liyaguang / DCRNN

Implementation of Diffusion Convolutional Recurrent Neural Network in Tensorflow
MIT License
1.17k stars 390 forks source link

building model for different dataset #48

Open b7amine opened 4 years ago

b7amine commented 4 years ago

Hello , I'd like to thank you first for the great work. I'm trying to use the model on a different dataset , which has average occupancy rate per hour , so I have nearly 2000 nodes , location (lat , long) for each node , and records each hour per node , If you could specify what should be modified exactly in order to use this dataset It would be very helpful thank you so much.

b7amine commented 4 years ago

Also I was thinking since my dataset only offer records hourly will the model be suitable for it ? I was thinking to set horizon for 3 and offsets for 3 or something like that since the traffic will not depend on all the 12 previous hours . If you have suggestions on this situation it would be very helpful , I'm thinking of some way to generate data of each 5 minutes based on the average on hour , but I will need a model only for that , the site where I'm getting my dataset provides a curve for the last 16 hours minute by minute in real time and the curve is pretty much the same except for unusual days (holidays etc) . Some of my questions are out of context but I'm a beginner and I'm trying to tackle the trafic forecasting in my city I tried few models before but now I'm trying to establish something professional and your repository is very clear and helpful so if you could help me with this questions I would be very grateful.

ThomasAFink commented 1 year ago

This may help: https://github.com/ThomasAFink/osmnx_adjacency_matrix_for_graph_convolutional_networks