我正在尝试tensorflow对象检测apis，我能够成功地微调fast_rcnn_resnet101并导出推理文件。
但是，在尝试使用我的模型检测图像时，我收到以下错误。
---------------------------------------------------------------------------
ResourceExhaustedError Traceback (most recent call last)
<ipython-input-19-ec0c1510b78c> in <module>()
30 (boxes, scores, classes, num_detections) = sess.run(
31 [boxes, scores, classes, num_detections],
---> 32 feed_dict={image_tensor: image_np_expanded})
33 print(" classes = "+str(classes)+" scores = "+str(scores)+" num# = "+str(num_detections))
34 print ("vizualizing boxes")
/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.pyc in run(self, fetches, feed_dict, options, run_metadata)
787 try:
788 result = self._run(None, fetches, feed_dict, options_ptr,
--> 789 run_metadata_ptr)
790 if run_metadata:
791 proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)
/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.pyc in _run(self, handle, fetches, feed_dict, options, run_metadata)
995 if final_fetches or final_targets:
996 results = self._do_run(handle, final_targets, final_fetches,
--> 997 feed_dict_string, options, run_metadata)
998 else:
999 results = []
/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.pyc in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)
1130 if handle is None:
1131 return self._do_call(_run_fn, self._session, feed_dict, fetch_list,
-> 1132 target_list, options, run_metadata)
1133 else:
1134 return self._do_call(_prun_fn, self._session, handle, feed_dict,
/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.pyc in _do_call(self, fn, *args)
1150 except KeyError:
1151 pass
-> 1152 raise type(e)(node_def, op, message)
1153
1154 def _extend_graph(self):
ResourceExhaustedError: OOM when allocating tensor with shape[300,14,14,1024]
[[Node: CropAndResize = CropAndResize[T=DT_FLOAT, extrapolation_value=0, method="bilinear", _device="/job:localhost/replica:0/task:0/gpu:0"](FirstStageFeatureExtractor/resnet_v1_101/resnet_v1_101/block3/unit_23/bottleneck_v1/Relu, Reshape_7, Reshape_8/_95, CropAndResize/crop_size)]]
[[Node: SecondStagePostprocessor/BatchMultiClassNonMaxSuppression/map/while/MultiClassNonMaxSuppression/Concatenate/concat_3/_151 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_2971_SecondStagePostprocessor/BatchMultiClassNonMaxSuppression/map/while/MultiClassNonMaxSuppression/Concatenate/concat_3", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"](^_cloopSecondStagePostprocessor/BatchMultiClassNonMaxSuppression/map/while/strided_slice/stack_2/_6)]]
Caused by op u'CropAndResize', defined at:
File "/home/ubuntu/anaconda2/lib/python2.7/runpy.py", line 174, in _run_module_as_main
"__main__", fname, loader, pkg_name)
File "/home/ubuntu/anaconda2/lib/python2.7/runpy.py", line 72, in _run_code
exec code in run_globals
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/__main__.py", line 3, in <module>
app.launch_new_instance()
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/traitlets/config/application.py", line 653, in launch_instance
app.start()
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/kernelapp.py", line 474, in start
ioloop.IOLoop.instance().start()
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/zmq/eventloop/ioloop.py", line 162, in start
super(ZMQIOLoop, self).start()
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/tornado/ioloop.py", line 887, in start
handler_func(fd_obj, events)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/tornado/stack_context.py", line 275, in null_wrapper
return fn(*args, **kwargs)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/zmq/eventloop/zmqstream.py", line 440, in _handle_events
self._handle_recv()
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/zmq/eventloop/zmqstream.py", line 472, in _handle_recv
self._run_callback(callback, msg)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/zmq/eventloop/zmqstream.py", line 414, in _run_callback
callback(*args, **kwargs)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/tornado/stack_context.py", line 275, in null_wrapper
return fn(*args, **kwargs)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/kernelbase.py", line 276, in dispatcher
return self.dispatch_shell(stream, msg)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/kernelbase.py", line 228, in dispatch_shell
handler(stream, idents, msg)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/kernelbase.py", line 390, in execute_request
