| Type: | Package |
| Title: | Vector Binary Tree to Make Your Data Management More Efficient |
| Version: | 0.1.1 |
| Date: | 2024-01-09 |
| Description: | Vector binary tree provides a new data structure, to make your data visiting and management more efficient. If the data has structured column names, it can read these names and factorize them through specific split pattern, then build the mappings within double list, vector binary tree, array and tensor mutually, through which the batched data processing is achievable easily. The methods of array and tensor are also applicable. Detailed methods are described in Chen Zhang et al. (2020) <doi:10.35566/isdsa2019c8>. |
| Imports: | tensorA |
| Depends: | R (≥ 2.10) |
| License: | GPL-3 |
| URL: | https://github.com/CubicZebra/VBTree |
| BugReports: | https://github.com/CubicZebra/VBTree/issues |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 7.2.3 |
| Suggests: | knitr, rmarkdown, testthat |
| VignetteBuilder: | knitr |
| NeedsCompilation: | no |
| Packaged: | 2024-01-11 09:20:13 UTC; Chen |
| Author: | Chen Zhang [aut, cre, cph] (0009-0007-7689-5030) |
| Maintainer: | Chen Zhang <chen.zhang_06sept@foxmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2024-01-12 09:00:13 UTC |
Vector Binary Tree to Make Your Data Management More Efficient
Description
Vector binary tree provides a new data structure, to make your data visiting and management more efficient. If the data has structured column names, it can read these names and factorize them through specific split pattern, then build the mappings within double list, vector binary tree, array and tensor mutually, through which the batched data processing is achievable easily. The methods of array and tensor are also applicable. Detailed methods are described in Chen Zhang et al. (2020) <doi:10.35566/isdsa2019c8>.
Details
This package provide an efficient approach to manage data by structurizing the column names. A column name is generally seen as a character object, while if it has a very organized pattern, such as "*-*-*-*" for example (each * mark presents a different condition), it must has a certain mapping relationship to a specific tensor. This package uses two data structure: double list and vector binary tree, to implement the conversion between the character vector and tensor. It affords various inquiry methods, which was mainly drived by vector binary tree, to extract the highly customizable subset from original data.
Author(s)
Chen Zhang [aut, cre, cph] (0009-0007-7689-5030)
Maintainer: Chen Zhang <chen.zhang_06sept@foxmail.com>
References
Sedgewick, Robert & Wayne, Kevin (2011). Algorithms, 4th Edition.. Addison-Wesley
Prakash, P. K. S. & Rao, Achyutuni Sri Krishna (2016). R Data Structures and Algorithms. Packt Publishing
See Also
Examples
#View the data to be visited:
summary(datatest)
colnames(datatest)
#Structurize colnames of data into vector binary tree:
dl <- chrvec2dl(colnames(datatest))
vbt <- dl2vbt(dl)
vbt
#Setting subset in different forms, for example the pattern
#"Strain-(900~1100)-(0.01, 1)-0.6" is desired:
subunregdl <- list(c(1), c(1:5), c(2,4), c(1)) # undifined double list
subregdl <- advbtinq(vbt, subunregdl) # regularized double list
subvbt <- dl2vbt(subregdl) # sub vector binary tree
subts <- vbt2ts(subvbt) # tensor
subarr <- vbt2arr(subvbt) # array
subchrvec <- as.vector(subarr) # character vector
#Visit the data through different methods:
datavisit(datatest, subunregdl) # by handmade double list
datavisit(datatest, subregdl) # by defined double list
datavisit(datatest, subvbt) # by vector binary tree
datavisit(datatest, subts) # by tensor
datavisit(datatest, subarr) # by array
datavisit(datatest, subchrvec) # by character vector
Using double list to visit vector binary tree
Description
Advanced visiting for the vector binary tree. Return a double list by specific assigment determined by
the argument inq.
Usage
advbtinq(x, inq)
Arguments
x |
The vector binary tree to be visited. Traversal is acheivable through invalid assignment in desired layer. |
inq |
An integer double list to determine the location to be visited. The length of |
Value
Return a double list according to the argument inq.
See Also
Examples
#Make vector binary tree:
colnamevbt <- dl2vbt(chrvec2dl(colnames(datatest)))
#Visit by specific assignment:
visit <- list(c(2), c(3:6), c(2,4), 1)
advbtinq(colnamevbt, visit)
#Traversal of the second layers:
visit <- list(c(2), colnamevbt$dims[2]+1, c(2,4), 1)
advbtinq(colnamevbt, visit)
#Invalid assignments in 1st and 3rd layers:
visit <- list(c(3), c(3:6), c(5), 1)
advbtinq(colnamevbt, visit)
Using double list to generate sub tree from vector binary tree
Description
Advanced visiting for the vector binary tree. Generating a sub tree from visited vector binary tree,
through specific assigment determined by the argument inq.
