Analysis Results #36

styoung3 · 2022-01-27T16:50:42Z

A couple code updates. I'm sure the code could be improved further, but the analysis results in the cost_and_time estimates notebook appear to be informative. I still don't have values for micro-mobility modes. Will ping the people with those values again.

Added a notebook to display cost impact estimates and added function stubs to scaffolding for calculating cost impact.

Created a separate file for testing functions in scaffolding.

Added cost_time.csv, read in through the mapping_dictionaries.ipynb and added the function call in cost_time_impact_estimates.

Basically doing all the same stuff for cost and time as for energy and CO2.

Just adding .DS_Store to .gitignore from the auxiliary files as well. .DS_Store in other directories already added.

commit 1986947 Author: Young, Stanley A <Stanley.A.Young@nrel.gov> Date: Fri Jan 21 14:42:35 2022 -0700 Update .gitignore *.DS_store Ignore any .DS_store file that is created by MacOS. commit d8445f7 Author: Young, Stanley A <Stanley.A.Young@nrel.gov> Date: Fri Jan 21 14:35:58 2022 -0700 Reset Notebook Output (again) I missed one the first time. I also think I removed the .DS_stores, but I don't see that in the commits here. commit b1fe3fa Author: Young, Stanley A <Stanley.A.Young@nrel.gov> Date: Fri Jan 21 14:24:26 2022 -0700 Create unit_tests.py Unit testing of energy_intensity and energy_impact functions in scaffolding with manually calculated expectations. commit 93fcc0b Author: Young, Stanley A <Stanley.A.Young@nrel.gov> Date: Tue Jan 18 13:06:47 2022 -0700 Reset Notebook Outputs commit a295529 Author: Young, Stanley A <Stanley.A.Young@nrel.gov> Date: Tue Jan 18 11:10:23 2022 -0700 Add .DS_Store to .gitignore

I thought I already did this. I guess I have to do it again though.

Why does this keep getting committed>? I added it to the .gitignore!

Defined speed function stub and started test of speed function. Loaded available constants for cost into cost_time.csv. Added code to mapping_dictionaries.ipynb for loading cost_time constants into dataframe.

Calculated duration in the unit_conversions function in scaffolding for input into the calc_avg_speed function.

Anticipate speeds in MPH, changing the duration feature to hours.

Using pre-existing duration feature now. Assuming it is in seconds.

Calculate average speed defined and tested.

Made most of the changes requested in the pull_request number 31. Did not separate the unit_test classes into separate files.

Forgot some documentation updates last time. Fixing this time around.

I also did some other rummaging around. Nothing that is finalized though.

Don't need time.csv, or the code that saves the mean/median duration to csv. Just return the average durations from the function.

Reformatted calc_avg_dura to return a pandas series with average duration for each mode instead of saving as a csv file.

Some work towards generalizing the energy, cost, and time calculations on per trip basis functions. Also delete row in cost.csv.

energy_intensity, cost, and time generalized with function eng_feat. Testing yet to be done!

Feature engine sounded better to me. I also made some changes to error checking and to the order of parameters.

Tested general function feat_eng. Did basic debugging to ensure passed basic functionality testing.

Erroneous file in the directory, deleted. Added testing for cost and time in TestEnergyIntensity.

Tested and debug feat_eng for use in functions energy_intensity, cost, and time.

These are the analysis results. The graphs are in a clean state. There is more cleaning that can be done in code, but analysis done.

shankari · 2022-01-27T18:46:46Z

@styoung3 I brought up the need for additional cost estimates at the last MITIE task update meeting, and both Bingrong and Alana were also interested in micromobility total cost of ownership (TCO). We discussed writing a short paper on the results since there does not appear to be any existing literature on the topic.

shankari

@styoung3 Quick review before a deeper dive into the code.

Please remove viz_scripts/.DS_Store
remove the now obsolete visualization code

shankari · 2022-01-27T18:49:00Z

viz_scripts/plots.py

@@ -181,16 +181,16 @@ def overeall_energy_impact(x,y,color,data,plot_title,file_name):



-def energy_impact(x,y,color,plot_title,file_name):
+def energy_impact(x,y,color,plot_title,file_name,xl='Energy_Impact(kWH)'):


please rename to impact_plot or similar, remove the existing CO2_impact and replace it by a call to this function.

All I did was add a parameter to this one. I didn't actually refactor it. Do you still want to change the name?

You also changed the way in which the labels are generated, so it works for any metric, not just energy, right?
So you should change the name, and also replace the implementation of the existing co2_impact function with a call to this one.

viz_scripts/plots.py

viz_scripts/energy_calculations.ipynb

shankari · 2022-01-27T18:53:25Z

viz_scripts/scaffolding.py

+def feat_eng(data, const, feats, prefs, mode='Mode_confirm', repm='Replaced_mode'):
+    """
+    Calculate trip aggregate results from constants and append to data (Feature Engine)
+
+    Parameters:


Ah, I see that you have refactored more code 😄
I will have to look at it in detail but glad to see that you are thinking about this.

This is actually the generalized function for calculating energy_intensity, cost, and time for each trip. I have yet to generalize the impact calculation function.

Yup. You are definitely on the right track with this change. I am actually OK with leaving *_impact functions unchanged - the number of duplicated lines is very small. But of course, unifying them would be better...

shankari · 2022-02-01T06:31:18Z

viz_scripts/mapping_dictionaries.ipynb

+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8a25d7e2",
+   "metadata": {},
+   "outputs": [],
+   "source": []


remove extra cell as well

shankari

Looks good overall. Most of my pending comments are around simple renaming/extraneous code and parameter deletion, and splitting into multiple sections.

shankari · 2022-02-01T06:40:51Z

viz_scripts/scaffolding.py

+
+    return data
+
+

 def energy_intensity(df,df1,distance,col1,col2):


If the distance is unused in energy_intensity, why not just remove it?

shankari · 2022-02-01T06:49:48Z

viz_scripts/scaffolding.py

+    # Feature engine!
+    for i in range(len(feats)):


Suggested change

# Feature engine!

for i in range(len(feats)):

# Feature engine!

# iterating by index (i) instead of `for feat in feats:` to index into prefs and create the new feature names

for i in range(len(feats)):

shankari · 2022-02-01T06:50:39Z

viz_scripts/scaffolding.py

+    df['duration_h'] = df['duration'] / 60 / 60 #seconds to hours
+
+
+def feat_eng(data, const, feats, prefs, mode='Mode_confirm', repm='Replaced_mode'):


is there a unit test for feat_eng or only for the wrapper functions?

shankari · 2022-02-01T06:55:48Z

viz_scripts/scaffolding.py

+            # Create new feature in data
+            fn = prefs[i]+m
+            data[fn] = data[m].map(dic)
+            print('Created ' + fn + ' feature in data.')


This is a bit tricky, so it would be nice to add some examples here - e.g.

Suggested change

# Create new feature in data

fn = prefs[i]+m

data[fn] = data[m].map(dic)

print('Created ' + fn + ' feature in data.')

# Create new feature in data

# prefs are [`cost__trip_`] or similar, and the mode and replaced_mode strings are passed in from the call

# location, so fn is `cost__trip_Mode_confirm`

fn = prefs[i]+m

# maps data["Mode_confirm"] against the dictionary

# and then sets `cost__trip_Mode_confirm` to it.

data[fn] = data[m].map(dic)

print('Created ' + fn + ' feature in data.')

shankari · 2022-02-01T06:57:23Z

viz_scripts/run_unit_tests.py

+import numpy as np
+import scaffolding
+
+class TestEnergyIntensity(unittest.TestCase):


Please remember to split the tests into multiple test files.

jpfleischer · 2024-09-07T04:06:28Z

I will inherit this PR

styoung3 and others added 30 commits January 11, 2022 15:42

Adding Cost Impact

70979df

Added a notebook to display cost impact estimates and added function stubs to scaffolding for calculating cost impact.

Create tests.py

f6b8f46

Created a separate file for testing functions in scaffolding.

Added Cost and Time Calculations by Mode

840ff24

Same as before?

1fa759a

Added time_impact and cost_impact functions.

18d4124

cost_time.csv updates

e9e1dfd

Added cost_time.csv, read in through the mapping_dictionaries.ipynb and added the function call in cost_time_impact_estimates.

Debugging

dca163d

debugging

545fdc0

debugging

636f0de

?

6ecfa26

blah

fc96fed

Mimicking Energy Calculations

406fbe5

Basically doing all the same stuff for cost and time as for energy and CO2.

Added energy_intensity() unit_test

5d8fdc1

Unit Test Energy Impact

55cd7cd

Merge branch 'stanley'

abe79ac

.DS_Store in aux_files

ae12f9d

Just adding .DS_Store to .gitignore from the auxiliary files as well. .DS_Store in other directories already added.

Add Unit Tests (again?)

e3c264e

I thought I already did this. I guess I have to do it again though.

Merge branch 'e-mission:main' into main

2600f92

Remove .DS_Store (again)

a269ee2

Why does this keep getting committed>? I added it to the .gitignore!

Merge branch 'main' of https://github.com/styoung3/em-public-dashboard

a67ac28

Function + constants + load

f118ab3

Defined speed function stub and started test of speed function. Loaded available constants for cost into cost_time.csv. Added code to mapping_dictionaries.ipynb for loading cost_time constants into dataframe.

unit_conversions + duration

01f583d

Calculated duration in the unit_conversions function in scaffolding for input into the calc_avg_speed function.

Convert duration to hours

c37fd3d

Anticipate speeds in MPH, changing the duration feature to hours.

Discovered pre-existing duration feature

afe129c

Using pre-existing duration feature now. Assuming it is in seconds.

Calculate Average Speed

45f568b

Calculate average speed defined and tested.

Made Some Changes

f951740

Made most of the changes requested in the pull_request number 31. Did not separate the unit_test classes into separate files.

Documentation Minor Updates

c825124

Forgot some documentation updates last time. Fixing this time around.

Fixed .gitignore Conflict

b42502e

I also did some other rummaging around. Nothing that is finalized though.

Trashed time.csv

4efd130

Don't need time.csv, or the code that saves the mean/median duration to csv. Just return the average durations from the function.

styoung3 added 9 commits January 25, 2022 14:35

Reformat calc_avg_dura

53450cd

Reformatted calc_avg_dura to return a pandas series with average duration for each mode instead of saving as a csv file.

Generalizing function (maybe?)

4c7b3b4

Some work towards generalizing the energy, cost, and time calculations on per trip basis functions. Also delete row in cost.csv.

Merge branch 'main' into generalize_functions

3cb697b

Generalized first functions

0a76b51

energy_intensity, cost, and time generalized with function eng_feat. Testing yet to be done!

Change eng_feat to feat_eng

ca63e31

Feature engine sounded better to me. I also made some changes to error checking and to the order of parameters.

Tested General Function

df70624

Tested general function feat_eng. Did basic debugging to ensure passed basic functionality testing.

Add tests cost and time

1d0b58b

Erroneous file in the directory, deleted. Added testing for cost and time in TestEnergyIntensity.

Applied feat_eng

0fcd995

Tested and debug feat_eng for use in functions energy_intensity, cost, and time.

Added Time and Cost Impact Graphs

8c19cbf

These are the analysis results. The graphs are in a clean state. There is more cleaning that can be done in code, but analysis done.

shankari requested changes Jan 27, 2022

View reviewed changes

shankari reviewed Feb 1, 2022

View reviewed changes

shankari requested changes Feb 1, 2022

View reviewed changes

shankari mentioned this pull request Jun 27, 2022

NREL-hosted version of the public dashboard e-mission/e-mission-docs#743

Open

shankari mentioned this pull request Aug 29, 2023

📈 New "Dashboard" tab - Design Considerations e-mission/e-mission-docs#961

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Analysis Results #36

Analysis Results #36

styoung3 commented Jan 27, 2022

shankari commented Jan 27, 2022 •

edited

Loading

shankari left a comment •

edited

Loading

shankari Jan 27, 2022

styoung3 Jan 28, 2022

shankari Jan 28, 2022

shankari Jan 27, 2022

styoung3 Jan 28, 2022

shankari Jan 28, 2022

shankari Feb 1, 2022

shankari left a comment

shankari Feb 1, 2022

shankari Feb 1, 2022

shankari Feb 1, 2022

shankari Feb 1, 2022

shankari Feb 1, 2022

jpfleischer commented Sep 7, 2024

		@@ -181,16 +181,16 @@ def overeall_energy_impact(x,y,color,data,plot_title,file_name):



		def energy_impact(x,y,color,plot_title,file_name):
		def energy_impact(x,y,color,plot_title,file_name,xl='Energy_Impact(kWH)'):

		df['duration_h'] = df['duration'] / 60 / 60 #seconds to hours


		def feat_eng(data, const, feats, prefs, mode='Mode_confirm', repm='Replaced_mode'):

Analysis Results #36

Are you sure you want to change the base?

Analysis Results #36

Conversation

styoung3 commented Jan 27, 2022

shankari commented Jan 27, 2022 • edited Loading

shankari left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

shankari left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jpfleischer commented Sep 7, 2024

shankari commented Jan 27, 2022 •

edited

Loading

shankari left a comment •

edited

Loading