Need testers for ES code (multiple model chaining/ arbitrary frames)

[asker]

The new version of ES support model chaining as well as inspecting multiple frames in an event.
I am looking for folks to:

1. Update to the new ES
2. Test if current operation works without making any config changes (that is, new code with old functionality)
3. Test new functionality.

Specifically, I want to know if folks update to the new ES and face any errors (without making changes to config)

Steps are here: https://github.com/pliablepixels/zmeven ... -749162186

Would appreciate feedback

[Magic919]

I'm in.

Have pulled and installed latest. No othe changes, just a restart of ZMES.

[asker]

I'm in.

Have pulled and installed latest. No othe changes, just a restart of ZMES.

Thanks! I've added the following note to Part I - when you test with no changes:

Specifically, make sure all your objectconfig changes you might have implemented earlier, like monitor specific waits, detection_pattern etc. continue to work with the changed values

[Magic919]

I haven't seen a single known face since the upgrade. Deleted my old faces.dat, but no help. Trying to retrain threw an error.

Tried it against an old alarm that 'knew' my face and got

Code: Select all

01/02/21 14:27:40 zmesdetect_mm9[3746936] FAT zm_detect.py:860 [Unrecoverable error:int() can't convert non-string with explicit base Traceback:Traceback (most recent call last):
  File "/var/lib/zmeventnotification/bin/zm_detect.py", line 857, in <module>
    main_handler()
  File "/var/lib/zmeventnotification/bin/zm_detect.py", line 348, in main_handler
    m = face.Face(logger=g.logger, options=g.config, upsample_times=g.config['face_upsample_times'],
  File "/usr/local/lib/python3.8/dist-packages/pyzm/ml/face.py", line 92, in __init__
    train.FaceTrain(options=self.options).train()
  File "/usr/local/lib/python3.8/dist-packages/pyzm/ml/face_train.py", line 35, in train
    num_jitters = int(self.options.get('face_num_jitters'),0)
TypeError: int() can't convert non-string with explicit base
]

[asker]

whoops - fixed in 0.3.19 pyzm

[Magic919]

Ok, just trained faces successfully. Will check on alarms shortly.

Thanks.

[Magic919]

I've emailed over some debug as I'm still getting the unknown faces, even on old events that showed as known.

[Magic919]

Thanks for the rapid fix, back on track now.

[tsp84]

Upgrading to pyzm 0.3.19 fixes that face detection error?

[Magic919]

Yes, it did.

[warcanoid]

How to fix this, have same erro with object in docker?

Code: Select all

root@16faf02ec93a:/# pip3 install pyzmq 0.3.19
Requirement already satisfied: pyzmq in /usr/local/lib/python3.8/dist-packages (20.0.0)
ERROR: Could not find a version that satisfies the requirement 0.3.19 (from versions: none)
ERROR: No matching distribution found for 0.3.19

Code: Select all

2021-01-05 17:31:22	zmesdetect_m5		1231	FAT	Unrecoverable error:'ml_sequence' Traceback:Traceback (most recent call last): File "/var/lib/zmeventnotification/bin/zm_detect.py", line 531, in <module> main_handler() File "/var/lib/zmeventnotification/bin/zm_detect.py", line 295, in main_handler if g.config['ml_sequence'] and g.config['use_sequence'] == 'yes':KeyError: 'ml_sequence'	zm_detect.py	534

[Magic919]

It’s not pyzmq.

[warcanoid]

added new config for ES in objectconfig.ini:

[ml]
use_sequence = yes
disable_locks= no
stream_sequence = {
'frame_strategy': 'most_models',
'frame_set': 'snapshot,alarm',
'contig_frames_before_error': 5,
'max_attempts': 3,
'sleep_between_attempts': 4,
'resize':800

}

ml_sequence= {
'general': {
'model_sequence': 'object'

},
'object': {
'general':{
'pattern':'(car|truck|bicycle|motorbike|bus|person|dog|cat)',
'same_model_sequence_strategy': 'first' # also 'most', 'most_unique's
},
'sequence': [{

# YoloV4 on GPU if TPU fails (because sequence strategy is 'first')
'object_config':'{{base_data_path}}/models/yolov4/yolov4.cfg',
'object_weights':'{{base_data_path}}/models/yolov4/yolov4.weights',
'object_labels': '{{base_data_path}}/models/yolov4/coco.names',
'object_min_confidence': 0.3,
'object_framework':'opencv',
'object_processor': 'gpu'
}]
}
Also tried to disable use_sequence: no, but dont work.

[Magic919]

Did you do the update of pyzm?

[warcanoid]

yes