user_expressions, allow_stdin)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/ipkernel.py", line 196, in do_execute
res = shell.run_cell(code, store_history=store_history, silent=silent)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/ipykernel/zmqshell.py", line 501, in run_cell
return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/IPython/core/interactiveshell.py", line 2717, in run_cell
interactivity=interactivity, compiler=compiler, result=result)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/IPython/core/interactiveshell.py", line 2821, in run_ast_nodes
if self.run_code(code, result):
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/IPython/core/interactiveshell.py", line 2881, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "<ipython-input-5-57652895f483>", line 7, in <module>
tf.import_graph_def(od_graph_def, name='')
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/framework/importer.py", line 311, in import_graph_def
op_def=op_def)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 2506, in create_op
original_op=self._default_original_op, op_def=op_def)
File "/home/ubuntu/anaconda2/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 1269, in __init__
self._traceback = _extract_stack()
ResourceExhaustedError (see above for traceback): OOM when allocating tensor with shape[300,14,14,1024]
[[Node: CropAndResize = CropAndResize[T=DT_FLOAT, extrapolation_value=0, method="bilinear", _device="/job:localhost/replica:0/task:0/gpu:0"](FirstStageFeatureExtractor/resnet_v1_101/resnet_v1_101/block3/unit_23/bottleneck_v1/Relu, Reshape_7, Reshape_8/_95, CropAndResize/crop_size)]]
[[Node: SecondStagePostprocessor/BatchMultiClassNonMaxSuppression/map/while/MultiClassNonMaxSuppression/Concatenate/concat_3/_151 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_2971_SecondStagePostprocessor/BatchMultiClassNonMaxSuppression/map/while/MultiClassNonMaxSuppression/Concatenate/concat_3", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"](^_cloopSecondStagePostprocessor/BatchMultiClassNonMaxSuppression/map/while/strided_slice/stack_2/_6)]]
下面是我的tf代码，它从test_images收集图像并尝试运行它。 我不认为它是一个内存问题因为，我使用的是高配置aws gpu服务器。 此外，我的测试图像尺寸小400x400。
tfconfig = tf.ConfigProto()
tfconfig.gpu_options.allocator_type = 'BFC'
tfconfig.gpu_options.per_process_gpu_memory_fraction = 0.40
tfconfig.gpu_options.allow_growth=True
with detection_graph.as_default():
with tf.Session(graph=detection_graph,config=tfconfig) as sess:
for image_path in TEST_IMAGE_PATHS:
image = Image.open(image_path)
# the array based representation of the image will be used later in order to prepare the
# result image with boxes and labels on it.
print("about to convert image ="+image_path+" into np")
image_np = load_image_into_numpy_array(image)
# Expand dimensions since the model expects images to have shape: [1, None, None, 3]
print ("expanding np ...")
image_np_expanded = np.expand_dims(image_np, axis=0)
print(image_np.shape)
print ("getting image tensor ...")
image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
# Each box represents a part of the image where a particular object was detected.
print ("detecting boxes ...")
boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
# Each score represent how level of confidence for each of the objects.
# Score is shown on the result image, together with the class label.
print ("getting scores, classes and num of detections ...")
scores = detection_graph.get_tensor_by_name('detection_scores:0')
classes = detection_graph.get_tensor_by_name('detection_classes:0')
num_detections = detection_graph.get_tensor_by_name('num_detections:0')
# Actual detection.
print ("building boxes")
(boxes, scores, classes, num_detections) = sess.run(
[boxes, scores, classes, num_detections],
feed_dict={image_tensor: image_np_expanded})
答案 0 :(得分：-1)
发现问题，即使我有足够的CPU内存，我不得不通过杀死在其下运行的进程来清除我的gpu内存（感谢quora https://www.quora.com/How-do-I-kill-all-the-computer-processes-shown-in-nvidia-smi）
(tfpy27) ubuntu@ip-10-0-0-9:~/Object-Detector-App/object_detection$ nvidia-smi
Sun Dec 10 12:29:04 2017
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 375.66 Driver Version: 375.66 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla K80 Off | 0000:00:1E.0 Off | 0 |
| N/A 67C P0 60W / 149W | 4483MiB / 11439MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 8200 C /home/ubuntu/anaconda2/bin/python 4481MiB |
+-----------------------------------------------------------------------------+
sudo kill -9 <PID>
但仍然不确定为什么在一个中等大小的图像上运行会话占用了大量内存