Theres added methods for reading the dataset from disjoint time periods (e.g. only from 2012 and 2017) and the result is generated as a whole. However, besides from doing aggregation on the whole period (e.g. the mean from that whole result) theres currently not any means to visualize such subsets.
Consider
the query: "I want the data from the first month of every year". There are multiple ways of presenting this data:
[x] Non-aggregated: All hourly values presented in yearly graph. Cons:
Big gaps of "missing" data on graph
Many concentrated datapoint in small area -> unreadable
[x] Aggregate togehter all points within month -> one yearly value. Pros: Most intuitive. Cons: few datapoints
[x] Aggregate together hourly values -> daily values. Still big gaps, but more readable
[x] Years are "merged" such that the mean for corresponding days (of the month) are aggregated. Values for each day are aggregations (e.g. mean) for ech year. Could possibly also use hourly values.
[x] Aggregated everything within period. (Already implemented in #17 )
Theres added methods for reading the dataset from disjoint time periods (e.g. only from 2012 and 2017) and the result is generated as a whole. However, besides from doing aggregation on the whole period (e.g. the mean from that whole result) theres currently not any means to visualize such subsets.
Consider
the query: "I want the data from the first month of every year". There are multiple ways of presenting this data: