[feature] Plot components of averaged df (and quants) if multiple IDs, but no df_name specified in plot_components() #888

judussoari · 2022-10-21T22:27:18Z

Closes #866

To Do:

Add mean plot of components across all dfs
Add Standard Deviation to plot of averaged df components (as a shade in both directions)
Migrate the code to plot_parameters() instead of plot_cmponents, makes more sense

Before: We raised an Exception whenever multiple time series were present in the input df and plot_paramenters() was called without specifying the time series to be plotted.
Now: plots mean and quants of all time series when time series is not specified in df_name.

Sceenshot 1: model parameters when df_name is specified.
Sceenshot 2: model parameters when df_name is not specified.

…ied in plot_components() *Use received_single_time_series and df_name as indicators *Adjust integration and uncertainy tests - no raise of Exception anymore *Changed former Errors to warnings in this special case * matplotlib and plotly

codecov-commenter · 2022-10-21T22:32:07Z

Codecov Report

Merging #888 (19ec697) into main (18e206b) will increase coverage by 0.61%.
The diff coverage is 97.77%.

@@            Coverage Diff             @@
##             main     #888      +/-   ##
==========================================
+ Coverage   89.19%   89.81%   +0.61%     
==========================================
  Files          18       18              
  Lines        4450     4632     +182     
==========================================
+ Hits         3969     4160     +191     
+ Misses        481      472       -9

Impacted Files	Coverage Δ
neuralprophet/forecaster.py	`88.01% <50.00%> (+0.30%)`	⬆️
neuralprophet/plot_forecast_plotly.py	`87.59% <86.66%> (+0.53%)`	⬆️
neuralprophet/plot_model_parameters_plotly.py	`94.86% <99.02%> (+2.71%)`	⬆️
neuralprophet/plot_model_parameters_matplotlib.py	`90.74% <99.04%> (+3.84%)`	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

…ied in plot_components() *Use received_single_time_series and df_name as indicators *Adjust integration and uncertainy tests - no raise of Exception anymore *Changed former Errors to warnings in this special case * matplotlib and plotly

…lobal

…ied in plot_components() *Use received_single_time_series and df_name as indicators *Adjust integration and uncertainy tests - no raise of Exception anymore *Changed former Errors to warnings in this special case * matplotlib and plotly

…specified *mean_std indicates whether multiple dfs have been passed, without specifying df_name *plot mean and std across all dfs *for plotly

…specified *mean_std indicates whether multiple dfs have been passed, without specifying df_name *plot mean and std across all dfs *for matplotlib

* only if global df, and df_name not specified * only for trend and seasonality * Plotly

* only if global df, and df_name not specified * only for trend and seasonality * Matplotlib

…be available for plot_parameters()

* df_name *global/local normalization *glocal/global model with multiple IDs

…lobal

alfonsogarciadecorral · 2022-10-27T09:39:09Z

HI @judussoari ! Really nice work, this is super useful!
I have a few comments, I will put them below.

Computation time. plotting parameters is talking long. But that is because of numpy on the .groupby().apply()

If you change:

            df_trend_std = df_trend.groupby("ds")[["trend"]].apply(lambda x: np.std(x))
            df_trend = df_trend.groupby("ds")[["trend"]].apply(lambda x: np.mean(x))

by:

            df_trend_std = df_trend.groupby("ds")[["trend"]].apply(lambda x: x.std())
            df_trend = df_trend.groupby("ds")[["trend"]].apply(lambda x: x.mean())

it is much faster.

Kevin-Chen0 · 2022-11-22T05:26:02Z

@judussoari Can you resolve the merge conflicts? Thx

# Conflicts: # neuralprophet/forecaster.py # neuralprophet/plot_forecast_plotly.py # neuralprophet/plot_model_parameters_matplotlib.py # neuralprophet/plot_model_parameters_plotly.py

Kevin-Chen0 · 2022-11-23T23:32:11Z

@judussoari can you resolve the merge conflicts? Thanks.

# Conflicts: # neuralprophet/forecaster.py # neuralprophet/plot_model_parameters_matplotlib.py # neuralprophet/plot_model_parameters_plotly.py # tests/test_integration.py # tests/test_plotting.py

Kevin-Chen0

LGTM

github-actions · 2022-11-27T14:44:08Z

fe1d759

Model Benchmark

Show benchmark results

Benchmark	Metric	main	current	diff
YosemiteTemps	MAE_val	1.72949	1.72949	0.0%	✅
YosemiteTemps	RMSE_val	2.27386	2.27386	0.0%	✅
YosemiteTemps	Loss_val	0.00096	0.00096	0.0%	✅
YosemiteTemps	RegLoss_val	0	0	0.0%	✅
YosemiteTemps	epoch	84	84	0.0%	✅
YosemiteTemps	MAE	1.45189	1.45189	0.0%	✅
YosemiteTemps	RMSE	2.16631	2.16631	0.0%	✅
YosemiteTemps	Loss	0.00066	0.00066	0.0%	✅
YosemiteTemps	RegLoss	0	0	0.0%	✅
PeytonManning	MAE_val	0.64636	0.64636	0.0%	✅
PeytonManning	RMSE_val	0.79276	0.79276	0.0%	✅
PeytonManning	Loss_val	0.01494	0.01494	0.0%	✅
PeytonManning	RegLoss_val	0	0	0.0%	✅
PeytonManning	epoch	37	37	0.0%	✅
PeytonManning	MAE	0.42701	0.42701	0.0%	✅
PeytonManning	RMSE	0.57032	0.57032	0.0%	✅
PeytonManning	Loss	0.00635	0.00635	0.0%	✅
PeytonManning	RegLoss	0	0	0.0%	✅
AirPassengers	MAE_val	15.2698	15.2698	0.0%	✅
AirPassengers	RMSE_val	19.4209	19.4209	0.0%	✅
AirPassengers	Loss_val	0.00195	0.00195	0.0%	✅
AirPassengers	RegLoss_val	0	0	0.0%	✅
AirPassengers	epoch	89	89	0.0%	✅
AirPassengers	MAE	9.82902	9.82902	0.0%	✅
AirPassengers	RMSE	11.7005	11.7005	0.0%	✅
AirPassengers	Loss	0.00056	0.00056	0.0%	✅
AirPassengers	RegLoss	0	0	0.0%	✅

Model Training

PeytonManning

YosemiteTemps

AirPassengers

judussoari added this to the Release 0.4.2 milestone Oct 21, 2022

judussoari self-assigned this Oct 21, 2022

Merge branch 'main' into comp_avg_global

2547553

judussoari added 4 commits October 21, 2022 15:34

Merge remote-tracking branch 'origin/comp_avg_global' into comp_avg_g…

6ea6903

…lobal

judussoari changed the title ~~[feature] Plot components of averaged df if multiple IDs, but no df_name specified in plot_components()~~ [feature] Plot components of averaged df (and std) if multiple IDs, but no df_name specified in plot_components() Oct 21, 2022

judussoari and others added 18 commits October 23, 2022 10:57

Plot components of std across all df if multiple IDs, but no df_name …

31e5d23

…specified *mean_std indicates whether multiple dfs have been passed, without specifying df_name *plot mean and std across all dfs *for plotly

Plot components of std across all df if multiple IDs, but no df_name …

33a22fd

…specified *mean_std indicates whether multiple dfs have been passed, without specifying df_name *plot mean and std across all dfs *for matplotlib

Plot components of std across all df if multiple IDs, but no df_name …

dd7b509

…specified *mean_std indicates whether multiple dfs have been passed, without specifying df_name *plot mean and std across all dfs *for matplotlib

Merge branch 'main' into comp_avg_global

3c5a82c

Plot components of std across all df if multiple IDs, but no df_name …

52366b5

…specified *mean_std indicates whether multiple dfs have been passed, without specifying df_name *plot mean and std across all dfs *for matplotlib

Merge branch 'main' into comp_avg_global

96a15c9

intermediate commit, base idea to continue the PR

04ebebb

[To be completed] Updating handling of df_name in plot_parameters

2865cbe

Plotting mean and std across all dfs in plot_parameters()

bd12ecf

* only if global df, and df_name not specified * only for trend and seasonality * Plotly

Plotting mean and std across all dfs in plot_parameters()

ae616cc

* only if global df, and df_name not specified * only for trend and seasonality * Matplotlib

Reverting the changes in plot_components() as the feature shall only …

ec9d5f6

…be available for plot_parameters()

