![](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F9360422%2Fd1e6dec41830216ab4e67bd8e6dd9a59%2F.png?generation=1709742806813673&alt=media\")
Hi,i have a small question regarding the dataset in this Kaggle competition. The time series dataset is available in two formats (CSV and cube data). I believe there might be a mismatch in the information provided for the cube dataset dimensions. In the data explanation, the time series cube dimensions are specified as (BAND, YEAR, QUARTER), while the monthly cube dimensions are specified as (RASTER-TYPE, YEAR, MONTH). However, the time series tensor cube data shape is torch.Size([6, 4, 21]), and the monthly cube data shape is torch.Size([4, 19, 12]). I assume that these dimensions correspond to time series (6 bands, 4 quarters, 21 years) and monthly (4 raster types, 19 years, 12 months).
\nWhat are your thoughts on this issue?
\n
![](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F9360422%2Fb7cd0270effbaabe1ecd0642e87b82be%2F.png?generation=1709742957090412&alt=media\")
Hi,i have a small question regarding the dataset in this Kaggle competition. The time series dataset is available in two formats (CSV and cube data). I believe there might be a mismatch in the information provided for the cube dataset dimensions. In the data explanation, the time series cube dimensions are specified as (BAND, YEAR, QUARTER), while the monthly cube dimensions are specified as (RASTER-TYPE, YEAR, MONTH). However, the time series tensor cube data shape is torch.Size([6, 4, 21]), and the monthly cube data shape is torch.Size([4, 19, 12]). I assume that these dimensions correspond to time series (6 bands, 4 quarters, 21 years) and monthly (4 raster types, 19 years, 12 months).
\nWhat are your thoughts on this issue?
\n
Hi there,
\nThank you for pointing this out. You are right about the \"mismatch\".
\nHowever, when I wrote it, I did not mean to write it in the actual order, but, now I understand it might be misleading. I did update it based on your feedback.
Just for clarification, the order is as follows:
\nLandsat: BAND, QUARTER, and YEAR
\nClimate: RASTER-TYPE, YEAR, and MONTH
Let us know in case you will need any assistance!
\nBest,
\nLukas
Thank you for responding to my question! I appreciate it. I also realized there might have been some misunderstandings on my part regarding the dataset description. Thank you for clarifying!
", "rawMarkdown": "Thank you for responding to my question! I appreciate it. I also realized there might have been some misunderstandings on my part regarding the dataset description. Thank you for clarifying!", "votes": null } ], "comments": [ { "id": 2684651, "author_name": "picekl", "author_url": "", "post_date": "03/06/2024 18:25:27", "content": "Hi there,
\nThank you for pointing this out. You are right about the \"mismatch\".
\nHowever, when I wrote it, I did not mean to write it in the actual order, but, now I understand it might be misleading. I did update it based on your feedback.
Just for clarification, the order is as follows:
\nLandsat: BAND, QUARTER, and YEAR
\nClimate: RASTER-TYPE, YEAR, and MONTH
Let us know in case you will need any assistance!
\nBest,
\nLukas
Thank you for responding to my question! I appreciate it. I also realized there might have been some misunderstandings on my part regarding the dataset description. Thank you for clarifying!
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "2684479": "Hi,i have a small question regarding the dataset in this Kaggle competition. The time series dataset is available in two formats (CSV and cube data). I believe there might be a mismatch in the information provided for the cube dataset dimensions. In the data explanation, the time series cube dimensions are specified as (BAND, YEAR, QUARTER), while the monthly cube dimensions are specified as (RASTER-TYPE, YEAR, MONTH). However, the time series tensor cube data shape is torch.Size([6, 4, 21]), and the monthly cube data shape is torch.Size([4, 19, 12]). I assume that these dimensions correspond to time series (6 bands, 4 quarters, 21 years) and monthly (4 raster types, 19 years, 12 months).\n\nWhat are your thoughts on this issue?\n\n\n\n", "2684651": "Hi there, \n\nThank you for pointing this out. You are right about the \"mismatch\".\nHowever, when I wrote it, I did not mean to write it in the actual order, but, now I understand it might be misleading. I did update it based on your feedback.\n\nJust for clarification, the order is as follows:\nLandsat: BAND, QUARTER, and YEAR\nClimate: RASTER-TYPE, YEAR, and MONTH\n\nLet us know in case you will need any assistance!\n\nBest,\nLukas", "2684787": "Thank you for responding to my question! I appreciate it. I also realized there might have been some misunderstandings on my part regarding the dataset description. Thank you for clarifying!" }, "source": "meta" }