Usage
advbtsub(x, inq)
Arguments
x |
The vector binary tree to be visited. Traversal is acheivable through invalid assignment in desired layers. |
inq |
An integer double list to determine the visiting location. The length of |
Value
Return a sub tree from visited vector binary tree, according to the argument inq.
See Also
Examples
#Make vector binary tree:
colnamevbt <- dl2vbt(chrvec2dl(colnames(datatest)))
#Visit by specific assignment:
visit <- list(c(2), c(3:6), c(2,4), 1)
advbtsub(colnamevbt, visit)
#Traversal of the second layers:
visit <- list(c(2), colnamevbt$dims[2]+1, c(2,4), 1)
advbtsub(colnamevbt, visit)
#Invalid assignments in 1st and 3rd layers:
visit <- list(c(3), c(3:6), c(5), 1)
advbtsub(colnamevbt, visit)
Convert a structured character array to double list
Description
Convert a structured character array to a double list. All character elements in array will be splited by a specific pattern then sorted intrinsically in each layer of the double list.
Usage
arr2dl(x, ...)
Arguments
x |
A structured character array to be converted. |
... |
Argument in |
Value
Return a double list based on the input array.
See Also
Examples
#Write the column names of datatest into a array:
arr <- dl2arr(chrvec2dl(colnames(datatest)))
#Recover the double list from character array:
arr2dl(arr)
Convert a structured character array to double list
Description
Convert a structured character array to a vector binary tree. All character elements in array will be splited by a specific pattern then sorted intrinsically in each layer of the vector binary tree.
Usage
arr2vbt(x, ...)
Arguments
x |
A structured character array to be converted. |
... |
Argument in |
Value
Return a vector binary tree based on the input array.
See Also
Examples
#Write the column names of datatest into a array:
arr <- dl2arr(chrvec2dl(colnames(datatest)))
#Recover the vector binary tree from character array:
arr2vbt(arr)
Convert character vector to a double list
Description
Structurize a character vector to a double list. Layers in the double list will be determined by the given pattern.
Usage
chrvec2dl(x, splt = "-")
Arguments
x |
a character vector to be converted. |
splt |
a string pattern to make defination for spliting each layer of double list. |
Value
return a character double list splited by defined pattern, the default pattern is "-".
Examples
#example using default dataset:
charvector <- colnames(datatest)
chrvec2dl(charvector, "-")
A test data structurized column names.
Description
A test data with 56 different columns.
Usage
data("datatest")Details
A test data structurized column names, with two data type "Strain" and "Stress", 7 different temperatures, 4 kinds strain rates and one level of compression rate.
Examples
datatest
Extract subset of data using different methods
Description
Extract the subset of data by column names using tensor, array, double list, integer vector, or vector binary tree.
Usage
datavisit(data, inq)
Arguments
data |
A data.frame with structured column names. |
inq |
An argument to determine the subset to be extracted by column names. A tensor, array, double list, integer vector and
vector binary tree is available format of |
Value
Return a list which contains the item index, column name, column coordinate and the data in corresponding column for each
element contained in the assignment of inq.
See Also
vbtinq, advbtinq, trvseleinq,
trvsidxinq, trvssubinq.
Examples
#View the data to be visited:
summary(datatest)
colnames(datatest)
#Structurize colnames of data into vector binary tree:
dl <- chrvec2dl(colnames(datatest))
vbt <- dl2vbt(dl)
vbt
#Setting subset in different forms, for example the pattern
#"Strain-(900~1100)-(0.01, 1)-0.6" is desired:
subunregdl <- list(c(1), c(1:5), c(2,4), c(1)) # undefined double list
subregdl <- advbtinq(vbt, subunregdl) # regularized double list
subvbt <- dl2vbt(subregdl) # sub vector binary tree
subts <- vbt2ts(subvbt) # tensor
subarr <- vbt2arr(subvbt) # array
subchrvec <- as.vector(subarr) # character vector
#Visit the data through different methods:
datavisit(datatest, subunregdl) # by handmade double list
datavisit(datatest, subregdl) # by defined double list
datavisit(datatest, subvbt) # by vector binary tree
datavisit(datatest, subts) # by tensor
datavisit(datatest, subarr) # by array
datavisit(datatest, subchrvec) # by character vector
Convert a double list to array
Description
Convert a double list to an array. The pure numeric layers will be sorted intrinsically then all elements will be bound in certain order as one character element, and filled into the proper location in the array.
Usage
dl2arr(x)
Arguments
x |
A double list to be converted. |
Value
Return an array filled with the binding character elements.
See Also
Examples
#Make column names of datatest into double list:
dl <- chrvec2dl(colnames(datatest), "-")
#Convert the double list to a tensor:
dl2arr(dl)
Convert a double list to tensor
Description
Convert a double list to a tensor. The pure numeric layers will be sorted intrinsically then all elements will be bound in certain order as one character element, and filled into the proper location in the tensor.