Defined edge cases for different inputs for:

bd3b7f1

* df_name *global/local normalization *glocal/global model with multiple IDs

Merge branch 'main' into comp_avg_global

d617e35

Test new feature

7fd6fb8

Merge remote-tracking branch 'origin/comp_avg_global' into comp_avg_g…

14737c5

…lobal

Test new feature

ea9005b

Test new feature

08996c0

Test new feature

6b7c6c8

alfonsogarciadecorral self-requested a review October 27, 2022 07:47

Kevin-Chen0 added this to the Release 0.5.0 milestone Nov 4, 2022

judussoari added the status: ready PR is ready to be merged label Nov 7, 2022

judussoari and others added 5 commits November 7, 2022 18:30

Merge branch 'main' into comp_avg_global

2ddb618

Merge branch 'main' into comp_avg_global

283dd92

Merge branch 'main' into comp_avg_global

bd69fa9

Merge branch 'main' into comp_avg_global

3596310

Merge branch 'main' into comp_avg_global

3026ed6

Kevin-Chen0 added status: needs update PR has outstanding comment(s) or PR test(s) that need to be resolved and removed status: ready PR is ready to be merged labels Nov 22, 2022

Merge branch 'main' into comp_avg_global

84ff5ad

# Conflicts: # neuralprophet/forecaster.py # neuralprophet/plot_forecast_plotly.py # neuralprophet/plot_model_parameters_matplotlib.py # neuralprophet/plot_model_parameters_plotly.py

judussoari added status: needs review PR needs to be reviewed by Reviewer(s) and removed status: needs update PR has outstanding comment(s) or PR test(s) that need to be resolved labels Nov 22, 2022

Kevin-Chen0 added status: needs update PR has outstanding comment(s) or PR test(s) that need to be resolved and removed status: needs review PR needs to be reviewed by Reviewer(s) labels Nov 23, 2022

judussoari added 8 commits November 23, 2022 23:55

Merge branch 'main' into comp_avg_global

f301847

# Conflicts: # neuralprophet/forecaster.py # neuralprophet/plot_model_parameters_matplotlib.py # neuralprophet/plot_model_parameters_plotly.py # tests/test_integration.py # tests/test_plotting.py

Resolve Merge conflict

518a862

Resolve Merge conflict

f796a39

Ensure codecoverage

cf467fb

Ensure codecoverage

ad6232f

Ensure codecoverage

ef664ca

Ensure codecoverage

d80848f

Linters

19ec697

judussoari added status: needs review PR needs to be reviewed by Reviewer(s) and removed status: needs update PR has outstanding comment(s) or PR test(s) that need to be resolved labels Nov 27, 2022

Kevin-Chen0 approved these changes Nov 27, 2022

View reviewed changes

Kevin-Chen0 merged commit fe1d759 into ourownstory:main Nov 27, 2022

judussoari deleted the comp_avg_global branch November 27, 2022 22:16

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[feature] Plot components of averaged df (and quants) if multiple IDs, but no df_name specified in plot_components() #888

[feature] Plot components of averaged df (and quants) if multiple IDs, but no df_name specified in plot_components() #888

judussoari commented Oct 21, 2022 •

edited

Loading

codecov-commenter commented Oct 21, 2022 •

edited

Loading

alfonsogarciadecorral commented Oct 27, 2022

Kevin-Chen0 commented Nov 22, 2022

Kevin-Chen0 commented Nov 23, 2022

Kevin-Chen0 left a comment

github-actions bot commented Nov 27, 2022

Model Training

PeytonManning

YosemiteTemps

AirPassengers

[feature] Plot components of averaged df (and quants) if multiple IDs, but no df_name specified in plot_components() #888

[feature] Plot components of averaged df (and quants) if multiple IDs, but no df_name specified in plot_components() #888

Conversation

judussoari commented Oct 21, 2022 • edited Loading

codecov-commenter commented Oct 21, 2022 • edited Loading

Codecov Report

alfonsogarciadecorral commented Oct 27, 2022

Kevin-Chen0 commented Nov 22, 2022

Kevin-Chen0 commented Nov 23, 2022

Kevin-Chen0 left a comment

Choose a reason for hiding this comment

github-actions bot commented Nov 27, 2022

Model Benchmark

Model Training

PeytonManning

YosemiteTemps

AirPassengers

judussoari commented Oct 21, 2022 •

edited

Loading

codecov-commenter commented Oct 21, 2022 •

edited

Loading