Comments (9)
Not sure if it is caused by PyTorch version difference. Can you run it normally with the Harbin dataset?
from deepgtt.
Thank you for your hastly reply, very appreciated.
Unfortunally, I have tried running the Harbin dataset, and it seems to work correctly, so no fault there.
On further inspection of my dataset, and the .h5 files resulting from the gentrain.jl, the 'S' (the tensorcore?) seems to have negative values, this cannot be correct, can it?
from deepgtt.
It should not be negative. This line is hard coded, perhaps you need to change the year value and also rename you file in the format "yy_mm_dd.h5" e.g. "15_10_03.h5".
from deepgtt.
Think you for your reply. I have tried cleaning up my dataset, sorting the dataset according to timestamps, which solved the negative values. Furthermore, I have changed such that the names are foramtted as you mentioned, and are parsed correctly as month and day.
Unfortunaly, I still recieve the same error messages as previously. I was looking into the differences of the Harbin and Chengdu (DiDi) data sets, and it seems the trips of Chengdu are vastly longer, i.e. more gps-points; this should not make a difference though, should it?
from deepgtt.
Also, should the tensor-core have specific dimensions?
from deepgtt.
It should not be a problem with high sampling rates because the preprocessing script will remove the redundant points, removereduantpoints
.
Maybe you can compare the small batch of data returned by the DataLoader
of the two cities.
from deepgtt.
I ended up hardcoding the size of the regions I, O S and C to fit the size of the one created in the Harbin-set (zeros(Float32, 138, 148)
).
I will admit, I am not sure if this will affect the final result of the trained model?
from deepgtt.
@ehvid Excuse me, do you have some results for deepGTT in the Chengdu DiDi dataset? I can share my code, and exchange experiences with you.
from deepgtt.
@NKzhangzhiwen, sorry, unfortunately not. Out project has since ended with the end of the semester. Hope your problems are resolved.
from deepgtt.
Related Issues (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 deepgtt.