Donot generated /test floder about randla-net

[Ailce8862]

@DRosemei @Ailce8862 Can you please check if you have successfully loaded the data?

NO

[v-shetty]

@hansen7 @QingyongHu

even iam trying to use pretrained model for S3DIS dataset

correct me if iam wrong

1. if iam using pretrained model then i dont have to run these lines
   sh jobs_6_fold_cv_s3dis.sh
   python utils/6_fold_cv.py
2. i just run python utils/data_prepare_s3dis.py and i succeded
   but when i run python -B main_S3DIS.py --gpu 0 --mode test --test_area 1
   i got into below result

---

Area_1_office_31_KDTree.pkl 7.4 MB loaded in 0.0s
Area_1_hallway_4_KDTree.pkl 0.7 MB loaded in 0.0s
Area_1_office_15_KDTree.pkl 0.9 MB loaded in 0.0s

Preparing reprojected indices for testing
Area_1_hallway_2 done in 0.0s
Area_1_hallway_5 done in 0.0s
Area_1_office_16 done in 0.0s
Area_1_office_18 done in 0.0s
Area_1_office_28 done in 0.0s
Area_1_office_20 done in 0.0s
Area_1_office_22 done in 0.0s
Area_1_hallway_1 done in 0.0s
Area_1_office_1 done in 0.0s
Area_1_office_14 done in 0.0s
Area_1_office_30 done in 0.0s
Area_1_office_7 done in 0.0s
Area_1_office_3 done in 0.0s
Area_1_office_26 done in 0.0s
Area_1_office_13 done in 0.0s
Area_1_office_5 done in 0.0s
Area_1_pantry_1 done in 0.0s
Area_1_office_11 done in 0.0s
Area_1_office_2 done in 0.0s
Area_1_office_9 done in 0.0s
Area_1_office_21 done in 0.0s
Area_1_office_10 done in 0.0s
Area_1_hallway_8 done in 0.0s
Area_1_conferenceRoom_1 done in 0.0s
Area_1_conferenceRoom_2 done in 0.0s
Area_1_hallway_6 done in 0.1s
Area_1_office_19 done in 0.0s
Area_1_WC_1 done in 0.0s
Area_1_office_17 done in 0.0s
Area_1_hallway_3 done in 0.0s
Area_1_hallway_7 done in 0.0s
Area_1_office_12 done in 0.0s
Area_1_office_4 done in 0.0s
Area_1_office_29 done in 0.0s
Area_1_copyRoom_1 done in 0.0s
Area_1_office_23 done in 0.0s
Area_1_office_8 done in 0.0s
Area_1_office_25 done in 0.0s
Area_1_office_27 done in 0.0s
Area_1_office_6 done in 0.0s
Area_1_office_24 done in 0.0s
Area_1_office_31 done in 0.1s
Area_1_hallway_4 done in 0.0s
Area_1_office_15 done in 0.0s
Initiating input pipelines
WARNING:tensorflow:From /home/vikas/RandLA-Net/RandLANet.py:265: softmax_cross_entropy_with_logits (from tensorflow.python.ops.nn_ops) is deprecated and will be removed in a future version.
Instructions for updating:

Future major versions of TensorFlow will allow gradients to flow
into the labels input on backprop by default.

See tf.nn.softmax_cross_entropy_with_logits_v2.

/home/vikas/anaconda3/envs/randlanet/lib/python3.5/site-packages/tensorflow/python/ops/gradients_impl.py:108: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory.
"Converting sparse IndexedSlices to a dense Tensor of unknown shape. "
results/Log_2020-06-22_11-52-29/snapshots
Model restored from results/Log_2020-06-22_11-52-29/snapshots/snap-26001
Killed

process got killed
let me know anyone can help

[QingyongHu]

Hi @honggesmile, thanks for your interest in our work!

• sh jobs_6_fold_cv_s3dis.sh
  But did not generate the test folder containing the ply file.so I can't continue to run the code.
  The test folder will be created when you evaluate the performance of the trained model on a specific area, e.g.,
  python -B main_S3DIS.py --gpu 0 --mode test --test_area 1
  Can you please give me more details about your problem?

[DRosemei]

@honggesmile I met the same problm with you. @QingyongHu Could you give some advice? I have generate these train_log, results and log*.txt.

[DRosemei]

I have found that this line will create /test folder while it did not worked. It only created results/log*

[DRosemei]

I have solved this problem.
I use python -B main_S3DIS.py --gpu 0 --mode test --test_area 1 --model_path results/Log_2020-03-31_07-53-08/snapshots/snap-33501 to test Area1 and test other areas respectively. But I still do not know why it did not work in "jobs_6_fold_cv_s3dis.sh"

[QingyongHu]

Hi @DRosemei, thanks for your interest in our work.

I have tried to run the code in my pc and it works very well. It is a little bit weird that it cannot generate the test folder on your computer. I suppose it could be caused by the permissions, but anyway, you have already found the way to make it work.

Good luck!

[DRosemei]

@honggesmile Have you generated model like below?

I found that your result path is so weird and it raised an error, "ValueError: attempt to get argmin of an empty sequence".

[Ailce8862]

@honggesmile Have you generated model like below?

I found that your result path is so weird and it raised an error, "ValueError: attempt to get argmin of an empty sequence".
My running results are the same, but I don’t know how to test the results.Can you tell me？

[QingyongHu]

@DRosemei @Ailce8862 Can you please check if you have successfully loaded the data?

[hansen7]

Hi, @honggesmile
I've encountered the same issue, my solution is to change:

RandLA-Net/main_S3DIS.py

Line 16 in 3aafa16

self.path = '/data/S3DIS'

into self.path='data/S3DIS', hope it helps:)

First thing is to check whether you have loaded the data successfully, if it has loaded the data, the console will print the following messages:

Preparing reprojected indices for testing
Area_1_office_1 done in 0.0s
Area_1_hallway_2 done in 0.0s
Area_1_office_24 done in 0.0s
Area_1_hallway_6 done in 0.1s
Area_1_office_2 done in 0.0s
Area_1_pantry_1 done in 0.0s
...

[hansen7]

hi @Vikas-Shetty this is warning, I've encountered, it is not the reason to kill your process

[fenfenglitech]

hi @Vikas-Shetty ,i got the testing process killed when i test on semantickitti sequences 13,19 and 21.have you sloved this problem,could you give me some suggestions?