我正在尝试通过跟随tensorflow对象检测API的running locally选项来训练对象检测模型。 遵循文档后:
INFO:tensorflow:开始会话。
INFO:tensorflow:将检查点保存到路径./model/train_logs/model.ckpt 信息:tensorflow:启动队列。
INFO:tensorflow:向协调员报告错误:,〜/ Documents / Projects / models / object_detection / data / pascal_train.record
[[Node:parallel_read / ReaderReadV2_2 = ReaderReadV2 [_device =" / job:localhost / replica:0 / task:0 / cpu:0"](parallel_read / TFRecordReaderV2_2,parallel_read / filenames)]]
INFO:tensorflow:global_step / sec:0 2017-08-29 11:41:59.852783:W tensorflow / core / framework / op_kernel.cc:1192]超出范围:FIFOQueue' _5_prefetch_queue'关闭且元素不足(请求1,当前大小0) [[Node:prefetch_queue_Dequeue = QueueDequeueV2component_types = [DT_INT32,DT_FLOAT,DT_INT32,DT_BOOL,DT_INT32,DT_BOOL,DT_INT32,DT_INT32,DT_STRING,DT_FLOAT,DT_FLOAT,DT_INT3 2,DT_INT64,DT_STRING,DT_INT32,DT_INT32,DT_INT32,DT_INT64,DT_INT32,DT_INT64,DT_STRING,DT_INT32],timeout_ms = -1,_device =" / job:localhost / replica:0 / task:0 / cpu:0& #34;]] 2017-08-29 11:41:59.852819:W tensorflow / core / framework / op_kernel.cc:1192]超出范围:FIFOQueue' _5_prefetch_queue'关闭且元素不足(请求1,当前大小0) [[Node:prefetch_queue_Dequeue = QueueDequeueV2component_types = [DT_INT32,DT_FLOAT,DT_INT32,DT_BOOL,DT_INT32,DT_BOOL,DT_INT32,DT_INT32,DT_STRING,DT_FLOAT,DT_FLOAT,DT_INT3 2,DT_INT64,DT_STRING,DT_INT32,DT_INT32,DT_INT32,DT_INT64,DT_INT32,DT_INT64,DT_STRING,DT_INT32],timeout_ms = -1,_device =" / job:localhost / replica:0 / task:0 / cpu:0& #34;]] 2017-08-29 11:41:59.852797:W tensorflow / core / framework / op_kernel.cc:1192]超出范围:FIFOQueue' _5_prefetch_queue'关闭且元素不足(请求1,当前大小0) 。 。 。
[[Node:prefetch_queue_Dequeue = QueueDequeueV2component_types = [DT_INT32,DT_FLOAT,DT_INT32,DT_BOOL,DT_INT32,DT_BOOL,DT_INT32,DT_INT32,DT_STRING,DT_FLOAT,DT_FLOAT,DT_INT32,DT_INT64,DT_STRING,DT_INT32,DT_INT32,DT_INT32,DT_INT64,DT_INT32, DT_INT64,DT_STRING,DT_INT32],timeout_ms = -1,_device =" / job:localhost / replica:0 / task:0 / cpu:0"]] 2017-08-29 11:42:00.353191:W tensorflow / core / framework / op_kernel.cc:1192]超出范围:FIFOQueue' _5_prefetch_queue'关闭且元素不足(请求1,当前大小0) [[Node:prefetch_queue_Dequeue = QueueDequeueV2component_types = [DT_INT32,DT_FLOAT,DT_INT32,DT_BOOL,DT_INT32,DT_BOOL,DT_INT32,DT_INT32,DT_STRING,DT_FLOAT,DT_FLOAT,DT_INT32,DT_INT64,DT_STRING,DT_INT32,DT_INT32,DT_INT32,DT_INT64,DT_INT32,DT_INT64,DT_STRING ,DT_INT32],timeout_ms = -1,_device =" / job:localhost / replica:0 / task:0 / cpu:0"]] 2017-08-29 11:42:00.353105:W tensorflow / core / framework / op_kernel.cc:1192]超出范围:FIFOQueue' _5_prefetch_queue'关闭且元素不足(请求1,当前大小0) [[Node:prefetch_queue_Dequeue = QueueDequeueV2component_types = [DT_INT32,DT_FLOAT,DT_INT32,DT_BOOL,DT_INT32,DT_BOOL,DT_INT32,DT_INT32,DT_STRING,DT_FLOAT,DT_FLOAT,DT_INT32,DT_INT64,DT_STRING,DT_INT32,DT_INT32,DT_INT32,DT_INT64,DT_INT32,DT_INT64,DT_STRING ,DT_INT32],timeout_ms = -1,_device =" / job:localhost / replica:0 / task:0 / cpu:0"]]
INFO:tensorflow:捕获OutOfRangeError。停止训练。
信息:tensorflow:完成培训!将模型保存到磁盘。
追踪(最近一次通话): 文件" object_detection / train.py",第199行,in tf.app.run()
文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/python/platform/app.py" ;,第48行,在运行中
_sys.exit(main(_sys.argv[:1] + flags_passthrough))
文件" object_detection / train.py",第195行,主要
worker_job_name, is_chief, FLAGS.train_dir)
文件" / home / skulhare / Documents / Projects / models / object_detection
/trainer.py" ;,第296行,在火车上
saver=saver)
文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/contrib/slim/python/slim/learning.py" ;,第767行,列车
sv.stop(threads, close_summary_writer=True)
文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/python/training/supervisor.py" ;,第792行,停止 stop_grace_period_secs = self._stop_grace_secs)
文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/python/training/coordinator.py",第389行,在加入 six.reraise(* self._exc_info_to_raise)
文件" /home/skulhare/.local/lib/python3.5/site-packages/six.py" ;,第686行,重新加入 提高价值 文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/python/training/queue_runner_impl.py" ;,第238行,在_run
enqueue_callable()
文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/python/client/session.py",第1235行,在_single_operation_run target_list_as_strings,status,None)
文件" /usr/lib/python3.5/contextlib.py",第66行,退出 下一个(self.gen) 文件" /home/skulhare/.local/lib/python3.5/site-packages/tensorflow/python/framework/errors_impl.py",第466行,在raise_exception_on_not_ok_status中 pywrap_tensorflow.TF_GetCode(状态))
tensorflow.python.framework.errors_impl.NotFoundError:〜/ Documents / Projects / models / object_detection / data / pascal_train.record [[Node:parallel_read / ReaderReadV2_2 = ReaderReadV2 [_device =" / job:localhost / replica:0 / task:0 / cpu:0"](parallel_read / TFRecordReaderV2_2,parallel_read / filenames)]]
我一直在Ubuntu 16.04上使用tensorflow 1.3.0和GTX 1080 ti。 Here是faster_rcnn_resnet101.config文件的内容。
提前致谢。