DeepSpeech testing Error

Hi,
Deepspeech training is done well, then durng testing, below error happened.
Can I know the reason??

Use standard file APIs to check for files with this prefix.
INFO:tensorflow:Restoring parameters from data/Zeroth/clips/ckpt4/best_dev-1219
I0131 15:29:07.220100 139722287343424 saver.py:1280] Restoring parameters from data/Zeroth/clips/ckpt4/best_dev-1219
I Restored variables from best validation checkpoint at data/Zeroth/clips/ckpt4/best_dev-1219, step 1219
Testing model on data/Zeroth/clips/test.csv
Test epoch | Steps: 0 | Elapsed Time: 0:00:00                                                                                                                                                                                               Fatal Python error: Segmentation fault

Thread 0x00007f13a14de740 (most recent call first):
  File "/home/beomgon2/deepspeech-train-venv/lib/python3.6/site-packages/ds_ctcdecoder/__init__.py", line 116 in ctc_beam_search_decoder_batch
  File "/home/beomgon2/DeepSpeech/evaluate.py", line 124 in run_test
  File "/home/beomgon2/DeepSpeech/evaluate.py", line 180 in evaluate
  File "./DeepSpeech.py", line 698 in test
  File "./DeepSpeech.py", line 967 in main
  File "/home/beomgon2/deepspeech-train-venv/lib/python3.6/site-packages/absl/app.py", line 250 in _run_main
  File "/home/beomgon2/deepspeech-train-venv/lib/python3.6/site-packages/absl/app.py", line 299 in run
  File "./DeepSpeech.py", line 990 in <module>

I am testing in Ubuntu 18.04, cuda10.0, cudnn7.5, and tensorflow 1.14-gpu(1.15 has same problem)

thanks

You can’t mix versions, so you need to use matching tensorflow version for the deepspeech you are using. 0.6.1 is r1.14, current master is r1.15.

Segfault at test step is likely mismatched ds_ctcdecoder python wheel, please ensure you use the proper version.