Usage
dl2ts(x)
Arguments
x |
A double list to be converted. |
Value
Return a tensor filled with the binding character elements.
See Also
Examples
#Make column names of datatest into double list:
dl <- chrvec2dl(colnames(datatest), "-")
#Convert the double list to a tensor:
dl2ts(dl)
Convert a double list to vector binary tree
Description
Convert a double list to vector binary tree. The pure numeric layers will be sorted intrinsically then all elements be exported in character form.
Usage
dl2vbt(x, regularize = TRUE, splt = "-")
Arguments
x |
A double list to be converted. |
regularize |
A boolean value to control the treatment of empty layers of double listed to be converted.
The default value |
splt |
A string pattern to split the binding elements in each layer if the sub-constructure exists. The default pattern uses "-". |
Value
Return a vector binary tree.
See Also
vbtinq, vbtsub, advbtinq,
advbtsub, trvssubinq, dl2ts,
dl2arr.
Examples
#Structurize the column names of datatest:
colname <- colnames(datatest)
colnamedl <- chrvec2dl(colname, "-")
colnamevbt <- dl2vbt(colnamedl)
#Simple data cleaning for sub-constructure existing double list;
#Make unregulated double list:
unregdl <- list(c("7", 2, 10), c("chr", "5"), c(),
c("var2", "var1", "var3"), c("M-8-9", "3-2"), c("6-3", "2-7"))
regvbt <- dl2vbt(unregdl)
regvbt2 <- dl2vbt(unregdl, FALSE) # not recommended
Welcome message
Description
Welcome message
Usage
hello()
Value
exit code of zero
Examples
hello()
Make traversal from vector binary tree
Description
Generating a table of traversal from given vector binary tree, in order to construct correct mapping relationships within double list, vector binary tree, array and tensor.
Usage
trvs(x)
Arguments
x |
A vector binary tree. |
Value
Return a traversal table from the given vector binary tree.
Examples
#Make vector binary tree:
colnamevbt <- dl2vbt(chrvec2dl(colnames(datatest)))
#Construct traversal table:
trvs(colnamevbt)
Using character element to visit the traversal table
Description
Visit the traversal table generated from a vector binary tree through the character element determined by the argument inq, and return
an inquiry result containing its numeric item index, the character pattern and its corresponding coordinate.
Usage
trvseleinq(trvs, inq)
Arguments
trvs |
The traversal table to be visited, which should be generated from the vector binary tree by the function trvs(). |
inq |
A desired character element to match the traversal table. |
Value
Return an inquiry result with a numeric item index, a character pattern and its coordinate in form of integer vector.
Examples
#Make traversal table:
trav <- trvs(dl2vbt(chrvec2dl(colnames(datatest))))
#Visit specific element by character pattern:
trvseleinq(trav,"Strain-1100-0.001-0.6")
Using vector to visit the traversal table
Description
Visit the traversal table generated from a vector binary tree through the coordinate determined by the argument inq, and return
an inquiry result containing its numeric item index, its corresponding character pattern and the coordinate.
Usage
trvsidxinq(trvs, inq)
Arguments
trvs |
The traversal table to be visited, which should be generated from the vector binary tree by the function trvs(). |
inq |
An integer vector to assign the coordinate corresponding to the element to be visited. |
Value
Return an inquiry result with a numeric item index, a character pattern and its coordinate in form of integer vector.
Examples
#Make traversal table:
trav <- trvs(dl2vbt(chrvec2dl(colnames(datatest))))
#Visit specific element by its coordinate:
trvsidxinq(trav,c(1,2,3,1))
Using sub vector binary tree to visit the traversal table
Description
Visit the traversal table generated from a vector binary tree through the sub vector binary tree determined by the argument inq, and
return an inquiry list containing the numeric index, the character pattern and the corresponding coordinate for each item.
Usage
trvssubinq(trvs, inq)
Arguments
trvs |
The traversal table to be visited, which should be generated from the vector binary tree by the function trvs(). |
inq |
A sub tree generated from the original vector binary tree, to determine the subset of elements to be visited. |
Value
Return a list containing the numeric index, the character pattern and the corresponding coordinate for each item.
See Also
Examples
#Make original vector binary tree and its traversal table:
vbt <- dl2vbt(chrvec2dl(colnames(datatest)))
trav <- trvs(vbt)
#Visit all elements defined by sub vector binary tree:
#example 1: visit all "Stress-*-*-*" patterns;
#make sub vector binary tree through vbtsub() then execute inquiry:
subvbt <- vbtsub(vbt, c(2,-1,-1,-1))
trvssubinq(trav, subvbt)
#example 2: visit all "Strain-("950", "1050")-("0.001", "0.1")-*" patterns;
#make sub vector binary tree through advbtsub() then execute inquiry:
subvbt <- advbtsub(vbt, list(1, c(2,4), c(1,3), 1))
trvssubinq(trav, subvbt)
Convert a structured character tensor to double list
Description
Convert a structured character tensor to a double list. All character elements in tensor will be splited by a specific pattern then sorted intrinsically in each layer of the double list.
