SSD-Tensorflow icon indicating copy to clipboard operation
SSD-Tensorflow copied to clipboard

NotFoundError: Failed to create a NewWriteableFile

Open JinwenJay opened this issue 7 years ago • 5 comments

I tried to convert the VOC2007 datasets to TF-Records using the tf_convert_data.py like this:

python tf_convert_data.py --dataset_name=pascalvoc --dataset_dir=VOCdevkit/VOC2007/test/ --output_name=voc_2007_test --output_dir=VOCdevkit/tfrecords

And I had download the VOC2007 datasets in the corresponding path. Following is the problem I am confused with.

Dataset directory: VOCdevkit/VOC2007/test/ Output directory: VOCdevkit/tfrecords Traceback (most recent call last): File "tf_convert_data.py", line 59, in tf.app.run() File "C:\Users\Jinwen Tan\Anaconda3\lib\site-packages\tensorflow\python\platform\app.py", line 48, in run _sys.exit(main(_sys.argv[:1] + flags_passthrough)) File "tf_convert_data.py", line 54, in main pascalvoc_to_tfrecords.run(FLAGS.dataset_dir, FLAGS.output_dir, FLAGS.output_name) File "C:\Users\Jinwen Tan\Documents\SSD-tensorflow\SSD-Tensorflow-master\datasets\pascalvoc_to_tfrecords.py", line 210, in run with tf.python_io.TFRecordWriter(tf_filename) as tfrecord_writer: File "C:\Users\Jinwen Tan\Anaconda3\lib\site-packages\tensorflow\python\lib\io\tf_record.py", line 106, in init compat.as_bytes(path), compat.as_bytes(compression_type), status) File "C:\Users\Jinwen Tan\Anaconda3\lib\site-packages\tensorflow\python\framework\errors_impl.py", line 473, in exit c_api.TF_GetCode(self.status.status)) tensorflow.python.framework.errors_impl.NotFoundError: Failed to create a NewWriteableFile: VOCdevkit/tfrecords/voc_2007_test_000.tfrecord : \u03f5\u0373\udcd5\u04b2\udcbb\udcb5\udcbd\u05b8\udcb6\udca8\udcb5\udcc4·\udcbe\udcb6\udca1\udca3 ; No such process

Besides, I run this code in windows cmd, does it matter? I mean whether this problem will disappear if I use ubuntu?

JinwenJay avatar Jan 14 '18 04:01 JinwenJay

I know the recommended form in https://github.com/balancap/SSD-Tensorflow like this:DATASET_DIR=./VOC2007/test/ OUTPUT_DIR=./tfrecords python tf_convert_data.py
--dataset_name=pascalvoc
--dataset_dir=${DATASET_DIR}
--output_name=voc_2007_train
--output_dir=${OUTPUT_DIR} But I use windows, it seems I can't run the script in this way.

JinwenJay avatar Jan 14 '18 04:01 JinwenJay

hey! Have you solve it? I have the same question as you.

dmy1997 avatar Jul 27 '18 09:07 dmy1997

You need to creat a dir by yourself, like VOCdevkit/tfrecords, at your running environment folder where you run this script. @dmy1997

liruixl avatar Jul 28 '18 13:07 liruixl

have the same question have you solved it?

murraykkkeed06 avatar Jul 05 '20 04:07 murraykkkeed06

You need to creat a dir by yourself, like VOCdevkit/tfrecords, at your running environment folder where you run this script. @dmy1997

i did add the folder, but has the same error again

murraykkkeed06 avatar Jul 05 '20 04:07 murraykkkeed06