osmnx VS pytorch_geometric_temporal

You want osmnx

OSM road network data was analyzed using osmnx (https://github.com/gboeing/osmnx). Plotted using QGIS.

I think you’re hitting this issue right now: https://github.com/gboeing/osmnx/issues/832

RDS-TMC transmits traffic data over FM radio. https://en.wikipedia.org/wiki/Traffic_message_channel Some is encrypted, some not.

It might be possible to model traffic with OSMnx to assign weights to roads by expected traffic levels https://github.com/gboeing/osmnx

Depending on where you live you might be able to get traffic data and maybe traffic models from public authorities using Freedom of Information requests.

Osmnx might also be a way to download data for the whole country. It breaks the area up in smaller parts automatically and also downloads the parts from Overpass. https://github.com/gboeing/osmnx

You can make your own with a variety of tools, but I like the results from OSMnx offhand.

I have no idea if this works for such large regions, but https://geoffboeing.com/publications/osmnx-complex-street-networks/ is an awesome tool to visualize street (and maybe bike) networks.

I have done this a few times now. Alone (e.g. https://github.com/paulmorio/geo2dr) and in collaboration with others (e.g. https://github.com/benedekrozemberczki/pytorch_geometric_tem...) primarily as a way to learn about the methods I was interested in from a research perspective whilst improving my skills in software engineering. I am still learning.

Starting out I would recommend implementing fundamental building blocks within whatever 'subculture' of ML you are interested in whether that be DL, kernel methods, probabilistic models, etc.

Let's say you are interested in deep learning methods (as that's something I could at least speak more confidently about). In that case build yourself an MLP layer, then an RNN layer, then a GNN layer, then a CNN layer, and an attention layer along with some full models with those layers on some case studies exhibiting different data modalities (images, graphs, signals). This should give you a feel for the assumptions driving the inductive biases in each layer and what motivates their existence (vs. an MLP). It also gives you the all the building blocks you can then extend to build every other DL layer+model out there. Another reason is that these fundamental building blocks have been implemented many times so you have a reference to look to when you get stuck.

On that note: here are some fun GNN papers to implement in order of increasing difficulty (try building using vanilla PyTorch/Jax instead of PyG).

Repo: https://github.com/benedekrozemberczki/pytorch_geometric_temporal