I'm trying to create a TFlite model using the ssd_mobilenet_v2_quantized_300x300_coco.config, but the following error occurs.
Tensorflow version : 1.14
CUDA version: 10.1
CudNN version : 7.6.5
Python version : 3.7.6
Previously with this version I was able to train successfully with faster_rcnn_inception_v2_pets.config
Windows fatal exception: access violation
Current thread 0x000098a4 (most recent call first):
File "C:\Users\mhtgn\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 648 in __init__
File "C:\Users\mhtgn\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 636 in NewCheckpointReader
File "C:\tensorflow1\models\research\object_detection\utils\variables_helper.py", line 139 in get_variables_available_in_checkpoint
File "C:\tensorflow1\models\research\object_detection\legacy\trainer.py", line 398 in train
File "train.py", line 181 in main
File "C:\Users\mhtgn\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\util\deprecation.py", line 324 in new_func
File "C:\Users\mhtgn\.conda\envs\tensorflow2\lib\site-packages\absl\app.py", line 250 in _run_main
File "C:\Users\mhtgn\.conda\envs\tensorflow2\lib\site-packages\absl\app.py", line 299 in run
File "C:\Users\mhtgn\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\platform\app.py", line 40 in run
File "train.py", line 185 in <module>
I'm trying to create a TFlite model using the ssd_mobilenet_v2_quantized_300x300_coco.config, but the following error occurs.