Mr.Crab

Question about the function "eval"?

1

When i run the code to `test_dataset = eval('datasets.'+config.DATASET.DATASET)( root=config.DATASET.ROOT, list_path=config.DATASET.TEST_SET, num_samples=None, num_classes=config.DATASET.NUM_CLASSES, multi_scale=False, flip=False, ignore_label=config.TRAIN.IGNORE_LABEL, base_size=config.TEST.BASE_SIZE, crop_size=test_size, downsample_rate=1)` what the function of eval? i cannot find any information about...

is there any method to compile the tf_ops into dll for the train on windows?

3

@charlesq34 i want to train the model in windows, is there any method to compile the tf_ops into dll for the train on windows?

not found tf_sampling_so.so during the training.

6

@charlesq34 Charles R. Qi, i follow your advice to compile the tf, but when i run the program, the software warns me that the tf_sampling_so.so can not be found and...

No OpKernel was registered to support Op 'FarthestPointSample' used by node

4

@charlesq34 how to use the .so files when i change the ckpt model to pb model? it seems that the .so can not be loaded when i used pb model...

error in NotFoundError: /home/fussen/Work_Area/02_DGASIS/utils/tf_ops/grouping/tf_grouping_so.so: undefined symbol: _ZN10tensorflow6StatusC1ENS_5error4CodeENS_11StringPieceE

1

when i try to run the train.py , i met the problem "NotFoundError: /home/fussen/Work_Area/02_DGASIS/utils/tf_ops/grouping/tf_grouping_so.so: undefined symbol: _ZN10tensorflow6StatusC1ENS_5error4CodeENS_11StringPieceE", i don't know why. The folder "/usr/local/lib" didn't contain the libcublas.so.9.0 and so...

what did nyu40ids represent for in scannet?  i want to train with my dataset, and i met some problem in computing votes. i'm confused about the meanning of nyu40ids....

A problem about tf complie

6

i succeed in compiling the file of .so, but when i run the program of train_val_seg.py, there is a warning that the tf_sampling_so.so could not be found for undefined symbol:...

how git you code?

3

how can i get your code, so i can run it on my local computer?

my image is only 300*800, i find thw iou is poor and i want to imporve the accuracy for segmentation. What should i do to realize it?

Data transform

1

@PingoLH I'm a little confused about the data preprocessing part, how did you determine the mean and standard deviation in tranforming. The code you use just like followes: > value_scale...