In [1]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
% matplotlib inline
Object Creation¶
In [2]:
s = pd.Series([1,3,5,np.nan,6,8])
s
Out[2]:
In [3]:
dates = pd.date_range('20180930', periods=6)
dates
Out[3]:
In [4]:
df = pd.DataFrame(np.random.randn(6,4), index=dates, columns=list('ABCD'))
df
Out[4]:
In [5]:
df2 = pd.DataFrame({ 'A' : 1.,
'B' : pd.Timestamp('20180930'),
'C' : pd.Series(1,index=list(range(4)),dtype='float32'),
'D' : np.array([3] * 4,dtype='int32'),
'E' : pd.Categorical(["test","train","test","train"]),
'F' : 'foo' })
df2
Out[5]:
In [6]:
df2.dtypes
Out[6]:
In [7]:
#df2.<TAB>
Viewing Data¶
In [8]:
df.head()
Out[8]:
In [9]:
df.tail(3)
Out[9]:
In [10]:
df.index
Out[10]:
In [11]:
df.columns
Out[11]:
In [12]:
df.values
Out[12]:
In [13]:
df.describe()
Out[13]:
In [14]:
df.T
Out[14]:
In [15]:
df.sort_index(axis=1, ascending=False)
Out[15]:
In [16]:
df.sort_values(by='B')
Out[16]:
Selection¶
In [17]:
df['A'] #df.A
Out[17]:
In [18]:
df[0:3]
Out[18]:
In [19]:
df['20181001':'20181002']
Out[19]:
Selection by Label¶
In [20]:
df.head()
Out[20]:
In [21]:
df.loc[dates[0]]
Out[21]:
In [22]:
df.loc[:,['A','B']]
Out[22]:
In [23]:
df.loc['20181001':'20181002',['A','B']]
Out[23]:
In [24]:
df.loc['20181002',['A','B']]
Out[24]:
In [25]:
df.loc[dates[0],'A']
Out[25]:
In [26]:
df.at[dates[0],'A']
Out[26]:
Selection by Position¶
In [27]:
df.iloc[3]
Out[27]:
In [28]:
df.iloc[3:5,0:2]
Out[28]:
In [29]:
df.iloc[[1,2,4],[0,2]]
Out[29]:
In [30]:
df.iloc[1:3,:]
Out[30]:
In [31]:
df.iloc[:,1:3]
Out[31]:
In [32]:
df.iloc[1,1]
Out[32]:
In [33]:
df.iat[1,1]
Out[33]:
Boolean Indexing¶
In [34]:
df[df.A > 0]
Out[34]:
In [35]:
df.A>0
Out[35]:
In [36]:
df[df > 0]
Out[36]:
In [37]:
df>0
Out[37]:
In [38]:
df2 = df.copy()
In [39]:
df2['E'] = ['one', 'one','two','three','four','three']
In [40]:
df2
Out[40]:
In [41]:
df2[df2['E'].isin(['two','four'])]
Out[41]:
In [42]:
df2['E'].isin(['two','four'])
Out[42]:
Setting¶
In [43]:
s1 = pd.Series([1,2,3,4,5,6], index=pd.date_range('20180930', periods=6))
s1
Out[43]:
In [44]:
df['F'] = s1
In [45]:
df.at[dates[0],'A'] = 0
In [46]:
df.iat[0,1] = 0
In [47]:
df.loc[:,'D'] = np.array([5] * len(df))
In [48]:
df
Out[48]:
In [49]:
df2 = df.copy()
In [50]:
df2[df2 > 0] = -df2
In [51]:
df2
Out[51]:
Missing Data¶
In [52]:
df1 = df.reindex(index=dates[0:4], columns=list(df.columns) + ['E'])
In [53]:
df1.loc[dates[0]:dates[1],'E'] = 1
In [54]:
df1
Out[54]:
In [55]:
df1.dropna(how='any')
Out[55]:
In [56]:
df1.fillna(value=5)
Out[56]:
In [57]:
pd.isna(df1)
Out[57]:
Operations¶
In [58]:
df.mean()
Out[58]:
In [59]:
df.mean(1)
Out[59]:
In [60]:
s = pd.Series([1,3,5,np.nan,6,8], index=dates).shift(2)
s
Out[60]:
In [61]:
df.sub(s, axis='index')
Out[61]:
Apply¶
In [62]:
df.apply(np.cumsum)
Out[62]:
In [63]:
df.apply(lambda x: x.max() - x.min())
Out[63]:
Histogramming¶
In [64]:
s = pd.Series(np.random.randint(0, 7, size=10))
s
Out[64]:
In [65]:
s.value_counts()
Out[65]:
In [66]:
plt.hist(s)
Out[66]:
String Methods¶
In [67]:
s = pd.Series(['A', 'B', 'C', 'Aaba', 'Baca', np.nan, 'CABA', 'dog', 'cat'])
s
Out[67]:
In [68]:
s.str.lower()
Out[68]:
Merge¶
Concat¶
In [69]:
df = pd.DataFrame(np.random.randn(10, 4))
df
Out[69]:
In [70]:
pieces = [df[:3], df[3:7], df[7:]]
In [71]:
pd.concat(pieces)
Out[71]:
Join¶
In [72]:
left = pd.DataFrame({'key': ['foo', 'foo'], 'lval': [1, 2]})
In [73]:
right = pd.DataFrame({'key': ['foo', 'foo'], 'rval': [4, 5]})
In [74]:
left
Out[74]:
In [75]:
right
Out[75]:
In [76]:
pd.merge(left, right, on='key')
Out[76]:
In [77]:
left = pd.DataFrame({'key': ['foo', 'bar'], 'lval': [1, 2]})
In [78]:
