PSI-PROF weekly submission - 24 Mar 2024

[jturtle]

No description provided.

[kjsato]

Thanks for your submissions. I'd like to notify that your model-output file (including the former submissions) data is 1w behind our expected origin_date, and the output is not reflected on website(2024-03-24 for this week). We would appreciate it if you could reflect that in the data one week ahead,

Thank you,

[kjsato]

approved to merge

[jturtle]

Hi Koji, Sorry, I did not originally see this message. Thank you for reaching out. In our most recent submission, we used all available data (through March 16) and thus set origin_date to March 17. Is this incorrect? Are you saying that origin_date should be March 24? If this is the case, is horizon=0 the week containing March 17 or the week containing March 24? Thank you in advance for the clarifications, Jamie

…

On Mon, Mar 25, 2024 at 2:14 PM Koji SATO ***@***.***> wrote: Thanks for your submissions. I'd like to notify that your model-output file (including the former submissions) data is 1w behind our expected origin_date, and the output is not reflected on website(2024-03-24 for this week). We would appreciate it if you could reflect that in the data one week ahead, Thank you, — Reply to this email directly, view it on GitHub <#33 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVOD26IFWVLNZSLYF43MG3Y2CHUNAVCNFSM6AAAAABFGKLKRGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMJYHEZDSNJZGM> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[kjsato]

@jturtle sorry for my late response and the inconvienence. Yes, in this week's submission (2024-3-24), we would have expected to have data with origin_date 2024-03-24 of which filename would be 2024-03-24-PSI-PROF.parquet, and it contains the week from 03-24 with horizon=0 in your case.

In this case, at the time of data processing, filename was a date in the past (if origin_date was a date in the past, the file name would be the same of course), so it was excluded from processing.
At least filename and data cannot go back in time, so I would appreciate it if you could set filename (e.g, 2024-03-24-PSI-PROF.parquet) and the data to match this week's (with origin_date "2024-03-24").

Thank you,

[jturtle]

Thank you for the clarifications. I have just submitted a pull request that includes a properly dated forecast for this week. Our pipeline has been updated, so next week's forecast should be on-time and properly dated. Best, Jamie

…

On Fri, Mar 29, 2024 at 7:03 AM Koji SATO ***@***.***> wrote: @jturtle <https://github.com/jturtle> sorry for my late response and the inconvienence. Yes, in this week's submission (2024-3-24), we would have expected to have data with origin_date 2024-03-24 of which filename would be 2024-03-24-PSI-PROF.parquet, and it contains the week from 03-24 with horizon=0 in your case. In this case, at the time of data processing, filename was a date in the past (if origin_date was a date in the past, the file name would be the same of course), so it was excluded from processing. At least filename and data cannot go back in time, so I would appreciate it if you could set filename (e.g, 2024-03-24-PSI-PROF.parquet) and the data to match this week's (with origin_date "2024-03-24"). Thank you, — Reply to this email directly, view it on GitHub <#33 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVOD2653LPRWNDBDVDKQSLY2VYCXAVCNFSM6AAAAABFGKLKRGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMRXGI4TANZRG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>