1, 512, 104, 168) 0 block_2d_bn_3[0][0] block_2c_relu[0][0] __________________________________________________________________________________________________ block_2d_relu (Activation) (1, 512, 104, 168) 0 add_6[0][0] __________________________________________________________________________________________________ block_3a_conv_1 (Conv2D) (1, 256, 52, 84) 131072 block_2d_relu[0][0] __________________________________________________________________________________________________ block_3a_bn_1 (BatchNormalizati (1, 256, 52, 84) 1024 block_3a_conv_1[0][0] __________________________________________________________________________________________________ block_3a_relu_1 (Activation) (1, 256, 52, 84) 0 block_3a_bn_1[0][0] __________________________________________________________________________________________________ block_3a_conv_2 (Conv2D) (1, 256, 52, 84) 589824 block_3a_relu_1[0][0] __________________________________________________________________________________________________ block_3a_bn_2 (BatchNormalizati (1, 256, 52, 84) 1024 block_3a_conv_2[0][0] __________________________________________________________________________________________________ block_3a_relu_2 (Activation) (1, 256, 52, 84) 0 block_3a_bn_2[0][0] __________________________________________________________________________________________________ block_3a_conv_3 (Conv2D) (1, 1024, 52, 84) 262144 block_3a_relu_2[0][0] __________________________________________________________________________________________________ block_3a_conv_shortcut (Conv2D) (1, 1024, 52, 84) 524288 block_2d_relu[0][0] __________________________________________________________________________________________________ block_3a_bn_3 (BatchNormalizati (1, 1024, 52, 84) 4096 block_3a_conv_3[0][0] __________________________________________________________________________________________________ block_3a_bn_shortcut (BatchNorm (1, 1024, 52, 84) 4096 block_3a_conv_shortcut[0][0] __________________________________________________________________________________________________ add_7 (Add) (1, 1024, 52, 84) 0 block_3a_bn_3[0][0] block_3a_bn_shortcut[0][0] __________________________________________________________________________________________________ block_3a_relu (Activation) (1, 1024, 52, 84) 0 add_7[0][0] __________________________________________________________________________________________________ block_3b_conv_1 (Conv2D) (1, 256, 52, 84) 262144 block_3a_relu[0][0] __________________________________________________________________________________________________ block_3b_bn_1 (BatchNormalizati (1, 256, 52, 84) 1024 block_3b_conv_1[0][0] __________________________________________________________________________________________________ block_3b_relu_1 (Activation) (1, 256, 52, 84) 0 block_3b_bn_1[0][0] __________________________________________________________________________________________________ block_3b_conv_2 (Conv2D) (1, 256, 52, 84) 589824 block_3b_relu_1[0][0] __________________________________________________________________________________________________ block_3b_bn_2 (BatchNormalizati (1, 256, 52, 84) 1024 block_3b_conv_2[0][0] __________________________________________________________________________________________________ block_3b_relu_2 (Activation) (1, 256, 52, 84) 0 block_3b_bn_2[0][0] __________________________________________________________________________________________________ block_3b_conv_3 (Conv2D) (1, 1024, 52, 84) 262144 block_3b_relu_2[0][0] __________________________________________________________________________________________________ block_3b_bn_3 (BatchNormalizati (1, 1024, 52, 84) 4096 block_3b_conv_3[0][0] __________________________________________________________________________________________________ add_8 (Add) (1, 1024, 52, 84) 0 block_3b_bn_3[0][0] block_3a_relu[0][0] __________________________________________________________________________________________________ block_3b_relu (Activation) (1, 1024, 52, 84) 0 add_8[0][0] __________________________________________________________________________________________________ block_3c_conv_1 (Conv2D) (1, 256, 52, 84) 262144 block_3b_relu[0][0] __________________________________________________________________________________________________ block_3c_bn_1 (BatchNormalizati (1, 256, 52, 84) 1024 block_3c_conv_1[0][0] __________________________________________________________________________________________________ block_3c_relu_1 (Activation) (1, 256, 52, 84) 0 block_3c_bn_1[0][0] __________________________________________________________________________________________________ block_3c_conv_2 (Conv2D) (1, 256, 52, 84) 589824 block_3c_relu_1[0][0] __________________________________________________________________________________________________ block_3c_bn_2 (BatchNormalizati (1, 256, 52, 84) 1024 block_3c_conv_2[0][0] __________________________________________________________________________________________________ block_3c_relu_2 (Activation) (1, 