DMVST-Net
DMVST-Net copied to clipboard
huaxiuyao
Would you update the code anymore?
Hi~ I have read your paper accepted by AAAI 2018, nice work! i just found your code repository for this work, i am very instersted in reproducing your work via real data experiment. could you update your code to show the data format of network inputs?
in detail: the format of input parameters of function build_model?
def build_model(trainX, trainY, testX, testY, trainimage, testimage, traintopo, testtopo, feature_len)
besides, No relevant code can be found about the graph embedding part mentioned in paper. hope to get your new updates ^_^ .
As I shown in the readme, I will update soon. In addition, I simply use LINE to do the network embedding. If you are Chinese, you can add my wechat: Huaxiu_Yao66
@still2009 did you solve for a real world example? Do you have an example ?
I am working on a same problem as this paper described, but using an open source dataset produced by DiDi China from real world taxi calling scenario.
The author has mentioned the dataset before, in README: https://outreach.didichuxing.com/appEn-vue/Personal.
@still2009 Hi, I am also working on this kinds of problem. I visited the URL (https://outreach.didichuxing.com/appEn-vue/Personal) and requested the dataset of Xi'an and Chengdu, but still didn't get the results. I found that you mentioned there is an open source dataset by Didi China. Which dataset do you refer to?
@LeiBAI Hi~ The dataset i mentioned before is exactly what you applied for in the website you listed above. Be patient, your application would probably be accepted just in one day.
hi, i was wondering if anyone understood how to use the model in training? is each point i calculated seperately or is it possible to calculate all points at the same time?
hi, i was wondering if anyone understood how to use the model in training? is each point i calculated seperately or is it possible to calculate all points at the same time?
This model is point independent, and should apply to any (i, t) and corresponding data in your dataset. Is this what you were asking about?
P.S. I recommend you to read the pesudocode in the paper. It's pretty straightfoward.
Hi~ I have read your paper accepted by AAAI 2018, nice work! i just found your code repository for this work, i am very instersted in reproducing your work via real data experiment. could you update your code to show the data format of network inputs?
in detail: the format of input parameters of function
build_model?def build_model(trainX, trainY, testX, testY, trainimage, testimage, traintopo, testtopo, feature_len)besides, No relevant code can be found about the graph embedding part mentioned in paper. hope to get your new updates ^_^ .
hi~recently i read this paper ,and have the same question, if you have solved this problem,could you please tell me something about it? Thank you very much!!!!!
