-
Notifications
You must be signed in to change notification settings - Fork 53
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Refactor storage and access of results
This commit changes storage and access of results in order to improve performance. DataFrames are replaced by DenseAxisArrays. It also optimizes the generation of realized results.
- Loading branch information
1 parent
edd4a80
commit 21f6dff
Showing
28 changed files
with
825 additions
and
871 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,84 @@ | ||
| struct ResultsByTime{T} | ||
| key::OptimizationContainerKey | ||
| data::SortedDict{Dates.DateTime, T} | ||
| resolution::Dates.Period | ||
| column_names::Vector{String} | ||
| end | ||
|
|
||
| function ResultsByTime(key, data, resolution, column_names) | ||
| _check_column_consistency(data, column_names) | ||
| ResultsByTime(key, data, resolution, column_names) | ||
| end | ||
|
|
||
| function _check_column_consistency( | ||
| data::SortedDict{Dates.DateTime, DenseAxisArray{Float64, 2}}, | ||
| cols, | ||
| ) | ||
| for val in values(data) | ||
| if axes(val)[1] != cols | ||
| error("Mismatch in DenseAxisArray column names: $(axes(val)[1]) $cols") | ||
| end | ||
| end | ||
| end | ||
|
|
||
| function _check_column_consistency(data::SortedDict{Dates.DateTime, Matrix{Float64}}, cols) | ||
| for val in values(data) | ||
| if size(val)[2] != length(cols) | ||
| error("Mismatch in length of Matrix columns: $(size(val)[2]) $(length(cols))") | ||
| end | ||
| end | ||
| end | ||
|
|
||
| # This struct behaves like a dict, delegating to its 'data' field. | ||
| Base.length(res::ResultsByTime) = length(res.data) | ||
| Base.iterate(res::ResultsByTime) = iterate(res.data) | ||
| Base.iterate(res::ResultsByTime, state) = iterate(res.data, state) | ||
| Base.getindex(res::ResultsByTime, i) = getindex(res.data, i) | ||
| Base.setindex!(res::ResultsByTime, v, i) = setindex!(res.data, v, i) | ||
| Base.firstindex(res::ResultsByTime) = firstindex(res.data) | ||
| Base.lastindex(res::ResultsByTime) = lastindex(res.data) | ||
|
|
||
| get_column_names(x::ResultsByTime) = x.column_names | ||
| get_num_rows(::ResultsByTime{DenseAxisArray{Float64, 2}}, data) = length(axes(data)[2]) | ||
| get_num_rows(::ResultsByTime{Matrix{Float64}}, data) = size(data)[1] | ||
|
|
||
| function _add_timestamps!(df::DataFrames.DataFrame, results::ResultsByTime, timestamp, data) | ||
| time_col = | ||
| range(timestamp; length = get_num_rows(results, data), step = results.resolution) | ||
| DataFrames.insertcols!(df, 1, :DateTime => time_col) | ||
| end | ||
|
|
||
| function make_dataframe( | ||
| results::ResultsByTime{DenseAxisArray{Float64, 2}}, | ||
| timestamp::Dates.DateTime, | ||
| ) | ||
| array = results.data[timestamp] | ||
| df = DataFrames.DataFrame(permutedims(array.data), axes(array)[1]) | ||
| _add_timestamps!(df, results, timestamp, array) | ||
| return df | ||
| end | ||
|
|
||
| function make_dataframe(results::ResultsByTime{Matrix{Float64}}, timestamp::Dates.DateTime) | ||
| array = results.data[timestamp] | ||
| df = DataFrames.DataFrame(array, results.column_names) | ||
| _add_timestamps!(df, results, timestamp, array) | ||
| return df | ||
| end | ||
|
|
||
| function make_dataframes(results::ResultsByTime) | ||
| return SortedDict(k => make_dataframe(results, k) for k in keys(results.data)) | ||
| end | ||
|
|
||
| struct ResultsByKeyAndTime | ||
| "Contains all keys stored in the model." | ||
| result_keys::Vector{OptimizationContainerKey} | ||
| "Contains the results that have been read from the store and cached." | ||
| cached_results::Dict{OptimizationContainerKey, ResultsByTime} | ||
| end | ||
|
|
||
| ResultsByKeyAndTime(result_keys) = ResultsByKeyAndTime( | ||
| collect(result_keys), | ||
| Dict{OptimizationContainerKey, ResultsByTime}(), | ||
| ) | ||
|
|
||
| Base.empty!(res::ResultsByKeyAndTime) = empty!(res.cached_results) |
Oops, something went wrong.