Useful Shortcuts
Control Q |
access documentation |
Control D |
duplicate code block |
dir() |
see the list of methods |
Tab (in Git bash) |
to toggle the files in the directory |
Useful OS commands
os.getcwd() |
get current directory |
os.chdir('path') |
change directory |
os.listdir('folder_path') |
list directory |
os.path.join('path1', 'path2') |
join file path |
Set up Env
conda env create --prefix P:\env\myenv --file='P:\filepath\env.yml' |
conda create env_name |
create a new env |
pip install -r "P:\filepath\requirements.txt" |
pip install 'P:\filepath\package.whl' --force-reinstall |
reinstall whl file |
Pandas Display
pd.set_option(‘display.max_colwidth’, 10_000) |
pd.set_option(‘display.max_rows’, 1000) |
ps.set_option('display.max_columns', 100) |
Neat Functions
[function(i) for i in list if logic_expression] |
list comprehension |
lambda arguments: expression
(e.g.) x = lambda a, b : a * b
print(x(5, 6)) |
a lambda function is a small anonymous function |
Explore data
df.col.describe() |
df.col.value_counts() |
normalize=True |
|
|
Reshape dataframe
df.groupby('col1').col2.count() |
count / sum / min / max / mean |
df.groupby(['col1', 'col2']).['col3', 'col4'].count() |
df1 = df.groupby('col').agg({'value1':['min','max'], 'value2':['mean']}) |
df.pivot_table(index='col1', columns='col2', values=['col3', 'col4'], aggfunc={'col3':np.mean, 'col4':[min, max]}) |
df1.columns = ['_'.join(col) for col in df1.columns] |
to flatten multi index df after groupby / pivot |
df1.merge(df2, left_on='l_key', right_on='r_key', how='left') |
how: {‘left’, ‘right’, ‘outer’, ‘inner’, ‘cross’}, default ‘inner’ |
pd.concat([df1, df2]) |
union 2 dataframes |
Statements
For Loop for var in range:
statements
|
While Loop while expression:
statements
|
If Statement if expression:
statements
elif expression:
statements
else:
statements
|
Exception Handling try:
print(x)
except NameError:
print("variable x is not defined")
except:
print("sth else went wrong")
else:
print("nth went wrong")
finally:
print("the 'try except' is finished")
|
|
