Fatal data load error

[karlienfranco]

Fatal data load error: "Something went wrong while fetching destroy/solutions.tsv" for destroy models

[AntheSevenants]

Can you send me your dataset? I'll try to look into why this is happening.

[karlienfranco]

I didn't use an external one, but an internal one that should be available at: https://qlvl.github.io/NephoVis/#model/destroy Op wo 25 jan. 2023 om 15:10 schreef Anthe Sevenants < ***@***.***>:

…

Can you send me your dataset? I'll try to look into why this is happening. — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQS5BUX32JNWVLDVKDWFXITWUEX5XANCNFSM6AAAAAAT43MDGU> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[AntheSevenants]

I think the destroy lemma was part of one of Mariana's early tests. It is no longer included in the overview because the file structure evolved since then. Where did you get the link from?

[karlienfranco]

I see! The link is contained in a chapter that we published actually. Is it impossible to make it work again? Op wo 25 jan. 2023 om 18:58 schreef Anthe Sevenants < ***@***.***>:

…

I think the *destroy* lemma was part of one of Mariana's early tests. It is no longer included in the overview <https://qlvl.github.io/NephoVis/> because the file structure evolved since then. Where did you get the link from? — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQS5BUU3PDJ3YRW6OSEJLT3WUFSTRANCNFSM6AAAAAAT43MDGU> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[AntheSevenants]

I'm working on getting it fixed! It's been a while since I've touched the code but I hope to get it working 🙂.

[AntheSevenants]

OK through some miracle I fixed it. For some reason though, some models for destroy (probably also for the other legacy datasets) do not have any defined coordinates in the tokens dataset. The models that do have defined coordinates now work fine. I'll create a pull request soon. I guess it's then up to Mariana to merge it (I don't know who else has write permissions).

[karlienfranco]

Great! I think the defined coordinates are only a problem in the destroy models - we only created tokens datasets for a subset (the medoids) if I recall correctly. Thanks for your help!