256, 52, 84) 0 block_3c_bn_2[0][0] __________________________________________________________________________________________________ block_3c_conv_3 (Conv2D) (1, 1024, 52, 84) 262144 block_3c_relu_2[0][0] __________________________________________________________________________________________________ block_3c_bn_3 (BatchNormalizati (1, 1024, 52, 84) 4096 block_3c_conv_3[0][0] __________________________________________________________________________________________________ add_9 (Add) (1, 1024, 52, 84) 0 block_3c_bn_3[0][0] block_3b_relu[0][0] __________________________________________________________________________________________________ block_3c_relu (Activation) (1, 1024, 52, 84) 0 add_9[0][0] __________________________________________________________________________________________________ block_3d_conv_1 (Conv2D) (1, 256, 52, 84) 262144 block_3c_relu[0][0] __________________________________________________________________________________________________ block_3d_bn_1 (BatchNormalizati (1, 256, 52, 84) 1024 block_3d_conv_1[0][0] __________________________________________________________________________________________________ block_3d_relu_1 (Activation) (1, 256, 52, 84) 0 block_3d_bn_1[0][0] __________________________________________________________________________________________________ block_3d_conv_2 (Conv2D) (1, 256, 52, 84) 589824 block_3d_relu_1[0][0] __________________________________________________________________________________________________ block_3d_bn_2 (BatchNormalizati (1, 256, 52, 84) 1024 block_3d_conv_2[0][0] __________________________________________________________________________________________________ block_3d_relu_2 (Activation) (1, 256, 52, 84) 0 block_3d_bn_2[0][0] __________________________________________________________________________________________________ block_3d_conv_3 (Conv2D) (1, 1024, 52, 84) 262144 block_3d_relu_2[0][0] __________________________________________________________________________________________________ block_3d_bn_3 (BatchNormalizati (1, 1024, 52, 84) 4096 block_3d_conv_3[0][0] __________________________________________________________________________________________________ add_10 (Add) (1, 1024, 52, 84) 0 block_3d_bn_3[0][0] block_3c_relu[0][0] __________________________________________________________________________________________________ block_3d_relu (Activation) (1, 1024, 52, 84) 0 add_10[0][0] __________________________________________________________________________________________________ block_3e_conv_1 (Conv2D) (1, 256, 52, 84) 262144 block_3d_relu[0][0] __________________________________________________________________________________________________ block_3e_bn_1 (BatchNormalizati (1, 256, 52, 84) 1024 block_3e_conv_1[0][0] __________________________________________________________________________________________________ block_3e_relu_1 (Activation) (1, 256, 52, 84) 0 block_3e_bn_1[0][0] __________________________________________________________________________________________________ block_3e_conv_2 (Conv2D) (1, 256, 52, 84) 589824 block_3e_relu_1[0][0] __________________________________________________________________________________________________ block_3e_bn_2 (BatchNormalizati (1, 256, 52, 84) 1024 block_3e_conv_2[0][0] __________________________________________________________________________________________________ block_3e_relu_2 (Activation) (1, 256, 52, 84) 0 block_3e_bn_2[0][0] __________________________________________________________________________________________________ block_3e_conv_3 (Conv2D) (1, 1024, 52, 84) 262144 block_3e_relu_2[0][0] __________________________________________________________________________________________________ block_3e_bn_3 (BatchNormalizati (1, 1024, 52, 84) 4096 block_3e_conv_3[0][0] __________________________________________________________________________________________________ add_11 (Add) (1, 1024, 52, 84) 0 block_3e_bn_3[0][0] block_3d_relu[0][0] __________________________________________________________________________________________________ block_3e_relu (Activation) (1, 1024, 52, 84) 0 add_11[0][0] __________________________________________________________________________________________________ block_3f_conv_1 (Conv2D) (1, 256, 52, 84) 262144 block_3e_relu[0][0] __________________________________________________________________________________________________ block_3f_bn_1 (BatchNormalizati (1, 256, 52, 84) 1024 block_3f_conv_1[0][0] __________________________________________________________________________________________________ block_3f_relu_1 (Activation) (1, 256, 52, 84) 0 block_3f_bn_1[0][0] __________________________________________________________________________________________________ block_3f_conv_2 (Conv2D) (1, 256, 52, 84) 589824 block_3f_relu_1[0][0] __________________________________________________________________________________________________ block_3f_bn_2 (BatchNormalizati (1, 256, 52, 84) 1024 