-
Notifications
You must be signed in to change notification settings - Fork 50
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Run the model on scans from ScanNet #5
Comments
@deli4iled Hi/ have you figured out how to generate tfrecord files for ScanComplete running? I've tried to train this Net using different dataset, but Have some error in my tfrecord files: |
@deli4iled @angeladai |
I've found the error in my code for tfrecords generation it was in using np array of ints to fill bytes feature in the record. ^) |
@KyriaAnnwyn Hi, I'm working on this net with my own data, but I can't find the way to transfer my own data to the tfrecords, could you please share the code? Thanks. |
@livc Here is my code for tfrecords creation: _RESOLUTIONS = ['5cm', '9cm', '19cm'] def float_feature(value): def int64_feature(value): def bytes_feature(value): def createTFRecord(args, pair, number): key_input = RESOLUTIONS[hierarchy_level - 1] + '' + _INPUT_FEATURE #read full and hole tsdfs ans save then to example center = [insample.shape[0]/2, insample.shape[1]/2, insample.shape[2]/2] #cut areas with specific size fron the input tsdfs, consider that the model shoul be somewhere in the center for numsample in xrange(10*args.batch_size):
writer.close() def convertList(args): for name in tsdfNames: for pair in name_pairs: def parse_arguements(argv): parser.add_argument('--infolder', type=str, help='Folder with input *.npy tsdfs', return parser.parse_args(argv) if name == 'main': |
@KyriaAnnwyn Thank you! |
thank you @KyriaAnnwyn for sharing this code. I got a question about how can i get numpy array of tsdfs ? |
hello, |
@qiji77 |
@KyriaAnnwyn can u please share the code file to convert obj file to tfrecords because I'm getting lot of errors with the above code. |
Hi, I'd like to run the inference on ScanNet dataset, but I can't figure out how to generate the tfrecords. Could you please share the code to generate them on ScanNet dataset? Thanks
The text was updated successfully, but these errors were encountered: