import numpy as np
import pandas as pd
import matplotlib
file = "U.S._natural_gas_production.csv"
gas_df = pd.read_csv(file, decimal=',')
gas_df.head(3)
Monthobject
U.S.int64
0
2008-06-01
70626
1
2008-07-01
71137
2
2008-08-01
70760
gas_df.shape
gas_df.columns
gas_df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 121 entries, 0 to 120
Data columns (total 20 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Month 121 non-null object
1 U.S. 121 non-null int64
2 Alaska 121 non-null int64
3 Arkansas 121 non-null int64
4 Kansas 121 non-null int64
5 Other States 121 non-null int64
6 Federal Offshore--Gulf of Mexico 121 non-null int64
7 Wyoming 121 non-null int64
8 West Virginia 121 non-null int64
9 Utah 121 non-null int64
10 Pennsylvania 121 non-null int64
11 Texas 121 non-null int64
12 Oklahoma 121 non-null int64
13 Ohio 121 non-null int64
14 North Dakota 121 non-null int64
15 New Mexico 121 non-null int64
16 Montana 121 non-null int64
17 Louisiana 121 non-null int64
18 Colorado 121 non-null int64
19 California 121 non-null int64
dtypes: int64(19), object(1)
memory usage: 19.0+ KB
gas_df["Month"] = pd.to_datetime(gas_df["Month"])
month_type = gas_df['Month'].dtype
gas_df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 121 entries, 0 to 120
Data columns (total 20 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Month 121 non-null datetime64[ns]
1 U.S. 121 non-null int64
2 Alaska 121 non-null int64
3 Arkansas 121 non-null int64
4 Kansas 121 non-null int64
5 Other States 121 non-null int64
6 Federal Offshore--Gulf of Mexico 121 non-null int64
7 Wyoming 121 non-null int64
8 West Virginia 121 non-null int64
9 Utah 121 non-null int64
10 Pennsylvania 121 non-null int64
11 Texas 121 non-null int64
12 Oklahoma 121 non-null int64
13 Ohio 121 non-null int64
14 North Dakota 121 non-null int64
15 New Mexico 121 non-null int64
16 Montana 121 non-null int64
17 Louisiana 121 non-null int64
18 Colorado 121 non-null int64
19 California 121 non-null int64
dtypes: datetime64[ns](1), int64(19)
memory usage: 19.0 KB
gas_df['Month'].dtype
gas_df['Month'].dt.year.head()
gas_df['Month'].dt.month.tail()
yearly_gas_df = gas_df.groupby(gas_df["Month"].dt.year).sum()
yearly_gas_df
U.S.int64
489473 - 1090730
Alaskaint64
54887 - 108986
2008
489473
64732
2009
856876
108986
2010
881622
105219
2011
936163
104154
2012
968632
103856
2013
970617
105776
2014
1032439
104298
2015
1082158
104447
2016
1070085
105900
2017
1090730
106969
plot = yearly_gas_df.filter(items=['U.S.']).plot(kind="bar")
plot.set_xlabel("Year");
filtered_yearly_gas_df = yearly_gas_df[1:-1]
filtered_yearly_gas_df
U.S.int64
856876 - 1090730
Alaskaint64
103856 - 108986
2009
856876
108986
2010
881622
105219
2011
936163
104154
2012
968632
103856
2013
970617
105776
2014
1032439
104298
2015
1082158
104447
2016
1070085
105900
2017
1090730
106969
plot = filtered_yearly_gas_df.filter(items=['U.S.']).plot(kind="bar")
plot.set_xlabel("Year");
filtered_yearly_gas_df.columns[1:].sort_values()
plot = filtered_yearly_gas_df.filter(items=['Colorado', 'Louisiana', 'Ohio', 'Utah']).plot()
plot.set_xlabel("Year");
file = "U.S._crude_oil_production.csv"
oil_df = pd.read_csv(file, decimal=',')
oil_df.head(3)
Monthobject
U.S. Crude Oil int64
0
2008-06-01
5138
1
2008-07-01
5177
2
2008-08-01
5003
oil_df["Month"] = pd.to_datetime(oil_df["Month"])
month_type_oil = oil_df['Month'].dtype
yearly_oil_df = oil_df.groupby(oil_df["Month"].dt.year).sum()
yearly_oil_df
U.S. Crude Oil int64
34211 - 113179
Alabamaint64
99 - 342
2008
34211
149
2009
64180
236
2010
65741
234
2011
67829
275
2012
78007
312
2013
89585
342
2014
105064
323
2015
113179
318
2016
105970
266
2017
112205
224
yearly_oil_df.columns = yearly_oil_df.columns.str.strip()
yearly_oil_df
U.S. Crude Oilint64
34211 - 113179
Alabamaint64
99 - 342
2008
34211
149
2009
64180
236
2010
65741
234
2011
67829
275
2012
78007
312
2013
89585
342
2014
105064
323
2015
113179
318
2016
105970
266
2017
112205
224
yearly_oil_df.filter(items=['U.S. Crude Oil']).plot(kind='bar');
filtered_yearly_oil_df = yearly_oil_df[1:-1]
total_gas = filtered_yearly_gas_df.filter(items=['U.S.'])
total_gas.columns = [ 'Gas' ]
display(total_gas.head(3))
total_oil = filtered_yearly_oil_df.filter(items=['U.S. Crude Oil'])
total_oil.columns = [ 'Crude Oil' ]
total_oil.head(3)
Crude Oilint64
2009
64180
2010
65741
2011
67829
merged_df = pd.concat([total_oil, total_gas], axis='columns')
merged_df
Crude Oilint64
64180 - 113179
Gasint64
856876 - 1090730
2009
64180
856876
2010
65741
881622
2011
67829
936163
2012
78007
968632
2013
89585
970617
2014
105064
1032439
2015
113179
1082158
2016
105970
1070085
2017
112205
1090730
plot = merged_df.plot(kind="bar")
plot.set_xlabel("Year")
plot.legend(['Gas (Millions of Cubic feet)', 'Crude Oil (Thousands of barrels)']);