block_3f_conv_2[0][0] __________________________________________________________________________________________________ block_3f_relu_2 (Activation) (1, 256, 52, 84) 0 block_3f_bn_2[0][0] __________________________________________________________________________________________________ block_3f_conv_3 (Conv2D) (1, 1024, 52, 84) 262144 block_3f_relu_2[0][0] __________________________________________________________________________________________________ block_3f_bn_3 (BatchNormalizati (1, 1024, 52, 84) 4096 block_3f_conv_3[0][0] __________________________________________________________________________________________________ add_12 (Add) (1, 1024, 52, 84) 0 block_3f_bn_3[0][0] block_3e_relu[0][0] __________________________________________________________________________________________________ block_3f_relu (Activation) (1, 1024, 52, 84) 0 add_12[0][0] __________________________________________________________________________________________________ block_4a_conv_1 (Conv2D) (1, 512, 26, 42) 524288 block_3f_relu[0][0] __________________________________________________________________________________________________ block_4a_bn_1 (BatchNormalizati (1, 512, 26, 42) 2048 block_4a_conv_1[0][0] __________________________________________________________________________________________________ block_4a_relu_1 (Activation) (1, 512, 26, 42) 0 block_4a_bn_1[0][0] __________________________________________________________________________________________________ block_4a_conv_2 (Conv2D) (1, 512, 26, 42) 2359296 block_4a_relu_1[0][0] __________________________________________________________________________________________________ block_4a_bn_2 (BatchNormalizati (1, 512, 26, 42) 2048 block_4a_conv_2[0][0] __________________________________________________________________________________________________ block_4a_relu_2 (Activation) (1, 512, 26, 42) 0 block_4a_bn_2[0][0] __________________________________________________________________________________________________ block_4a_conv_3 (Conv2D) (1, 2048, 26, 42) 1048576 block_4a_relu_2[0][0] __________________________________________________________________________________________________ block_4a_conv_shortcut (Conv2D) (1, 2048, 26, 42) 2097152 block_3f_relu[0][0] __________________________________________________________________________________________________ block_4a_bn_3 (BatchNormalizati (1, 2048, 26, 42) 8192 block_4a_conv_3[0][0] __________________________________________________________________________________________________ block_4a_bn_shortcut (BatchNorm (1, 2048, 26, 42) 8192 block_4a_conv_shortcut[0][0] __________________________________________________________________________________________________ add_13 (Add) (1, 2048, 26, 42) 0 block_4a_bn_3[0][0] block_4a_bn_shortcut[0][0] __________________________________________________________________________________________________ block_4a_relu (Activation) (1, 2048, 26, 42) 0 add_13[0][0] __________________________________________________________________________________________________ block_4b_conv_1 (Conv2D) (1, 512, 26, 42) 1048576 block_4a_relu[0][0] __________________________________________________________________________________________________ block_4b_bn_1 (BatchNormalizati (1, 512, 26, 42) 2048 block_4b_conv_1[0][0] __________________________________________________________________________________________________ block_4b_relu_1 (Activation) (1, 512, 26, 42) 0 block_4b_bn_1[0][0] __________________________________________________________________________________________________ block_4b_conv_2 (Conv2D) (1, 512, 26, 42) 2359296 block_4b_relu_1[0][0] __________________________________________________________________________________________________ block_4b_bn_2 (BatchNormalizati (1, 512, 26, 42) 2048 block_4b_conv_2[0][0] __________________________________________________________________________________________________ block_4b_relu_2 (Activation) (1, 512, 26, 42) 0 block_4b_bn_2[0][0] __________________________________________________________________________________________________ block_4b_conv_3 (Conv2D) (1, 2048, 26, 42) 1048576 block_4b_relu_2[0][0] __________________________________________________________________________________________________ block_4b_bn_3 (BatchNormalizati (1, 2048, 26, 42) 8192 block_4b_conv_3[0][0] __________________________________________________________________________________________________ add_14 (Add) (1, 2048, 26, 42) 0 block_4b_bn_3[0][0] block_4a_relu[0][0] __________________________________________________________________________________________________ block_4b_relu (Activation) (1, 2048, 26, 42) 0 add_14[0][0] __________________________________________________________________________________________________ block_4c_conv_1 (Conv2D) (1, 512, 26, 42) 1048576 block_4b_relu[0][0] __________________________________________________________________________________________________ block_4c_bn_1 (BatchNormalizati (1, 512, 26, 42) 2048 block_4c_conv_1[0][0] __________________________________________________________________________________________________ block_4c_relu_1 (Activation) (1, 512, 26, 42) 0 block_4c_bn_1[0][0] __________________________________________________________________________________________________ block_4c_conv_2 (Conv2D) (1, 512, 26, 42) 2359296 block_4c_relu_1[0][0] __________________________________________________________________________________________________ block_4c_bn_2 (BatchNormalizati (1, 512, 26, 42) 2048 block_4c_conv_2[0][0] __________________________________________________________________________________________________ block_4c_relu_2 (Activation) (1, 512, 26, 42) 0 block_4c_bn_2[0][0] __________________________________________________________________________________________________ block_4c_conv_3 (Conv2D) (1, 2048, 26, 42) 1048576 block_4c_relu_2[0][0] __________________________________________________________________________________________________ block_4c_bn_3 (BatchNormalizati (1, 2048, 26, 42) 8192 block_4c_conv_3[0][0] __________________________________________________________________________________________________ add_15 (Add) (1, 2048, 26, 42) 0 block_4c_bn_3[0][0] block_4b_relu[0][0] __________________________________________________________________________________________________ block_4c_relu (Activation) (1, 2048, 26, 42) 0 add_15[0][0] __________________________________________________________________________________________________ l5 (Conv2D) (1, 256, 26, 42) 524544 block_4c_relu[0][0] __________________________________________________________________________________________________ l4 (Conv2D) (1, 256, 52, 84) 262400 block_3f_relu[0][0] __________________________________________________________________________________________________ FPN_up_4 (UpSampling2D) (1, 256, 52, 84) 0 l5[0][0] __________________________________________________________________________________________________ FPN_add_4 (Add) (1, 256, 52, 84) 0 l4[0][0] FPN_up_4[0][0] __________________________________________________________________________________________________ l3 (Conv2D) (1, 256, 104, 168) 131328 block_2d_relu[0][0] __________________________________________________________________________________________________ FPN_up_3 (UpSampling2D) (1, 256, 104, 168) 0 FPN_add_4[0][0] __________________________________________________________________________________________________ FPN_add_3 (Add) (1, 256, 104, 168) 0 l3[0][0] FPN_up_3[0][0] __________________________________________________________________________________________________ l2 (Conv2D) (1, 256, 208, 336) 65792 block_1c_relu[0][0] __________________________________________________________________________________________________ FPN_up_2 (UpSampling2D) (1, 256, 208, 336) 0 FPN_add_3[0][0] __________________________________________________________________________________________________ FPN_add_2 (Add) (1, 256, 208, 336) 0 l2[0][0] FPN_up_2[0][0] __________________________________________________________________________________________________ post_hoc_d5 (Conv2D) (1, 256, 26, 42) 590080 l5[0][0] __________________________________________________________________________________________________ post_hoc_d2 (Conv2D) (1, 256, 208, 336) 590080 FPN_add_2[0][0] __________________________________________________________________________________________________ post_hoc_d3 (Conv2D) (1, 256, 104, 168) 590080 FPN_add_3[0][0] __________________________________________________________________________________________________ post_hoc_d4 (Conv2D) (1, 256, 52, 84) 590080 FPN_add_4[0][0] __________________________________________________________________________________________________ p6 (MaxPooling2D) (1, 256, 13, 21) 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) (1, 208, 336, 3) 0 rpn-class[0][0] __________________________________________________________________________________________________ permute_2 (Permute) (1, 104, 168, 3) 0 rpn-class[1][0] __________________________________________________________________________________________________ permute_4 (Permute) (1, 52, 84, 3) 0 rpn-class[2][0] __________________________________________________________________________________________________ permute_6 (Permute) (1, 26, 42, 3) 0 rpn-class[3][0] __________________________________________________________________________________________________ permute_8 (Permute) (1, 13, 21, 3) 0 rpn-class[4][0] __________________________________________________________________________________________________ permute_1 (Permute) (1, 208, 336, 12) 0 rpn-box[0][0] __________________________________________________________________________________________________ permute_3 (Permute) (1, 104, 168, 12) 0 rpn-box[1][0] __________________________________________________________________________________________________ permute_5 (Permute) (1, 52, 84, 12) 0 rpn-box[2][0] __________________________________________________________________________________________________ permute_7 (Permute) (1, 26, 42, 12) 0 rpn-box[3][0] __________________________________________________________________________________________________ permute_9 (Permute) (1, 13, 21, 12) 0 rpn-box[4][0] __________________________________________________________________________________________________ anchor_layer (AnchorLayer) OrderedDict([(2, (20 0 image_input[0][0] __________________________________________________________________________________________________ info_input (InfoInput) [(1, 5)] 0 __________________________________________________________________________________________________ MLP (MultilevelProposal) ((1, 1000), (1, 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 (1, 1000, 256, 16, 1 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 (1000, 65536) 0 multilevel_crop_resize[0][0] __________________________________________________________________________________________________ fc6 (Dense) (1000, 1024) 67109888 box_head_reshape1[0][0] __________________________________________________________________________________________________ fc7 (Dense) (1000, 1024) 1049600 fc6[0][0] __________________________________________________________________________________________________ class-predict (Dense) (1000, 4) 4100 fc7[0][0] __________________________________________________________________________________________________ box-predict (Dense) (1000, 16) 16400 fc7[0][0] __________________________________________________________________________________________________ box_head_reshape2 (ReshapeLayer (1, 1000, 4) 0 class-predict[0][0] __________________________________________________________________________________________________ box_head_reshape3 (ReshapeLayer (1, 1000, 16) 0 box-predict[0][0] __________________________________________________________________________________________________ gpu_detections (GPUDetections) ((1,), (1, 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 (1, 100, 256, 32, 32 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 (100, 256, 32, 32) 0 multilevel_crop_resize_1[0][0] __________________________________________________________________________________________________ mask-conv-l0 (Conv2D) (100, 256, 32, 32) 590080 mask_head_reshape_1[0][0] __________________________________________________________________________________________________ mask-conv-l1 (Conv2D) (100, 256, 32, 32) 590080 mask-conv-l0[0][0] __________________________________________________________________________________________________ mask-conv-l2 (Conv2D) (100, 256, 32, 32) 590080 mask-conv-l1[0][0] __________________________________________________________________________________________________INFO:tensorflow:Done calling model_fn. 2022-07-12 15:28:25,096 [INFO] tensorflow: Done calling model_fn. INFO:tensorflow:Graph was finalized. 2022-07-12 15:28:26,230 [INFO] tensorflow: Graph was finalized. INFO:tensorflow:Restoring parameters from /tmp/tmpjt9xkve_/model.ckpt-403 2022-07-12 15:28:26,232 [INFO] tensorflow: Restoring parameters from /tmp/tmpjt9xkve_/model.ckpt-403 INFO:tensorflow:Running local_init_op. 2022-07-12 15:28:27,324 [INFO] tensorflow: Running local_init_op. INFO:tensorflow:Done running local_init_op. 2022-07-12 15:28:27,418 [INFO] tensorflow: Done running local_init_op. INFO:tensorflow:Saving checkpoints for 403 into /tmp/tmp5u8lau5j/model.ckpt. 2022-07-12 15:28:32,133 [INFO] tensorflow: Saving checkpoints for 403 into /tmp/tmp5u8lau5j/model.ckpt. WARNING:tensorflow:From /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/export/exporter.py:252: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead. 2022-07-12 15:28:47,031 [WARNING] tensorflow: From /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/export/exporter.py:252: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead. INFO:tensorflow:Restoring parameters from /tmp/tmp5u8lau5j/model.ckpt-403 2022-07-12 15:28:48,433 [INFO] tensorflow: Restoring parameters from /tmp/tmp5u8lau5j/model.ckpt-403 INFO:tensorflow:Froze 307 variables. 2022-07-12 15:28:49,844 [INFO] tensorflow: Froze 307 variables. INFO:tensorflow:Converted 307 variables to const ops. 2022-07-12 15:28:50,353 [INFO] tensorflow: Converted 307 variables to const ops. WARNING:tensorflow:From /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/export/exporter.py:292: The name tf.reset_default_graph is deprecated. Please use tf.compat.v1.reset_default_graph instead. 2022-07-12 15:28:59,569 [WARNING] tensorflow: From /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/export/exporter.py:292: The name tf.reset_default_graph is deprecated. Please use tf.compat.v1.reset_default_graph instead. 2022-07-12 15:28:59,570 [INFO] numba.cuda.cudadrv.driver: init WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/graphsurgeon/StaticGraph.py:125: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead. 2022-07-12 15:29:00,001 [WARNING] tensorflow: From /usr/local/lib/python3.6/dist-packages/graphsurgeon/StaticGraph.py:125: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead. WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/uff/converters/tensorflow/converter.py:179: The name tf.AttrValue is deprecated. Please use tf.compat.v1.AttrValue instead. 