right = pd.DataFrame({'key': ['foo', 'bar'], 'rval': [4, 5]})
In [79]:
left
Out[79]:
In [80]:
right
Out[80]:
In [81]:
pd.merge(left, right, on='key')
Out[81]:
Append¶
In [82]:
df = pd.DataFrame(np.random.randn(8, 4), columns=['A','B','C','D'])
df
Out[82]:
In [83]:
s = df.iloc[3]
In [84]:
df.append(s, ignore_index=True)
Out[84]:
Grouping¶
In [85]:
df = pd.DataFrame({'A' : ['foo', 'bar', 'foo', 'bar',
'foo', 'bar', 'foo', 'foo'],
'B' : ['one', 'one', 'two', 'three',
'two', 'two', 'one', 'three'],
'C' : np.random.randn(8),
'D' : np.random.randn(8)})
In [86]:
df
Out[86]:
In [87]:
df.groupby('A').sum()
Out[87]:
In [88]:
df.groupby(['A','B']).sum()
Out[88]:
Reshaping¶
In [89]:
tuples = list(zip(*[['bar', 'bar', 'baz', 'baz',
'foo', 'foo', 'qux', 'qux'],
['one', 'two', 'one', 'two',
'one', 'two', 'one', 'two']]))
In [90]:
index = pd.MultiIndex.from_tuples(tuples, names=['first', 'second'])
In [91]:
df = pd.DataFrame(np.random.randn(8, 2), index=index, columns=['A', 'B'])
In [92]:
df2 = df[:4]
In [93]:
df2
Out[93]:
In [94]:
stacked = df2.stack()
In [95]:
stacked
Out[95]:
In [96]:
stacked.unstack()
Out[96]:
In [97]:
stacked.unstack(1)
Out[97]:
In [98]:
stacked.unstack(0)
Out[98]:
Pivot Tables¶
In [99]:
df = pd.DataFrame({'A' : ['one', 'one', 'two', 'three'] * 3,
'B' : ['A', 'B', 'C'] * 4,
'C' : ['foo', 'foo', 'foo', 'bar', 'bar', 'bar'] * 2,
'D' : np.random.randn(12),
'E' : np.random.randn(12)})
In [100]:
df
Out[100]:
In [101]:
pd.pivot_table(df, values='D', index=['A', 'B'], columns=['C'])
Out[101]:
Time Series¶
In [102]:
rng = pd.date_range('30/09/2018', periods=100, freq='S')
rng
Out[102]:
In [103]:
ts = pd.Series(np.random.randint(0, 500, len(rng)), index=rng)
ts.head()
Out[103]:
In [104]:
ts.resample('5Min').sum()
Out[104]:
In [105]:
rng = pd.date_range('30/09/2018 00:00', periods=5, freq='D')
rng
Out[105]:
In [106]:
ts = pd.Series(np.random.randn(len(rng)), rng)
ts
Out[106]:
In [107]:
ts_utc = ts.tz_localize('UTC')
ts_utc
Out[107]:
In [108]:
ts_utc.tz_convert('US/Eastern')
Out[108]:
In [109]:
rng = pd.date_range('30/09/2018', periods=5, freq='M')
In [110]:
ts = pd.Series(np.random.randn(len(rng)), index=rng)
In [111]:
ts
Out[111]:
In [112]:
ps = ts.to_period()
ps
Out[112]:
In [113]:
ps.to_timestamp()
Out[113]:
In [114]:
prng = pd.period_range('1990Q1', '2000Q4', freq='Q-NOV')
prng
Out[114]:
In [115]:
ts = pd.Series(np.random.randn(len(prng)), prng)
ts.head()
Out[115]:
In [116]:
ts.index = (prng.asfreq('M', 'e') + 1).asfreq('H', 's') + 9
ts.head()
Out[116]:
Categoricals¶
In [117]:
df = pd.DataFrame({"id":[1,2,3,4,5,6], "raw_grade":['a', 'b', 'b', 'a', 'a', 'e']})
In [118]:
df["grade"] = df["raw_grade"].astype("category")
In [119]:
df["grade"]
Out[119]:
In [120]:
df["grade"].cat.categories = ["very good", "good", "very bad"]
df['grade']
Out[120]:
In [121]:
df["grade"] = df["grade"].cat.set_categories(["very bad", "bad", "medium", "good", "very good"])
df.grade
Out[121]:
In [122]:
df.sort_values(by="grade")
Out[122]:
In [123]:
df.groupby("grade").size()
Out[123]:
Plotting¶
In [124]:
ts = pd.Series(np.random.randn(1000), index=pd.date_range('1/1/2000', periods=1000))
In [125]:
ts = ts.cumsum()
In [126]:
ts.plot()
Out[126]:
In [127]:
df = pd.DataFrame(np.random.randn(1000, 4), index=ts.index,
columns=['A', 'B', 'C', 'D'])
In [128]:
df = df.cumsum()
In [129]:
df.plot()
Out[129]:
CSV¶
In [130]:
df.to_csv('foo.csv')
In [131]:
pd.read_csv('foo.csv').head()
Out[131]:
Excel¶
In [132]:
df.to_excel('foo.xlsx', sheet_name='Sheet1')
In [133]:
pd.read_excel('foo.xlsx', 'Sheet1', index_col=None, na_values=['NA']).head() ##requires xlrd
Out[133]: