Ilastik giving Not a location (invalid object ID) error

Hi!

I’m using ilastik (ilastik-1.3.3-Linux) boundary based segmentation with multicut. I provide it with the raw data and ilastik generated probabilities. I then chose the watershed parameters and proceed to training by selecting in multiple stacks which edges should be dropped and which should be preserved. After a couple of stacks, when trying to save, I get the “Not a location (invalid object ID) error” message. In the log file I find:

INFO 2019-10-22 10:09:57,879 opEdgeTraining 3708 140544323864320 Converting edge probabilities to dict…
INFO 2019-10-22 10:09:57,972 opEdgeTraining 3708 140544323864320 …done
WARNING 2019-10-22 10:09:58,580 opEdgeTraining 3708 140543375947520 Superpixels changed. Deleting all labels in lane 2.
INFO 2019-10-22 10:09:58,580 opEdgeTraining 3708 140543375947520 Old labels were: {(1, 648): 2, (1, 2927): 2, (1, 3910): 2, (1, 5774): 2, (1, 339): 2, (48, 648): 2, (48, 1858): 2, (2927, 3064): 2, (48, 3471): 2, (2579, 2984): 2, (1, 7290): 2, (6560, 6989): 2, (1, 6848): 1, (1, 4519): 1, (1, 1802): 2, (3910, 3962): 1, (3471, 3910): 2, (1858, 2662): 1, (5109, 5110): 1, (5109, 5868): 1, (6560, 6618): 1, (1800, 1804): 1, (1802, 1858): 1, (2662, 2927): 1, (5004, 5057): 1, (6058, 6629): 2, (1007, 1476): 2, (48, 2984): 2, (4750, 6070): 2, (6070, 6131): 2, (7521, 7970): 2, (7505, 8420): 2, (2799, 2855): 2, (1031, 1635): 2, (2855, 3635): 2, (1, 2): 1, (4282, 5821): 2, (4282, 6105): 2, (5548, 5722): 2, (4282, 5722): 2, (7425, 7547): 2, (6674, 8517): 2, (2326, 3807): 2, (2326, 3911): 2, (5722, 6263): 2, (56, 1101): 2, (1091, 1643): 2, (2, 7088): 2, (2, 5819): 2, (2, 6372): 2, (1, 4517): 2, (1, 3773): 2, (1, 3286): 2, (2, 1992): 2, (2, 1562): 2, (5325, 5722): 2, (1524, 2728): 1, (3435, 4090): 2, (8294, 8375): 2, (8707, 9061): 2, (6801, 8047): 2, (4282, 6742): 2, (7088, 8233): 1, (8191, 8611): 1, (8191, 8233): 1, (5819, 6372): 1, (5283, 5651): 1, (5168, 5229): 1, (3773, 4517): 1, (3772, 3773): 1, (3287, 3410): 1, (4458, 4517): 1, (3410, 3666): 1, (2973, 3010): 1, (3178, 3341): 1, (1992, 1994): 1, (1, 8233): 2, (9012, 9401): 2, (697, 1566): 2, (1101, 1524): 2, (2, 5320): 2, (2, 6257): 2, (1887, 2426): 2, (4, 1887): 2, (2, 4): 2, (2, 1887): 2, (3527, 5013): 2, (1887, 4282): 2, (3527, 4282): 2, (1240, 3527): 2, (5013, 6854): 2, (4282, 5013): 2, (6204, 6801): 2, (8047, 9061): 2, (6757, 8406): 2, (4282, 6204): 1, (4282, 5823): 1, (4282, 5320): 2, (2, 8815): 2, (2, 7730): 2, (6801, 7730): 2, (6801, 7595): 2, (1887, 4164): 1, (4282, 4328): 1, (1887, 4328): 2, (2, 2426): 2, (4456, 4829): 1, (2, 4217): 2, (5001, 5320): 1, (36, 1240): 2, (1240, 1292): 1, (5013, 6757): 2}
INFO 2019-10-22 10:09:59,177 opEdgeTraining 3708 140544332257024 Converting edge probabilities to dict…
INFO 2019-10-22 10:09:59,283 opEdgeTraining 3708 140544332257024 …done
WARNING 2019-10-22 10:09:59,985 opEdgeTraining 3708 140543375947520 Superpixels changed. Deleting all labels in lane 1.
INFO 2019-10-22 10:09:59,986 opEdgeTraining 3708 140543375947520 Old labels were: {(3816, 5222): 2, (5379, 5810): 2, (3822, 4573): 2, (4502, 4880): 2, (3618, 4502): 2, (6775, 9103): 1, (6775, 8798): 2, (6775, 8213): 2, (6775, 7146): 2, (6861, 7303): 2, (6343, 7303): 2, (3181, 4325): 2, (4325, 5674): 2, (4325, 4513): 2, (3833, 4513): 2, (4502, 4513): 2, (3816, 4357): 2, (3592, 4357): 2, (8094, 9060): 1, (7223, 9060): 1, (6354, 7223): 1, (5405, 7223): 1, (5405, 6354): 1, (5405, 6356): 1, (3770, 5405): 1, (3770, 3853): 1, (3451, 3770): 1, (2985, 3451): 1, (2916, 2985): 1, (2916, 3117): 1, (2558, 2916): 1, (3190, 4841): 2, (4841, 5405): 2, (4841, 4973): 2, (3058, 3394): 2, (2, 1124): 2, (1193, 2187): 2, (2187, 2858): 2, (84, 1193): 2, (3614, 4502): 2, (1, 2): 1, (1, 6293): 1, (143, 6951): 1, (143, 321): 1, (4802, 7753): 2, (5640, 8417): 2, (5640, 7753): 2, (5640, 8418): 2, (2929, 2934): 1, (2994, 3404): 1, (2293, 3497): 2, (1539, 2293): 2, (2293, 2545): 2, (1774, 1969): 1, (1593, 1770): 1, (1530, 1591): 1, (1591, 1770): 1, (1591, 1969): 1, (1037, 1775): 1, (1031, 1037): 1, (1384, 1591): 2, (1591, 2088): 2, (2387, 2875): 2, (434, 1695): 2, (104, 434): 2, (434, 1077): 1, (434, 959): 1, (424, 434): 1, (2872, 2941): 1, (2872, 3209): 1, (3654, 3718): 1, (3596, 4215): 2, (3596, 4557): 2, (7740, 8486): 2, (6249, 7566): 2, (3335, 3711): 2, (3335, 4348): 2, (3204, 3335): 2, (5212, 5521): 2, (824, 1943): 2, (824, 947): 2, (74, 824): 2, (824, 927): 1}
INFO 2019-10-22 10:10:00,417 opEdgeTraining 3708 140544332257024 Converting edge probabilities to dict…
INFO 2019-10-22 10:10:00,500 opEdgeTraining 3708 140544332257024 …done
WARNING 2019-10-22 10:10:01,017 opEdgeTraining 3708 140543375947520 Superpixels changed. Deleting all labels in lane 0.
INFO 2019-10-22 10:10:01,017 opEdgeTraining 3708 140543375947520 Old labels were: {(115, 464): 2, (138, 777): 1, (464, 1422): 1, (464, 2452): 1, (464, 2907): 1, (951, 1026): 1, (464, 1339): 1, (464, 951): 1, (464, 3027): 1, (2907, 3288): 2, (2536, 3288): 2, (1011, 1618): 2, (1014, 1552): 1, (2617, 2765): 1, (3467, 3517): 1, (1552, 2617): 1, (1552, 1827): 2, (2152, 2617): 2, (1, 3517): 2, (3517, 4049): 2, (3641, 3964): 2, (1552, 2258): 2, (4705, 5682): 2, (105, 397): 2, (42, 863): 1, (1183, 1229): 1, (2617, 3385): 1, (3517, 3634): 1, (5568, 5650): 1, (5516, 5650): 1, (1927, 3641): 1, (1927, 2675): 1, (1927, 2258): 1, (1370, 1927): 1, (1927, 2685): 2, (3494, 3543): 2, (3494, 3904): 2, (397, 2372): 2, (16, 1014): 2, (1552, 1732): 2, (16, 1053): 2, (1732, 2008): 2, (2152, 2940): 2, (3, 722): 1, (1, 3663): 1, (1, 2008): 1, (3886, 3994): 2, (5704, 6455): 2, (4852, 5219): 2, (4049, 5516): 1, (3725, 3899): 1, (4162, 4206): 1, (1927, 2521): 1, (2233, 2445): 1, (3006, 3047): 1, (2884, 3220): 1, (3720, 4061): 1, (4445, 4633): 1, (3636, 3719): 1, (1927, 3720): 1, (2639, 2733): 1, (2685, 3494): 1, (4239, 4551): 2, (1720, 1994): 2, (2536, 2685): 2, (4554, 5304): 2, (4554, 5682): 2, (41, 717): 1, (41, 3409): 2, (1, 6560): 1, (2, 3): 1, (1, 3): 1, (3663, 3994): 2, (41, 1927): 2, (41, 3174): 2, (41, 2685): 2, (1, 41): 1, (1, 2152): 2, (41, 3208): 2, (41, 3780): 2, (1, 4760): 2, (4049, 5033): 2, (4722, 4774): 2, (41, 5682): 2, (5580, 6063): 2, (5682, 6226): 2}
INFO 2019-10-22 10:10:03,578 opEdgeTraining 3708 140544340649728 Converting edge probabilities to dict…
INFO 2019-10-22 10:10:03,690 opEdgeTraining 3708 140544340649728 …done
INFO 2019-10-22 10:10:03,948 wsDtSegmentation 3708 140545481996032 wsDtSegmentation…
DEBUG 2019-10-22 10:10:03,948 wsDtSegmentation 3708 140545481996032 signed_distance_transform…
