Suggestions for some of the Pandas solutions

[samukweku]

Hi @Atrebas , great article comparing the two data processing tools. The concise syntax of Rdatatable is appealing - R is not my strong suit, so hopefully my time with #pydatatable will be fun. The code below are excerpts of some parts of your Pandas code I believe could be better written, and made more performant. For some scenarios, Python's verbose syntax is unavoidable. I put comments for each of them to refer to the particular part of the code. For some, I have no idea (rleid, %inrange%,...). When I have some time again, I will have a look at it, check some codes online (the rdatatable explanation wasnt too clear for me) and then I may revisit it. Either ways, thanks for your article. Was a good read. Same for the dplyr comparision as well. Cheers.

import pandas as pd
import numpy as np

pd.__version__

'1.1.0'

df = pd.DataFrame(
  {"V1" : [1, 2, 1, 2, 1, 2, 1, 2, 1],
   "V2" : [1, 2, 3, 4, 5, 6, 7, 8, 9], 
   "V3" : [0.5, 1.0, 1.5] * 3, 
   "V4" : ['A', 'B', 'C'] * 3}) 
df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V3	V4
0	1	1	0.5	A
1	2	2	1.0	B
2	1	3	1.5	C
3	2	4	0.5	A
4	1	5	1.0	B
5	2	6	1.5	C
6	1	7	0.5	A
7	2	8	1.0	B
8	1	9	1.5	C

#Discard rows using negative indices
df.loc[~df.index.isin(range(2,7))] #no need for the list constructor
#or df.query("not index.between(2,6)", engine='python')

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V3	V4
0	1	1	0.5	A
1	2	2	1.0	B
7	2	8	1.0	B
8	1	9	1.5	C

#Filter rows using multiple conditions
df.loc[(df.V1==1) & (df.V4=="A")]
# or df.query("V1==1 and V4=='A' ")

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V3	V4
0	1	1	0.5	A
6	1	7	0.5	A

#other filters
#tuples used when selecting multiple starts (("B","C",...))
df.loc[df.V4.str.startswith("B")]

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V3	V4
1	2	2	1.0	B
4	1	5	1.0	B
7	2	8	1.0	B

#DT[V2 %inrange% list(-1:1, 1:3)]
#dont understand this yet, will look into it

#Select one column using an index (not recommended)
df.iloc[:,[2]] #DF.iloc[:, 2].to_frame() unneccessary

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V3
0	0.5
1	1.0
2	1.5
3	0.5
4	1.0
5	1.5
6	0.5
7	1.0
8	1.5

cols = ['V2', 'V3']
df.loc[:, cols]

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V2	V3
0	1	0.5
1	2	1.0
2	3	1.5
3	4	0.5
4	5	1.0
5	6	1.5
6	7	0.5
7	8	1.0
8	9	1.5

#summarise one column
df.loc[:, ["V1"]].agg(['sum']) # unnecessary : DF[['V1']].sum().to_frame(name = 'sumV1')

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1
sum	13

#Summarise several columns
#Pandas version 1.1.0 supports renaming using namedtuples
#previously available only for aggregations after groupby
#not as pretty as rdatatable
#Pandas places the aggregations as indices
df.agg(sumV1=("V1",'sum'), sdv3=("V3","std"))

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V3
sumV1	13.0	NaN
sdv3	NaN	0.433013

##Summarise a subset of rows
df.loc[:3,"V1"].sum() #0 is redundant

6

df.loc[0,'V3'] #df.head(1).V3 not necessary and inefficient

0.5

df.at[df.index[-1], 'V3'] #df.tail(1).V3 not necessary and inefficient
#at method useful and a bit more performant than loc if interested in only scalars

1.5

#trying to keep up with changes from here
df.loc[:, "V1"] = df.loc[:, "V1"] ** 2

df = df.assign(v5 = np.log(df.V1))

df = df.assign(v6 = np.sqrt(df.V1), v7 = 'X')

