'Tensorflow Object Detection API gives Tfrecord files of 0 bytes size?
I am trying to train a custom object detection model using TFOD, following this documentation. When I run the generate_tfrecord.py file, it gives generates tfrecord files of 0 bytes, I don't know what wrong it used to work previously.
# Create train data:
!python generate_tfrecord.py -x /content/drive/MyDrive/training_demo/images/train -l /content/drive/MyDrive/training_demo/annotations/label_map.pbtxt -o /content/drive/MyDrive/training_demo/annotations/train.record
# Create test data:
!python generate_tfrecord.py -x /content/drive/MyDrive/training_demo/images/test -l /content/drive/MyDrive/training_demo/annotations/label_map.pbtxt -o /content/drive/MyDrive/training_demo/annotations/test.record
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|