DEBUG 2019-10-22 10:10:04,117 wsDtSegmentation 3708 140545481996032 remove_wrongly_sized_connected_components…
DEBUG 2019-10-22 10:10:04,117 wsDtSegmentation 3708 140545481996032 remove_wrongly_sized_connected_components took (0.00 seconds)
DEBUG 2019-10-22 10:10:04,117 wsDtSegmentation 3708 140545481996032 positive distance transform…
DEBUG 2019-10-22 10:10:04,523 wsDtSegmentation 3708 140545481996032 inverting pmap membrane pixels…
DEBUG 2019-10-22 10:10:04,534 wsDtSegmentation 3708 140545481996032 signed_distance_transform took (0.59 seconds)
DEBUG 2019-10-22 10:10:04,534 wsDtSegmentation 3708 140545481996032 binary_seeds_from_distance_transform…
DEBUG 2019-10-22 10:10:04,539 wsDtSegmentation 3708 140545481996032 localMaximaND…
DEBUG 2019-10-22 10:10:04,972 wsDtSegmentation 3708 140545481996032 localMaximaND took (0.43 seconds)
DEBUG 2019-10-22 10:10:04,977 wsDtSegmentation 3708 140545481996032 binary_seeds_from_distance_transform took (0.44 seconds)
DEBUG 2019-10-22 10:10:04,978 wsDtSegmentation 3708 140545481996032 group_seeds_by_distance…
DEBUG 2019-10-22 10:10:04,989 wsDtSegmentation 3708 140545481996032 Number of seed points: 12172
DEBUG 2019-10-22 10:10:08,411 wsDtSegmentation 3708 140545481996032 Extracting connected seeds…
DEBUG 2019-10-22 10:10:08,438 wsDtSegmentation 3708 140545481996032 … took 0.026305 seconds
DEBUG 2019-10-22 10:10:08,441 wsDtSegmentation 3708 140545481996032 group_seeds_by_distance took (3.46 seconds)
DEBUG 2019-10-22 10:10:08,443 wsDtSegmentation 3708 140545481996032 iterative_inplace_watershed…
DEBUG 2019-10-22 10:10:09,545 wsDtSegmentation 3708 140545481996032 remove_wrongly_sized_connected_components…
DEBUG 2019-10-22 10:10:09,595 wsDtSegmentation 3708 140545481996032 remove_wrongly_sized_connected_components took (0.05 seconds)
DEBUG 2019-10-22 10:10:10,013 wsDtSegmentation 3708 140545481996032 Max Watershed Label: 5174
DEBUG 2019-10-22 10:10:10,013 wsDtSegmentation 3708 140545481996032 iterative_inplace_watershed took (1.57 seconds)
INFO 2019-10-22 10:10:10,013 wsDtSegmentation 3708 140545481996032 wsDtSegmentation took (6.07 seconds)

This has happened multiple times. I have tried choosing several edges without turning “Live Predict” on, and also the opposite (Live Predict on all the time).

Does anyone know if there is there anything I could be doing wrong to cause this?

Hi @Buono,

this looks like a bug, but I haven’t seen this one before. The content from the log-file, however does look fairly normal. Did a window pop up or something? How many images did you have? Is this something one can produce with just two lanes?

Cheers
Dominik

I do get a window saying:
“Not a location (invalid object ID)”
And telling me to check the log file (that I posted).
The difficult thing is that I don’t know how to reproduce the problem. It eventually happens.
I do have 19 z-stacks of 10-25MB each copied into the file. These are chunks of a larger dataset.

I ended up downloading the beta (ilastik-1.3.3b2-Linux) and am using it. That error did not happen again. But now, from time to time, and for reasons I cannot reproduce, the software will freeze when I am training the multicut. When this happens, I have to kill it and start from a previous saving point. No error messages in this case.