Comments (8)
Hi,
First, as mentioned, you can always use your own data loader.
The shape of the dataset I prepared is (samples, time, joints, dimension), which should be something like (38086, 300, 50, 3). '300' means the length of the sequence, where the true length is indicated in the file lenname=datasets+'_len.npy'. '50' is the number of joints and '3' is the xyz dimensions.
It should also be able to be derived from the relevant code.
data_handle=np.load(dataname) line 23
dataset=data_handle[shufflevideoindex] line 79
sample=dataset[startid:startid+self.seq_len] line 86
from indrnn_pytorch.
Thanks alot for your answer.. I'm having few more questions. Sorry I'm still in the learning phase. when considers '50' the maximum number of joints in a given time frame is 25. But since it is a 3D shaped image, in total there are 75 (25x3) columns. In your example may I know why you took it as 50. Is it like (25x2), which means 50 joints are feeding to the network? Does 300 means 300 data rows are feeding to the network. Here I elaborate it in the following image in more details. Does 38086 means total number of rows taking as the sample size? Following image shows the .mat file of a particular person. So to make 38086, we would combine the time frames of multiple people and creating a single dataframe? In the following image there were 4450 rows. By combining more .mat files, I'm planning to make a large sample. Can you correct if I'm wrong.
from indrnn_pytorch.
Hi,
50 joints meaning two persons with each person having 25 joints.
from indrnn_pytorch.
Thanks alot :)
from indrnn_pytorch.
Hi, Is there any possibility to attach your dataname=datasets+'.npy'
labelname=datasets+'_label.npy'
lenname=datasets+'_len.npy'
files here? I really appreciate if it is possible. And also I'm not clear what the lenname array looks like. Kindly help
from indrnn_pytorch.
Hi,
The file is too large.
As I mentioned repeatedly, you don't have to use my data loader. You can just write one of your own according to you understanding.
datasets+'_len.npy' is used because the lengths of different samples are different.
datasets+'_label.npy' apparently are the labels.
from indrnn_pytorch.
Hi,
Thank you verymuch for sharing your knowledge.
I downloaded the NTU skeleton dataset and the files are in .skeleton format.
Did you convert that .skeleton files to np array inorder to make inputs ?
Thank you.
from indrnn_pytorch.
Hi, @WangMuT
Yes.
from indrnn_pytorch.
Related Issues (11)
- Question about action recognition experiment on this pytorch implementation!!!! HOT 12
- Do not work on pytorch 1.9
- LSTM instead of RNN
- issue on grad
- Not abel to use this on packed sequence , AttributeError: 'PackedSequence' object has no attribute 'size' HOT 1
- Not able to reproduce the Results HOT 3
- Input sequence padding HOT 4
- Word level PTB repo? HOT 5
- Settings to reproduce resIndRNN results on PTB HOT 4
- cuda version of IndRNNCell? HOT 7
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from indrnn_pytorch.