Related
{'id': {0: 'x1',
1: 'x2',
2: 'x3',
3: 'x4',
4: 'x5',
5: 'x6',
6: 'x7',
7: 'x8',
8: 'x9',
9: 'x10'},
'trt': {0: 'cnt',
1: 'cnt',
2: 'tr',
3: 'tr',
4: 'tr',
5: 'cnt',
6: 'tr',
7: 'tr',
8: 'cnt',
9: 'cnt'},
'work.T1': {0: 0.6516556669957936,
1: 0.567737752571702,
2: 0.1135089821182191,
3: 0.5959253052715212,
4: 0.3580499750096351,
5: 0.4288094183430075,
6: 0.0519033221062272,
7: 0.2641776674427092,
8: 0.3987907308619469,
9: 0.8361341434065253},
'play.T1': {0: 0.8647212258074433,
1: 0.6153524168767035,
2: 0.7751098964363337,
3: 0.3555686913896352,
4: 0.4058499720413238,
5: 0.7066469138953835,
6: 0.8382876652758569,
7: 0.2395891312044114,
8: 0.7707715332508087,
9: 0.3558977444190532},
'talk.T1': {0: 0.5355970377568156,
1: 0.0930881295353174,
2: 0.169803041499108,
3: 0.8998324507847428,
4: 0.4226376069709658,
5: 0.7477464678231627,
6: 0.8226525799836963,
7: 0.9546536463312804,
8: 0.6854445093777031,
9: 0.5005032296758145},
'work.T2': {0: 0.2754838624969125,
1: 0.2289039448369294,
2: 0.0144339059479534,
3: 0.7289645625278354,
4: 0.2498804717324674,
5: 0.1611832766793668,
6: 0.0170426501426845,
7: 0.4861003451514989,
8: 0.1029001718852669,
9: 0.8015470046084374},
'play.T2': {0: 0.3543280649464577,
1: 0.9364325392525644,
2: 0.2458663922734558,
3: 0.4731414613779634,
4: 0.191560871200636,
5: 0.5832219698932022,
6: 0.4594731898978352,
7: 0.467434047954157,
8: 0.3998325555585325,
9: 0.5052855962421745},
'talk.T2': {0: 0.0318881559651345,
1: 0.1144675880204886,
2: 0.468935475917533,
3: 0.3969867376144975,
4: 0.8336191941052675,
5: 0.7611217433586717,
6: 0.5733564489055425,
7: 0.447508045937866,
8: 0.0838020080700516,
9: 0.2191385473124683}}
mydt = pd.DataFrame(dt, columns = ['id’, ‘trt’, 'work.T1', '', ‘play.T1’, ‘talk.T1’'work.T2', '', ‘play.T2’, ‘talk.T2’])
So I have the above dataset. How can I convert the above dataset so that it looks like the following where each unique id has a an observation at both T1 and T2? Please let me know.
You can use pd.wide_to_long:
names = ['play', 'talk', 'work']
out = (pd.wide_to_long(df, names, i='id', j='time', sep='.', suffix='T\d')
.sort_index().reset_index())
Output:
>>> out
id time trt play talk work
0 x1 T1 cnt 0.010651 0.400500 0.774567
1 x1 T2 cnt 0.758084 0.147504 0.831137
2 x2 T1 tr 0.540774 0.960239 0.332556
3 x2 T2 tr 0.882142 0.884740 0.020044
4 x3 T1 tr 0.184434 0.842003 0.909759
5 x3 T2 tr 0.108465 0.234251 0.954427
6 x4 T1 tr 0.457186 0.567753 0.095170
7 x4 T2 tr 0.553769 0.768538 0.401176
8 x5 T1 cnt 0.787308 0.036411 0.179075
9 x5 T2 cnt 0.002033 0.745652 0.638177
10 x6 T1 tr 0.954682 0.163233 0.074820
11 x6 T2 tr 0.044682 0.875726 0.538481
I have the following dataframes (this is just test data), in real samples, I have index values that are repeated a few times inside dataframe 1 and dataframe 2 - this causes the repeated/duplicate rows inside final dataframe.
DataFrame 1:
pd.DataFrame({'id': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'first_name': {0: 'Jennee',
1: 'Dagny',
2: 'Correy',
3: 'Pall',
4: 'Julie',
5: 'Janene',
6: 'Lemmy',
7: 'Coleman',
8: 'Beck',
9: 'Che'},
'last_name': {0: 'Strelitzki',
1: 'Dunsire',
2: 'Wickrath',
3: 'Jopp',
4: 'Gheeraert',
5: 'Gawith',
6: 'Farrow',
7: 'Legging',
8: 'Beckwith',
9: 'Burgoin'},
'email': {0: 'jstrelitzki0#google.de',
1: 'ddunsire1#geocities.com',
2: 'cwickrath2#github.com',
3: 'pjopp3#infoseek.co.jp',
4: 'jgheeraert4#theatlantic.com',
5: 'jgawith5#sciencedirect.com',
6: 'lfarrow6#wikimedia.org',
7: 'clegging7#businessinsider.com',
8: 'bbeckwith8#zdnet.com',
9: 'cburgoin9#reference.com'},
'gender': {0: 'Male',
1: 'Female',
2: 'Female',
3: 'Female',
4: 'Female',
5: 'Female',
6: 'Male',
7: 'Female',
8: 'Polygender',
9: 'Male'},
'ip_address': {0: '8.99.68.120',
1: '188.238.129.48',
2: '87.159.243.249',
3: '66.37.174.94',
4: '233.77.128.104',
5: '190.202.131.98',
6: '84.175.231.196',
7: '140.178.100.5',
8: '81.211.179.167',
9: '31.219.69.206'},
'Boolean': {0: False,
1: False,
2: True,
3: True,
4: False,
5: True,
6: True,
7: False,
8: False,
9: False}})
DataFrame 2:
pd.DataFrame({'id': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'Model': {0: 2005,
1: 2007,
2: 2011,
3: 2003,
4: 1998,
5: 1992,
6: 1992,
7: 1992,
8: 2008,
9: 1996},
'Make': {0: 'Cadillac',
1: 'Lexus',
2: 'Dodge',
3: 'Dodge',
4: 'Oldsmobile',
5: 'Volkswagen',
6: 'Chevrolet',
7: 'Suzuki',
8: 'Ford',
9: 'Mazda'},
'Colour': {0: 'Red',
1: 'Red',
2: 'Crimson',
3: 'Red',
4: 'Purple',
5: 'Crimson',
6: 'Red',
7: 'Aquamarine',
8: 'Puce',
9: 'Maroon'}})
The two dataframes should be connected based on common Index values found in both dataframes only. Which means, any index values that don't match in those two dataframes; should not appear in the final combined/merged dataframe.
I want to ensure that the final dataframe is unique, and only captures combinations of columns, based on unique Index values.
When I try using the following code, the output is supposed to 'inner join' based on the unique index found in both dataframes.
final = pd.merge(df1, df2, left_index=True, right_index=True)
However, when I try applying the above merge technique on my larger (other) pandas dataframes, there are many rows being repeated/duplicated multiple times. When the merging happpens a few times with more dataframes, the rows gets repeated very frequently, with the same Index value.
I am expecting to see one Index value returned per row (with all the column combinations from each dataframe).
I am not sure why this happens. I can confirm that there is nothing wrong with the datasets.
Is there a better technique of merging those two dataframes, based on only common index values, and at the same time ensure that I don't repeat any rows (with the same index) in my final dataframe ? I often find that this merging often creates a giant final CSV file around 20GB in size too. The source files are only around 15MB into total.
Any help is much appreciated.
My end output should look like this (please copy and use this as Pandas DF):
pd.DataFrame({'id': {0: 1, 1: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10},
'first_name': {0: 'Jennee',
1: 'Dagny',
2: 'Correy',
3: 'Pall',
4: 'Julie',
5: 'Janene',
6: 'Lemmy',
7: 'Coleman',
8: 'Beck',
9: 'Che'},
'last_name': {0: 'Strelitzki',
1: 'Dunsire',
2: 'Wickrath',
3: 'Jopp',
4: 'Gheeraert',
5: 'Gawith',
6: 'Farrow',
7: 'Legging',
8: 'Beckwith',
9: 'Burgoin'},
'email': {0: 'jstrelitzki0#google.de',
1: 'ddunsire1#geocities.com',
2: 'cwickrath2#github.com',
3: 'pjopp3#infoseek.co.jp',
4: 'jgheeraert4#theatlantic.com',
5: 'jgawith5#sciencedirect.com',
6: 'lfarrow6#wikimedia.org',
7: 'clegging7#businessinsider.com',
8: 'bbeckwith8#zdnet.com',
9: 'cburgoin9#reference.com'},
'gender': {0: 'Male',
1: 'Female',
2: 'Female',
3: 'Female',
4: 'Female',
5: 'Female',
6: 'Male',
7: 'Female',
8: 'Polygender',
9: 'Male'},
'ip_address': {0: '8.99.68.120',
1: '188.238.129.48',
2: '87.159.243.249',
3: '66.37.174.94',
4: '233.77.128.104',
5: '190.202.131.98',
6: '84.175.231.196',
7: '140.178.100.5',
8: '81.211.179.167',
9: '31.219.69.206'},
'Boolean': {0: False,
1: False,
2: True,
3: True,
4: False,
5: True,
6: True,
7: False,
8: False,
9: False},
'Model': {0: 2005,
1: 2007,
2: 2011,
3: 2003,
4: 1998,
5: 1992,
6: 1992,
7: 1992,
8: 2008,
9: 1996},
'Make': {0: 'Cadillac',
1: 'Lexus',
2: 'Dodge',
3: 'Dodge',
4: 'Oldsmobile',
5: 'Volkswagen',
6: 'Chevrolet',
7: 'Suzuki',
8: 'Ford',
9: 'Mazda'},
'Colour': {0: 'Red',
1: 'Red',
2: 'Crimson',
3: 'Red',
4: 'Purple',
5: 'Crimson',
6: 'Red',
7: 'Aquamarine',
8: 'Puce',
9: 'Maroon'}})
This is expected behavior with non-unique idx values. Since you have 3 ID1 rows in one df and 2 ID1 in the other, you end up with 6 ID1 rows in your merged df. If you add validate="one_to_one" to pd.merge() you will get this Error. MergeError: Merge keys are not unique in either left or right dataset; not a one-to-one mergeAll other validations fail except for many to many.
If it makes sense for your data, you can use the left_on, and right_on parameters to find unique combinations and give you a one-to-one if that's what you're after.
Edit after your new data:
Now that you have unique ids, this should work for you. Notice it doesn't throw a validation error.
final = pd.merge(df1, df2, left_on=['id'], right_on=['id'], validate='one_to_one')
I want to plot by Step Typ: Traction and Stribeck. The different load stages should have his own plot. At the respective load level, the line plots should be broken down by temperature. y-axis is Traction (-) and x-axis the respective counterpart SRR (%) or Rolling speed (mm/s) (for Traction and Stribeck respectively). At the end, I should have four different plots.
Example, how it should look like:
My attempt so far, which leads to an empty plot.
import pandas as pd
import matplotlib.pyplot as plt
data = {'Step 1': {'Step Typ': 'Traction', 'SRR (%)': {1: 8.384, 2: 9.815, 3: 7.531, 4: 10.209, 5: 7.989, 6: 7.331, 7: 5.008, 8: 2.716, 9: 9.6, 10: 7.911}, 'Traction (-)': {1: 5.602, 2: 6.04, 3: 2.631, 4: 2.952, 5: 8.162, 6: 9.312, 7: 4.994, 8: 2.959, 9: 10.075, 10: 5.498}, 'Temperature': 30, 'Load': 40}, 'Step 3': {'Step Typ': 'Traction', 'SRR (%)': {1: 2.909, 2: 5.552, 3: 5.656, 4: 9.043, 5: 3.424, 6: 7.382, 7: 3.916, 8: 2.665, 9: 4.832, 10: 3.993}, 'Traction (-)': {1: 9.158, 2: 6.721, 3: 7.787, 4: 7.491, 5: 8.267, 6: 2.985, 7: 5.882, 8: 3.591, 9: 6.334, 10: 10.43}, 'Temperature': 80, 'Load': 40}, 'Step 5': {'Step Typ': 'Traction', 'SRR (%)': {1: 4.765, 2: 9.293, 3: 7.608, 4: 7.371, 5: 4.87, 6: 4.832, 7: 6.244, 8: 6.488, 9: 5.04, 10: 2.962}, 'Traction (-)': {1: 6.656, 2: 7.872, 3: 8.799, 4: 7.9, 5: 4.22, 6: 6.288, 7: 7.439, 8: 7.77, 9: 5.977, 10: 9.395}, 'Temperature': 30, 'Load': 70}, 'Step 7': {'Step Typ': 'Traction', 'SRR (%)': {1: 9.46, 2: 2.83, 3: 3.249, 4: 9.273, 5: 8.792, 6: 9.673, 7: 6.784, 8: 3.838, 9: 8.779, 10: 4.82}, 'Traction (-)': {1: 5.245, 2: 8.491, 3: 10.088, 4: 9.988, 5: 4.886, 6: 4.168, 7: 8.628, 8: 5.038, 9: 7.712, 10: 3.961}, 'Temperature': 80, 'Load': 70}, 'Step 2': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 4.862, 2: 4.71, 3: 4.537, 4: 6.35, 5: 6.691, 6: 5.337, 7: 8.419, 8: 10.303, 9: 5.018, 10: 10.195}, 'Traction (-)': {1: 6.674, 2: 10.137, 3: 2.822, 4: 5.494, 5: 9.986, 6: 9.095, 7: 3.53, 8: 6.96, 9: 8.251, 10: 7.836}, 'Temperature': 30, 'Load': 40}, 'Step 4': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 4.04, 2: 8.288, 3: 3.731, 4: 10.137, 5: 5.32, 6: 8.504, 7: 5.917, 8: 9.677, 9: 8.641, 10: 7.685}, 'Traction (-)': {1: 9.522, 2: 4.749, 3: 3.46, 4: 3.21, 5: 5.005, 6: 9.886, 7: 8.023, 8: 5.935, 9: 8.74, 10: 5.117}, 'Temperature': 80, 'Load': 40}, 'Step 6': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 6.244, 2: 7.015, 3: 5.998, 4: 4.894, 5: 6.117, 6: 6.644, 7: 7.619, 8: 10.477, 9: 9.61, 10: 2.958}, 'Traction (-)': {1: 7.353, 2: 7.98, 3: 6.675, 4: 8.853, 5: 7.537, 6: 5.256, 7: 4.923, 8: 10.293, 9: 2.873, 10: 10.407}, 'Temperature': 30, 'Load': 70}, 'Step 8': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 3.475, 2: 2.756, 3: 7.809, 4: 9.449, 5: 2.72, 6: 4.133, 7: 10.139, 8: 10.0, 9: 3.71, 10: 8.267}, 'Traction (-)': {1: 6.307, 2: 2.83, 3: 9.258, 4: 3.405, 5: 9.659, 6: 6.662, 7: 6.413, 8: 6.488, 9: 7.972, 10: 6.288}, 'Temperature': 80, 'Load': 70} }
df = pd.DataFrame(data)
items = list()
series = list()
for item, d in data.items():
items.append(item)
series.append(pd.DataFrame.from_dict(d))
df = pd.concat(series, keys=items)
df.set_index(['Step Typ', 'Load', 'Temperature'], inplace=True)
df.loc[('Stribeck')]
for force, _ in df.groupby(level=1):
plt.figure(figsize=(15, 12))
for i, row in df.loc[('Traction'), force].iterrows():
plt.ylim(0, 0.1)
plt.ylabel('Traction Coeff (-)')
plt.xlabel('Rolling Speed (mm/s)')
plt.title('Title comes later', loc='left')
plt.plot(row['Rolling Speed (mm/s)'], row['Traction (-)'], label=f"{i} - {force}")
print(f"{i} - {force}")
plt.show()
I have changed your plotting loop. The code below will generate two plots for Traction (one for each Load value), where each has two curves (one for each temperature). I have commented the line where you set the ylim(a, b) because this could lead to empty plot if data fall out of (a, b) range.
import pandas as pd
import matplotlib.pyplot as plt
data = {'Step 1': {'Step Typ': 'Traction', 'SRR (%)': {1: 8.384, 2: 9.815, 3: 7.531, 4: 10.209, 5: 7.989, 6: 7.331, 7: 5.008, 8: 2.716, 9: 9.6, 10: 7.911}, 'Traction (-)': {1: 5.602, 2: 6.04, 3: 2.631, 4: 2.952, 5: 8.162, 6: 9.312, 7: 4.994, 8: 2.959, 9: 10.075, 10: 5.498}, 'Temperature': 30, 'Load': 40}, 'Step 3': {'Step Typ': 'Traction', 'SRR (%)': {1: 2.909, 2: 5.552, 3: 5.656, 4: 9.043, 5: 3.424, 6: 7.382, 7: 3.916, 8: 2.665, 9: 4.832, 10: 3.993}, 'Traction (-)': {1: 9.158, 2: 6.721, 3: 7.787, 4: 7.491, 5: 8.267, 6: 2.985, 7: 5.882, 8: 3.591, 9: 6.334, 10: 10.43}, 'Temperature': 80, 'Load': 40}, 'Step 5': {'Step Typ': 'Traction', 'SRR (%)': {1: 4.765, 2: 9.293, 3: 7.608, 4: 7.371, 5: 4.87, 6: 4.832, 7: 6.244, 8: 6.488, 9: 5.04, 10: 2.962}, 'Traction (-)': {1: 6.656, 2: 7.872, 3: 8.799, 4: 7.9, 5: 4.22, 6: 6.288, 7: 7.439, 8: 7.77, 9: 5.977, 10: 9.395}, 'Temperature': 30, 'Load': 70}, 'Step 7': {'Step Typ': 'Traction', 'SRR (%)': {1: 9.46, 2: 2.83, 3: 3.249, 4: 9.273, 5: 8.792, 6: 9.673, 7: 6.784, 8: 3.838, 9: 8.779, 10: 4.82}, 'Traction (-)': {1: 5.245, 2: 8.491, 3: 10.088, 4: 9.988, 5: 4.886, 6: 4.168, 7: 8.628, 8: 5.038, 9: 7.712, 10: 3.961}, 'Temperature': 80, 'Load': 70}, 'Step 2': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 4.862, 2: 4.71, 3: 4.537, 4: 6.35, 5: 6.691, 6: 5.337, 7: 8.419, 8: 10.303, 9: 5.018, 10: 10.195}, 'Traction (-)': {1: 6.674, 2: 10.137, 3: 2.822, 4: 5.494, 5: 9.986, 6: 9.095, 7: 3.53, 8: 6.96, 9: 8.251, 10: 7.836}, 'Temperature': 30, 'Load': 40}, 'Step 4': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 4.04, 2: 8.288, 3: 3.731, 4: 10.137, 5: 5.32, 6: 8.504, 7: 5.917, 8: 9.677, 9: 8.641, 10: 7.685}, 'Traction (-)': {1: 9.522, 2: 4.749, 3: 3.46, 4: 3.21, 5: 5.005, 6: 9.886, 7: 8.023, 8: 5.935, 9: 8.74, 10: 5.117}, 'Temperature': 80, 'Load': 40}, 'Step 6': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 6.244, 2: 7.015, 3: 5.998, 4: 4.894, 5: 6.117, 6: 6.644, 7: 7.619, 8: 10.477, 9: 9.61, 10: 2.958}, 'Traction (-)': {1: 7.353, 2: 7.98, 3: 6.675, 4: 8.853, 5: 7.537, 6: 5.256, 7: 4.923, 8: 10.293, 9: 2.873, 10: 10.407}, 'Temperature': 30, 'Load': 70}, 'Step 8': {'Step Typ': 'Stribeck', 'Rolling Speed (mm/s)': {1: 3.475, 2: 2.756, 3: 7.809, 4: 9.449, 5: 2.72, 6: 4.133, 7: 10.139, 8: 10.0, 9: 3.71, 10: 8.267}, 'Traction (-)': {1: 6.307, 2: 2.83, 3: 9.258, 4: 3.405, 5: 9.659, 6: 6.662, 7: 6.413, 8: 6.488, 9: 7.972, 10: 6.288}, 'Temperature': 80, 'Load': 70} }
df = pd.DataFrame(data)
items = list()
series = list()
for item, d in data.items():
items.append(item)
series.append(pd.DataFrame.from_dict(d))
df = pd.concat(series, keys=items)
df.set_index(['Step Typ', 'Load', 'Temperature'], inplace=True)
for force, _ in df.groupby(level=1):
fig, ax = plt.subplots(figsize=(8, 6))
df_step = df.loc[('Traction'), force]
for temperature in df_step.index.unique():
df_temp = df_step.loc[temperature].sort_values('SRR (%)')
# ax.set_ylim(0, 0.1)
ax.set_ylabel('Traction Coeff (-)')
ax.set_xlabel('SRR (%)')
ax.set_title('Title comes later', loc='left')
ax.plot(df_temp['SRR (%)'], df_temp['Traction (-)'], label = f'T = {df_temp.index.unique().values[0]}°C - Load = {force}')
ax.legend(frameon = True)
plt.show()
I am trying to plot a line graph to show the trends of each key of a dictionary in Jupyter Notebook with Python. This is what I have in the k_rmse_values variable as shown below:
k_rmse_values =
{'bore': {1: 8423.759328233446,
3: 6501.928933614838,
5: 6807.187615513473,
7: 6900.29659028346,
9: 7134.8868708101645},
'city-mpg': {1: 4265.365592771621,
3: 3865.0178306330113,
5: 3720.409335758634,
7: 3819.183283405616,
9: 4219.677972675927},
'compression-rate': {1: 7016.906657495168,
3: 7319.354017489066,
5: 6301.624922763969,
7: 6133.006310754547,
9: 6417.253959732598},
'curb-weight': {1: 3950.9888180049306,
3: 4201.343428000144,
5: 4047.052502155118,
7: 3842.0974736649846,
9: 3943.9478256384205},
'engine-size': {1: 2853.7338453331627,
3: 2793.6254775629623,
5: 3123.320055069605,
7: 2941.73029681235,
9: 2931.996240628853},
'height': {1: 6330.178232877807,
3: 7049.500497198366,
5: 6869.570862695864,
7: 6738.641089739572,
9: 6344.062937760911},
'highway-mpg': {1: 4826.0580187146525,
3: 3510.253629329685,
5: 3379.2250123364083,
7: 4044.271135312068,
9: 4462.027046251678},
'horsepower': {1: 3623.6389886411143,
3: 4294.825669466819,
5: 4778.254807521257,
7: 4730.538701514935,
9: 4662.8601512508885},
'length': {1: 4952.798701744297,
3: 5403.624431188139,
5: 5500.731909846179,
7: 5103.4515274528885,
9: 4471.077661709427},
'normalized-losses': {1: 9604.929081466453,
3: 7494.820436511842,
5: 6391.912634697067,
7: 6699.853883298577,
9: 6861.6389834002875},
'peak-rpm': {1: 8041.2366213164005,
3: 7502.080095843049,
5: 6521.863037752326,
7: 6869.602542315512,
9: 6884.533017667794},
'stroke': {1: 10330.231237489314,
3: 8947.585146097614,
5: 6973.912792744113,
7: 7266.333478250421,
9: 7026.017456146411},
'wheel-base': {1: 2797.4144312203725,
3: 3392.8627620671928,
5: 4238.25624378706,
7: 4456.687059524217,
9: 4426.032222634904},
'width': {1: 2849.2691940215127,
3: 4076.59327053035,
5: 3979.9751617315405,
7: 3845.3326184519606,
9: 3687.926625900343}}
When I used this code to plot
for k,v in k_rmse_values.items():
x = list(v.keys())
y = list(v.values())
plt.plot(x,y)
plt.xlabel('k value')
plt.ylabel('RMSE')
and it doesn't plot from 1 to 9 in order; it gives this graph
it plots in this k-value order 1, 3, 9 , 5, 7
I have spent hours on this problem and still can't figure out a way to do it. Your help with this would be greatly appreciated.
One solution is to sort the keys and get the matching values:
for k,v in k_rmse_values.items():
xs = list(v.keys()).sort()
ys = [v[x] for x in xs]
# Note I renamed these arrays so following uses should be changed accordingly
I am trying to rename the key and subkey in python nested dictionary. However, I haven't got the result that I expected yet. Below is the original nested key that I have.
nested_dict = {
0: {0: 33.97, 1: 55.32, 2: 57.31, 3: 71.56},
1: {0: 27.31, 1: 23.32, 2: 32.25, 3: 60.21},
2: {0: 65.38, 1: 36.88, 2: 70.88, 3: 21.93},
3: {0: 35.44, 1: 21.21, 2: 40.72, 3: 51.35}
}
I am trying to change the key and subkey to another value into this.
nested_dict = {
4: {4: 33.97, 5: 55.32, 6: 57.31, 7: 71.56},
5: {4: 27.31, 5: 23.32, 6: 32.25, 7: 60.21},
6: {4: 65.38, 5: 36.88, 6: 70.88, 7: 21.93},
7: {4: 35.44, 5: 21.21, 6: 40.72, 7: 51.35}
}
What I have in mind is renaming the key using a list. I have tried to replace the key and subkey with a list below:
new_key = []
for i in range(4,8):
new_key.append(i)
However, I still haven't got it. Another idea is using pandas DataFrame to rename both key and subkey. I am not sure whether using lists or pandas is suitable for the given problem.
Code for renaming a key from here:
mydict[new_key] = mydict.pop(old_key)
You could use a (nested) dict comprehension ([Python]: PEP 274 -- Dict Comprehensions). Note that it generates a new dictionary (but you can assign it to the old variable):
>>> from pprint import pprint as pp
>>>
>>> nested_dict = {
... 0: {0: 33.97, 1: 55.32, 2: 57.31, 3: 71.56},
... 1: {0: 27.31, 1: 23.32, 2: 32.25, 3: 60.21},
... 2: {0: 65.38, 1: 36.88, 2: 70.88, 3: 21.93},
... 3: {0: 35.44, 1: 21.21, 2: 40.72, 3: 51.35}
... }
>>>
>>> pp(nested_dict)
{0: {0: 33.97, 1: 55.32, 2: 57.31, 3: 71.56},
1: {0: 27.31, 1: 23.32, 2: 32.25, 3: 60.21},
2: {0: 65.38, 1: 36.88, 2: 70.88, 3: 21.93},
3: {0: 35.44, 1: 21.21, 2: 40.72, 3: 51.35}}
>>>
>>> modified_nested_dict = {k0 + 4: {k1 + 4: v1 for k1, v1 in v0.items()} for k0, v0 in nested_dict.items()}
>>>
>>> pp(modified_nested_dict)
{4: {4: 33.97, 5: 55.32, 6: 57.31, 7: 71.56},
5: {4: 27.31, 5: 23.32, 6: 32.25, 7: 60.21},
6: {4: 65.38, 5: 36.88, 6: 70.88, 7: 21.93},
7: {4: 35.44, 5: 21.21, 6: 40.72, 7: 51.35}}
You can use Pandas Dataframe for the desired task, as follows:
import pandas as pd
nested_dict = {
0: {0: 33.97, 1: 55.32, 2: 57.31, 3: 71.56},
1: {0: 27.31, 1: 23.32, 2: 32.25, 3: 60.21},
2: {0: 65.38, 1: 36.88, 2: 70.88, 3: 21.93},
3: {0: 35.44, 1: 21.21, 2: 40.72, 3: 51.35}
}
print("Dictionary before renaming: ", nested_dict)
# Convert nested dictionary to Pandas Dataframe
my_dataframe = pd.DataFrame.from_dict(nested_dict)
new_keys = list(range(4, 8)) # List of new keys
my_dataframe.columns = new_keys # Set columns to the new keys
my_dataframe.set_index([new_keys], inplace=True) # Set index to the new keys
nested_dict = my_dataframe.to_dict() # Convert back to nested dictionary
print("Dictionary after renaming: ", nested_dict)
This gives you the following expected output:
Dictionary before renaming: {0: {0: 33.97, 1: 55.32, 2: 57.31, 3: 71.56}, 1: {0: 27.31, 1: 23.32, 2: 32.25, 3: 60.21}, 2: {0: 65.38, 1: 36.88, 2: 70.88, 3: 21.93}, 3: {0: 35.44, 1: 21.21, 2: 40.72, 3: 51.35}}
Dictionary after renaming: {4: {4: 33.97, 5: 55.32, 6: 57.31, 7: 71.56}, 5: {4: 27.31, 5: 23.32, 6: 32.25, 7: 60.21}, 6: {4: 65.38, 5: 36.88, 6: 70.88, 7: 21.93}, 7: {4: 35.44, 5: 21.21, 6: 40.72, 7: 51.35}}