dexplo¶

A data analysis library comparable to pandas

Installation¶

You must have cython installed. Run python setup.py build_ext --use-cython -i

Main Goals¶

• A minimal set of features
• Be as explicit as possible
• There should be one-- and preferably only one --obvious way to do it.

Data Structures¶

• Only DataFrames
• No Series

Only Scalar Data Types¶

All data types allow nulls

• [x] bool - always 8 bits
• [x] int
• [x] float
• [x] str - stored as a categorical
• [x] datetime
• [x] timedelta

Column Labels¶

• No hierarchical index
• Column names must be strings
• Column names must be unique

Row Labels¶

• No row labels for now
• Only a number display on the output

Subset Selection¶

• Only one way to select data - [ ]
• Subset selection will be explicit and necessitate both rows and columns
• Rows will be selected only by integer location
• Columns will be selected by either label or integer location. Since columns must be strings, this will not be amibguous
• Slice notation is also OK

Development¶

• Must use type hints
• Must use 3.6+ - fstrings
• numpy

Advantages over pandas¶

• Easier to write idiomatically
• String processing will be much faster
• Nulls allowed in each data type
• Nearly all operations will be faster

API¶

Attributes¶

• [x] size
• [x] shape
• [x] values
• [x] dtypes

Methods¶

Stats

• [x] abs
• [x] all
• [x] any
• [x] argmax
• [x] argmin
• [x] clip
• [x] corr
• [x] count
• [x] cov
• [x] cummax
• [x] cummin
• [x] cumprod
• [x] cumsum
• [x] describe
• [x] max
• [x] min
• [x] median
• [x] mean
• [x] mode
• [x] nlargest
• [x] nsmallest
• [x] prod
• [x] quantile
• [x] rank
• [x] round
• [x] std
• [x] streak
• [x] sum
• [x] var
• [x] unique
• [x] nunique
• [x] value_counts

Selection

• [x] drop
• [x] head
• [x] isin
• [x] rename
• [x] sample
• [x] select_dtypes
• [x] tail
• [x] where

Missing Data

• [x] isna
• [x] dropna
• [x] fillna
• [ ] interpolate

Other

• [x] append
• [x] astype
• [x] factorize
• [x] groupby
• [x] iterrows
• [ ] join
• [x] melt
• [x] pivot
• [x] replace
• [x] rolling
• [x] sort_values
• [x] to_csv

Other (after 0.1 release) - [ ] cut - [ ] plot - [ ] profile

Functions

• [x] read_csv
• [ ] read_sql
• [ ] concat

Group By - specifically with groupby method

• [x] agg
• [x] all
• [x] apply
• [x] any
• [x] corr
• [x] count
• [x] cov
• [x] cumcount
• [x] cummax
• [x] cummin
• [x] cumsum
• [x] cumprod
• [x] head
• [x] first
• [ ] fillna
• [x] filter
• [x] last
• [x] max
• [x] median
• [x] min
• [x] ngroups
• [x] nunique
• [x] prod
• [ ] quantile
• [ ] rank
• [ ] rolling
• [x] size
• [x] sum
• [x] tail
• [x] var

str - df.str.<method>

• [x] capitalize
• [x] cat
• [x] center
• [x] contains
• [x] count
• [x] endswith
• [x] find
• [x] findall
• [x] get
• [x] get_dummies
• [x] isalnum
• [x] isalpha
• [x] isdecimal
• [x] isdigit
• [x] islower
• [x] isnumeric
• [x] isspace
• [x] istitle
• [x] isupper
• [x] join
• [x] len
• [x] ljust
• [x] lower
• [x] lstrip
• [x] partition
• [x] repeat
• [x] replace
• [x] rfind
• [x] rjust
• [x] rpartition
• [x] rsplit
• [x] rstrip
• [x] slice
• [x] slice_replace
• [x] split
• [x] startswith
• [x] strip
• [x] swapcase
• [x] title
• [x] translate
• [x] upper
• [x] wrap
• [x] zfill

dt - df.dt.<method>

• [x] ceil
• [x] day
• [x] day_of_week
• [x] day_of_year
• [x] days_in_month
• [x] floor
• [ ] freq
• [x] hour
• [x] is_leap_year
• [x] is_month_end
• [x] is_month_start
• [x] is_quarter_end
• [x] is_quarter_start
• [x] is_year_end
• [x] is_year_start
• [x] microsecond
• [x] millisecond
• [x] minute
• [x] month
• [x] nanosecond
• [x] quarter
• [x] round
• [x] second
• [x] strftime
• [x] to_pydatetime
• [x] to_pytime
• [ ] tz
• [ ] tz_convert
• [ ] tz_localize
• [x] weekday_name
• [x] week_of_year
• [x] year

td - df.td.<method>

• [ ] ceil
• [ ] components
• [x] days
• [ ] floor
• [ ] freq
• [x] microseconds
• [x] milliseconds
• [x] nanoseconds
• [ ] round
• [x] seconds
• [ ] to_pytimedelta