2022-07-12 15:29:01,092 [WARNING] tensorflow: From /usr/local/lib/python3.6/dist-packages/uff/converters/tensorflow/converter.py:179: The name tf.AttrValue is deprecated. Please use tf.compat.v1.AttrValue instead. 2022-07-12 15:29:11,523 [INFO] iva.mask_rcnn.export.exporter: Converted model was saved into /workspace/tao-experiments/mask_rcnn/model.step-403.etlt [07/12/2022-15:29:22] [TRT] [E] 3: fc6/MatMul:kernel weights has count 67108864 but 12845056 was expected [07/12/2022-15:29:22] [TRT] [E] 3: fc6/MatMul:kernel weights has count 67108864 but 12845056 was expected [07/12/2022-15:29:22] [TRT] [E] 3: fc6/MatMul:kernel weights has count 67108864 but 12845056 was expected [07/12/2022-15:29:22] [TRT] [E] UffParser: Parser error: fc6/BiasAdd: The input to the Scale Layer is required to have a minimum of 3 dimensions. 2022-07-12 15:29:22,247 [ERROR] iva.common.export.trt_utils: Failed to parse UFF File File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/export/trt_utils.py", line 301, in _load_from_files mask-conv-l3 (Conv2D) (100, 256, 32, 32) 590080 mask-conv-l2[0][0] __________________________________________________________________________________________________ conv5-mask (Conv2DTranspose) (100, 256, 64, 64) 262400 mask-conv-l3[0][0] __________________________________________________________________________________________________ mask_fcn_logits (Conv2D) (100, 4, 64, 64) 1028 conv5-mask[0][0] __________________________________________________________________________________________________ mask_postprocess (MaskPostproce (1, 100, 64, 64) 0 mask_fcn_logits[0][0] gpu_detections[0][2] __________________________________________________________________________________________________ mask_sigmoid (Activation) (1, 100, 64, 64) 0 mask_postprocess[0][0] ================================================================================================== Total params: 98,303,207 Trainable params: 23,285,632 Non-trainable params: 75,017,575 __________________________________________________________________________________________________ WARNING: To create TensorRT plugin nodes, please use the `create_plugin_node` function instead. WARNING: To create TensorRT plugin nodes, please use the `create_plugin_node` function instead. NOTE: UFF has been tested with TensorFlow 1.14.0. WARNING: The version of TensorFlow installed on this system is not guaranteed to work with UFF. Warning: No conversion function registered for layer: MultilevelCropAndResize_TRT yet. Converting pyramid_crop_and_resize_mask as custom op: MultilevelCropAndResize_TRT Warning: No conversion function registered for layer: ResizeNearest_TRT yet. Converting nearest_upsampling_2 as custom op: ResizeNearest_TRT Warning: No conversion function registered for layer: ResizeNearest_TRT yet. Converting nearest_upsampling_1 as custom op: ResizeNearest_TRT Warning: No conversion function registered for layer: ResizeNearest_TRT yet. Converting nearest_upsampling as custom op: ResizeNearest_TRT Warning: No conversion function registered for layer: SpecialSlice_TRT yet. Converting mrcnn_detection_bboxes as custom op: SpecialSlice_TRT Warning: No conversion function registered for layer: GenerateDetection_TRT yet. Converting generate_detections as custom op: GenerateDetection_TRT Warning: No conversion function registered for layer: MultilevelProposeROI_TRT yet. Converting multilevel_propose_rois as custom op: MultilevelProposeROI_TRT Warning: No conversion function registered for layer: MultilevelCropAndResize_TRT yet. Converting pyramid_crop_and_resize_box as custom op: MultilevelCropAndResize_TRT DEBUG [/usr/local/lib/python3.6/dist-packages/uff/converters/tensorflow/converter.py:96] Marking ['generate_detections', 'mask_fcn_logits/BiasAdd'] as outputs 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/common/export/trt_utils.py", line 301, in _load_from_files AssertionError During handling of the above exception, another exception occurred: 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/export.py", line 12, in File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/export/app.py", line 268, in launch_export File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/export/app.py", line 250, in run_export 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/export/exporter.py", line 654, in export File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/export/trt_utils.py", line 291, in __init__ File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/export/trt_utils.py", line 164, in __init__ File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/export/trt_utils.py", line 309, in _load_from_files AssertionError: UFF parsing failed on line 301 in statement