#Create one column and remove the others
#no need to create another dataframe
df.loc[:, ["V3"]].add(1).rename(columns={"V3":"V8"})
# or df.loc[:, ["V3"]].add(1).set_axis(["V8"], axis = 1)

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V8
0	1.5
1	2.0
2	2.5
3	1.5
4	2.0
5	2.5
6	1.5
7	2.0
8	2.5

del df["v5"]

df = df.drop(["v6", "v7"],1)

cols = 'V3'
del df[cols]

#Replace values for rows matching a condition
df.loc[df.loc[:, "V2"] < 4, "V2"] = 0

df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	1	0	A
1	4	0	B
2	1	0	C
3	4	4	A
4	1	5	B
5	4	6	C
6	1	7	A
7	4	8	B
8	1	9	C

#by group
df.groupby(["V4"],as_index=False).agg(sumV2=("V2","sum")) #no need for the to frame and reset index 

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V4	sumV2
0	A	11
1	B	13
2	C	15

#several groups
df.groupby(["V4","V1"], as_index = False).agg(sumV2=("V2","sum")) # again, no need for the to frame and reset index 

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V4	V1	sumV2
0	A	1	7
1	A	4	4
2	B	1	5
3	B	4	8
4	C	1	9
5	C	4	6

#Calling function in by

#you could apply the function before grouping
df.assign(V4 = df.V4.str.lower()).groupby(["V4"], as_index=False).agg(sumV1=("V1","sum"))

# or df.groupby(df.V4.str.lower()).agg(sumV1=("V1","sum")).reset_index()
# reset_index becomes inevitable here

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V4	sumV1
0	a	6
1	b	9
2	c	6

#Assigning column name in by
#not possible in Pandas
#create the name before grouping
df.assign(abc = lambda x: x.V4.str.lower()).groupby(["abc"], as_index=False).agg(sumV1=("V1","sum"))

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	abc	sumV1
0	a	6
1	b	9
2	c	6

#Using a condition in by
#just trying to keep track here
df.groupby(df.V4=="A").V1.sum()

V4
False    15
True      6
Name: V1, dtype: int64

#on a subset of rows
df.iloc[:5].groupby("V4").agg(sumV1=("V1",'sum'))

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	sumV1
V4
A	5
B	5
C	1

#Add a column with number of observations for each group
df.assign(n = lambda x: x.groupby("V1").V4.transform("count"))

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4	n
0	1	0	A	5
1	4	0	B	4
2	1	0	C	5
3	4	4	A	4
4	1	5	B	5
5	4	6	C	4
6	1	7	A	5
7	4	8	B	4
8	1	9	C	5

#Summarise all the columns
#better to use built-in functions over apply where possible
#as they are optimised/vectorized
#apply is a for loop, handy in some situations, but relatively slow in a lot of cases
df.agg(["max"])

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
max	4	9	C

#Summarise several columns
df.loc[:, ["V1","V2"]].agg(["mean"]) #again, no need for apply

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2
mean	2.333333	4.333333

# Summarise several columns by group
# unpacking a dictionary comprehension suffices here
df.groupby("V4").agg(**{f"{col}_mean":(col, "mean") for col in ["V1","V2"]})

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1_mean	V2_mean
V4
A	2	3.666667
B	3	4.333333
C	2	5.000000

#Summarise using a condition
df.select_dtypes(include="number").agg(["mean"])

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2
mean	2.333333	4.333333

#Modify all the columns
#reverse on the 0 axis
#apply not needed 
df.iloc[::-1]

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
8	1	9	C
7	4	8	B
6	1	7	A
5	4	6	C
4	1	5	B
3	4	4	A
2	1	0	C
1	4	0	B
0	1	0	A

#Modify several columns (dropping the others)
df.filter(["V1","V2"]).agg(np.sqrt)

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2
0	1.0	0.000000
1	2.0	0.000000
2	1.0	0.000000
3	2.0	2.000000
4	1.0	2.236068
5	2.0	2.449490
6	1.0	2.645751
7	2.0	2.828427
8	1.0	3.000000