Usage
ts2dl(x, ...)
Arguments
x |
A structured character tensor to be converted. |
... |
Argument in |
Value
Return a double list based on the input tensor.
See Also
Examples
#Write the column names of datatest into a tensor:
ts <- dl2ts(chrvec2dl(colnames(datatest)))
#Recover the double list from character tensor:
ts2dl(ts)
Convert a structured character tensor to double list
Description
Convert a structured character tensor to a vector binary tree. All character elements in tensor will be splited by a specific pattern then sorted intrinsically in each layer of the vector binary tree.
Usage
ts2vbt(x, ...)
Arguments
x |
A structured character tensor to be converted. |
... |
Argument in |
Value
Return a vector binary tree based on the input tensor.
See Also
Examples
#Write the column names of datatest into a tensor:
ts <- dl2ts(chrvec2dl(colnames(datatest)))
#Recover the vector binary tree from character tensor:
ts2vbt(ts)
Convert a vector binary tree to array
Description
Convert a vector binary tree to an array. The pure numeric layers will be sorted intrinsically then all elements will be bound in certain order as one character element, and filled into the proper location in the array.
Usage
vbt2arr(x)
Arguments
x |
A vector binary tree to be converted. |
Value
Return an array filled with the binding character elements.
See Also
Examples
#Make column names of datatest into vector binary tree:
vbt <- dl2vbt(chrvec2dl(colnames(datatest), "-"))
#Convert the vector binary tree to an array:
vbt2arr(vbt)
Convert a vector binary tree to double list
Description
Recover a vector binary tree to double list for easy visualization. Empty layers in vector binary tree will be marked by the symbol "*" as default.
Usage
vbt2dl(x)
Arguments
x |
A vector binary tree to be converted. |
Value
Return a double list based on input vector binary tree.
See Also
vbtinq, vbtsub, advbtinq,
advbtsub, trvssubinq, vbt2ts,
vbt2arr.
Examples
#Recover vector binary tree to a double list for easy visualization:
vbt <- dl2vbt(chrvec2dl(colnames(datatest))) #make vector binary tree
vbt2dl(vbt)
Convert a vector binary tree to tensor
Description
Convert a vector binary tree to a tensor. The pure numeric layers will be sorted intrinsically then all elements will be bound in certain order as one character element, and filled into the proper location in the tensor.
Usage
vbt2ts(x)
Arguments
x |
A vector binary tree to be converted. |
Value
Return a tensor filled with the binding character elements.
See Also
Examples
#Make column names of datatest into vector binary tree:
vbt <- dl2vbt(chrvec2dl(colnames(datatest), "-"))
#Convert the vector binary tree to a tensor:
vbt2ts(vbt)
Using vector to visit vector binary tree
Description
Visit the vector binary tree and return a double list through specific assigment determined
by the argument inq.
Usage
vbtinq(x, inq)
Arguments
x |
The vector binary tree to be visited. Traversal is available by setting -1 in desired layer. |
inq |
An integer vector to determine desired location. The length of |
Value
Return a double list according to the argument inq.
See Also
Examples
#Make vector binary tree:
colnamevbt <- dl2vbt(chrvec2dl(colnames(datatest)))
#Visit by specific assignment:
vbtinq(colnamevbt, c(2, 3, 1, 1))
#Traversal of the second layers:
vbtinq(colnamevbt, c(2, -1, 1, 1))
#Invalid assignments in 1st and 3rd layers:
vbtinq(colnamevbt, c(4, 3, 7, 1))
Using vector to generate sub tree from vector binary tree
Description
Visit the vector binary tree and generate a sub tree from visited vector binary tree, through
specific assigment determined by the argument inq.
Usage
vbtsub(x, inq)
Arguments
x |
The vector binary tree to be visited. Traversal is available by setting -1 in desired layer. |
inq |
An integer vector to determine the visiting location. The length of |
Value
Return a sub tree from visited vector binary tree, according to the argument inq.
See Also
Examples
#Make vector binary tree:
colnamevbt <- dl2vbt(chrvec2dl(colnames(datatest)))
#Generating sub tree by specific assignment:
vbtsub(colnamevbt, c(2, 3, 1, 1))
#Generating sub tree with traversal in the second layers:
vbtsub(colnamevbt, c(2, -1, 1, 1))
#Generating sub tree with invalid assignments in 1st and 3rd layers:
vbtsub(colnamevbt, c(4, 3, 7, 1))