language2pose
language2pose copied to clipboard
chahuja
Language2Pose: Natural Language Grounded Pose Forecasting
You only give the commands for the second step of the training process, and the commands for training the seq2seq model in the first part are not given.
Hi:) Loading word2vec takes forever (aka a few minutes). Is it supposed to be like that? Is there a way to avoid this loading time? Thanks! https://github.com/chahuja/language2pose/blob/a65d6857d504b5c7cc154260ee946224d387da9d/src/nlp/lstm.py#L121
Hello, First of all, thanks for releasing your work/code. I might found an issue in the code, tell me what you think. In ``data/data.py``line 327, you define ``rifke_dict`` which seems...
why do i get this error :( ``` (torch) Seyeeet@Seyeeet:/media/Seyeeet/Office/language2pose/src$ python dataProcessing/meanVariance.py -mask '[0]' -feats_kind rifke -dataset KITMocap -path2data /media/Seyeeet/Office/language2pose/dataset/kit-mocap -f_new 8 Namespace(angles=[[90]], batch_size=[100], chunks=[1], clean_render=[1], config=[None], cpk=['m'], cuda=[0], curriculum=[0],...
I was not able to achieve the same number as provided in the paper for the evaluation part. is it possible to provide the code for the evaluation metric and...
File "sample_wordConditioned_newSentence.py", line 219, in argparseNloop(sample) File "/mnt/D4567C46567C2AFE/Work HDD/Research/Workflows/Pose-Generation/Repos/language2pose/src/argsUtils.py", line 137, in argparseNloop loop(args, i) File "sample_wordConditioned_newSentence.py", line 112, in sample f_new=f_new) File "/mnt/D4567C46567C2AFE/Work HDD/Research/Workflows/Pose-Generation/Repos/language2pose/src/dataUtils.py", line 81, in __init__ self.datasets...