#Modify several columns (dropping the others)
df.filter(regex='[^V4]').agg(np.exp)

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2
0	2.718282	1.000000
1	54.598150	1.000000
2	2.718282	1.000000
3	54.598150	54.598150
4	2.718282	148.413159
5	54.598150	403.428793
6	2.718282	1096.633158
7	54.598150	2980.957987
8	2.718282	8103.083928

#Modify several columns (keeping the others)
df.loc[:, "V1":"V2"] = df.filter(["V1","V2"]).agg(np.sqrt)
df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	1.0	0.000000	A
1	2.0	0.000000	B
2	1.0	0.000000	C
3	2.0	2.000000	A
4	1.0	2.236068	B
5	2.0	2.449490	C
6	1.0	2.645751	A
7	2.0	2.828427	B
8	1.0	3.000000	C

##Modify several columns (keeping the others)
cols = df.columns.difference(["V4"])
df.loc[:, cols] = df.filter(cols).agg(lambda x: pow(x,2))
df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	1.0	0.0	A
1	4.0	0.0	B
2	1.0	0.0	C
3	4.0	4.0	A
4	1.0	5.0	B
5	4.0	6.0	C
6	1.0	7.0	A
7	4.0	8.0	B
8	1.0	9.0	C

#Modify columns using a condition (dropping the others)
#no need for intermediate cols variable
df.select_dtypes("number").sub(1)

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2
0	0.0	-1.0
1	3.0	-1.0
2	0.0	-1.0
3	3.0	3.0
4	0.0	4.0
5	3.0	5.0
6	0.0	6.0
7	3.0	7.0
8	0.0	8.0

#Modify columns using a condition (keeping the others)
df.loc[:, cols] = df.loc[:, cols].astype(int)
df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	1	0	A
1	4	0	B
2	1	0	C
3	4	4	A
4	1	5	B
5	4	5	C
6	1	7	A
7	4	8	B
8	1	9	C

df.groupby("V4").head(2).assign(V2="X").sort_values("V4")

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	1	X	A
3	4	X	A
1	4	X	B
4	1	X	B
2	1	X	C
5	4	X	C

#not sure what this does
#will study it some more
#Use multiple expressions (with DT[,{j}])

#Expression chaining using DT[][] (recommended)
df.groupby(['V4'], as_index=False).agg(V1sum=("V1",np.sum)).query("V1sum > 5")

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V4	V1sum
0	A	6
1	B	9
2	C	6

df.set_index('V4', drop = False, inplace = True)
df.sort_index(inplace = True)
df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
V4
A	1	0	A
A	4	4	A
A	1	7	A
B	4	0	B
B	1	5	B
B	4	8	B
C	1	0	C
C	4	5	C
C	1	9	C

#Apply a function on the matching rows
df.loc[['A', 'C'],'V1'].sum() #faster than df.loc[['A', 'C']].V1.sum() as the data fetching is done once

12

#keeping pace with the material
#Modify values for matching rows
df.loc['A', 'V1'] = 0

#Use keys in by
df.query("index != 'B'").groupby(level=0).agg({"V1":np.sum})

# or df.loc[~(df.index=="B")].groupby(level=0).agg({"V1":"sum"})

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1
V4
A	0
C	6

#Set keys/indices for multiple columns
df.set_index(['V4', 'V1'], drop = False, inplace = True)
df.sort_index(inplace = True)

df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

		V1	V2	V4
V4	V1
A	0	0	0	A
	0	0	4	A
	0	0	7	A
B	1	1	5	B
	4	4	0	B
	4	4	8	B
C	1	1	0	C
	1	1	9	C
	4	4	5	C

#Subset using multiple keys/indices
df.loc[("C",1)]

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

		V1	V2	V4
V4	V1
C	1	1	0	C
	1	1	9	C

#Subset using multiple keys/indices
df.loc[(("B","C"),1),:]

#or query
#can be much clearer 
#df.query("V4 in ('B','C') and V1==1")
# you can rename the indices
# to separate it from the columns
# the columns will take precedence if index and columns have the same name

