Models: Error running train.py from lstm_object_detetcion

Created on 14 May 2019 · 10Comments · Source: tensorflow/models

Getting this following error while running train.py from lstm_object_detection.

Traceback (most recent call last):
File "lstm_object_detection/train.py", line 185, in
tf.app.run()
File "/home/kt-ml1/.local/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 125, in run
_sys.exit(main(argv))
File "lstm_object_detection/train.py", line 94, in main
FLAGS.pipeline_config_path)
File "/home/kt-ml1/models-master/models-master/research/lstm_object_detection/utils/config_util.py", line 46, in get_configs_from_pipeline_file
text_format.Merge(proto_str, pipeline_config)
File "/home/kt-ml1/.local/lib/python3.6/site-packages/google/protobuf/text_format.py", line 574, in Merge
descriptor_pool=descriptor_pool)
File "/home/kt-ml1/.local/lib/python3.6/site-packages/google/protobuf/text_format.py", line 631, in MergeLines
return parser.MergeLines(lines, message)
File "/home/kt-ml1/.local/lib/python3.6/site-packages/google/protobuf/text_format.py", line 654, in MergeLines
self._ParseOrMerge(lines, message)
File "/home/kt-ml1/.local/lib/python3.6/site-packages/google/protobuf/text_format.py", line 676, in _ParseOrMerge
self._MergeField(tokenizer, message)
File "/home/kt-ml1/.local/lib/python3.6/site-packages/google/protobuf/text_format.py", line 735, in _MergeField
'that message\'s _pb2 module must be imported as well' % name)
google.protobuf.text_format.ParseError: 18:26 : Extension "object_detection.protos.lstm_model" not registered. Did you import the _pb2 module which defines it? If you are trying to place the extension in the MessageSet field of another message that is in an Any or MessageSet field, that message's _pb2 module must be imported as well

research support

Source

Aaronreb

Most helpful comment

I'm getting the same issue when trying to train on a data set of still images. not sure if that's causing the issue though

Fixed it!
Replace object_detection.protos.lstm_model with lstm_object_detection.protos.lstm_model in the config file.

Aaronreb on 15 May 2019

👍2 🎉1

All 10 comments

I'm getting the same issue when trying to train on a data set of still images. not sure if that's causing the issue though

matt-virgo on 14 May 2019

I'm getting the same issue when trying to train on a data set of still images. not sure if that's causing the issue though

Fixed it!
Replace object_detection.protos.lstm_model with lstm_object_detection.protos.lstm_model in the config file.

Aaronreb on 15 May 2019

👍2 🎉1

That worked, thanks! Now I'm running into issues with my tf_record format. I'm trying to train using the same still image tf_records I had created for the normal object detection retraining.

Do you know of any good guidance on creating the sort of sequence records that this lstm training is expecting? Thanks again!

matt-virgo on 15 May 2019

👍2

That worked, thanks! Now I'm running into issues with my tf_record format. I'm trying to train using the same still image tf_records I had created for the normal object detection retraining.

Do you know of any good guidance on creating the sort of sequence records that this lstm training is expecting? Thanks again!

Well, I'm afraid i am having the same issue!

TypeError: Failed to convert object of type to Tensor. Contents: [, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ]. Consider casting elements to a supported type.

This is the error i am getting.
Any help would be appreciated.
Thanks.

Aaronreb on 16 May 2019

👍1

Will let you know if I figure anything out!

matt-virgo on 16 May 2019

Will let you know if I figure anything out!

Fixed it!
Comment out the line 165 from lstm_object_detection/meta_architectures/lstm_ssd_meta_arch.py

match_list = [matcher.Match(match) for match in tf.unstack(batch_match)]

Instead use the following code-
match_list=batch_match

Aaronreb on 17 May 2019

🎉1

Nice! I'm headed to a conference today so probably won't be able to test anything out until later next week. Were you able to get it to train all the way through on a set of still image tfrecord data?

matt-virgo on 17 May 2019

Nice! I'm headed to a conference today so probably won't be able to test anything out until later next week. Were you able to get it to train all the way through on a set of still image tfrecord data?

Did you solve the sequence tfrecord issue?

Aaronreb on 17 Jun 2019

Hi There,
We are checking to see if you still need help on this, as this seems to be an old issue. Please update this issue with the latest information, code snippet to reproduce your issue and error you are seeing.
If we don't hear from you in the next 7 days, this issue will be closed automatically. If you don't need help on this issue any more, please consider closing this.