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

		V1	V2	V4
V4	V1
B	1	1	5	B
C	1	1	0	C
	1	1	9	C

df.query("V4 in ('B','C') and V1==1")

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

		V1	V2	V4
V4	V1
B	1	1	5	B
C	1	1	0	C
	1	1	9	C

#remove keys/indices
df.reset_index(inplace = True, drop = True)

df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	0	0	A
1	0	4	A
2	0	7	A
3	1	5	B
4	4	0	B
5	4	8	B
6	1	0	C
7	1	9	C
8	4	5	C

#trying to keep pace
df.iloc[0, 1] = 3
df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2	V4
0	0	3	A
1	0	4	A
2	0	7	A
3	1	5	B
4	4	0	B
5	4	8	B
6	1	0	C
7	1	9	C
8	4	5	C

df.sort_values(['V4','V1'], ascending = [True, False], inplace = True)

df.rename(columns = {'V2':'v2'}, inplace = True)
cols = df.columns.values; cols[1] = 'V2'
df.columns = cols

df = df[['V4', 'V1', 'V2']]

df

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V4	V1	V2
0	A	0	3
1	A	0	4
2	A	0	7
4	B	4	0
5	B	4	8
3	B	1	5
8	C	4	5
6	C	1	0
7	C	1	9

#Get row number of first (and last) observation by group
pd.DataFrame(df.groupby("V4").indices).melt() #returns all the indices

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	variable	value
0	A	0
1	A	1
2	A	2
3	B	3
4	B	4
5	B	5
6	C	6
7	C	7
8	C	8

pd.DataFrame(df.groupby("V4").indices).loc[1] #get the second row per group

A    1
B    4
C    7
Name: 1, dtype: int64

pd.DataFrame(df.groupby("V4").indices).iloc[[0,-1]].melt() #get first and last indices per group

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	variable	value
0	A	0
1	A	2
2	B	3
3	B	5
4	C	6
5	C	8

df.to_csv("test.csv", index = False)

#drop columns when reading csv
pd.read_csv("test.csv", usecols = lambda x: x != "V4")

# or pd.read_csv("test.csv", usecols = lambda x: x not in ["V4"])

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V1	V2
0	0	3
1	0	4
2	0	7
3	4	0
4	4	8
5	1	5
6	4	5
7	1	0
8	1	9

mdf = df.melt(id_vars = "V4", value_vars=("V1","V2"))
mdf

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	V4	variable	value
0	A	V1	0
1	A	V1	0
2	A	V1	0
3	B	V1	4
4	B	V1	4
5	B	V1	1
6	C	V1	4
7	C	V1	1
8	C	V1	1
9	A	V2	3
10	A	V2	4
11	A	V2	7
12	B	V2	0
13	B	V2	8
14	B	V2	5
15	C	V2	5
16	C	V2	0
17	C	V2	9

#cast data from long to wide
pd.crosstab(mdf.V4, mdf.variable)

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

variable	V1	V2
V4
A	3	3
B	3	3
C	3	3

#cast data from long to wide
pd.crosstab(mdf.V4, mdf.variable, values = mdf.value, aggfunc='sum')

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

variable	V1	V2
V4
A	0	14
B	9	13
C	6	14

#cast data from long to wide
pd.crosstab(mdf.V4, mdf.value > 5)

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

value	False	True
V4
A	5	1
B	5	1
C	5	1

[Atrebas]

Thank you very much, this is really helpful.
I will look at it into more details shortly.

[Atrebas]

The post has been updated, integrating your suggestions and crediting you, if you don't mind.
Going through your comments, I realized that it was a really careful review. Thank you, I sincerely appreciate.
Looking forward the pydatatable version ;-)
Cheers!

[samukweku]

Great work @Atrebas I will definitely make out time to write a blog comparing datatable to pydatatable for users who want to work within python and still use datatable. Pydatatable has features that have not yet been added - rdatatable has way more features. I added some documentation to the pydatatable docs (fread, groupby, ...); I will definitely add more over time. Cheers and stay safe