Difference between revisions of "User:Jaired"

From REU@MU
Jump to: navigation, search
(Week 6 (2 Jul - 6 Jul))
(Week 10 (30 Jul - 3 Aug))
 
(46 intermediate revisions by the same user not shown)
Line 155: Line 155:
  
 
== Tuesday ==
 
== Tuesday ==
* Work on long short-term memory recurrent neural network
+
* Implemented long short-term memory recurrent neural network, peaks too often
 +
* Tried several configurations of LS-SVM NAR
 +
* Created "artificial" data with extra troughs in it to improve accuracy of troughed data with LS-SVM
 +
* Literature search on time series/forecast*/machine learning
 +
 
 +
== Wednesday ==
 +
* Created "artificial" data with extra peaks in it to improve accuracy of peaked data with LS-SVM
 +
* Manual ensemble of peaked, trough, and normal LS-SVM models for forecasting
 +
* Created several figures to show LS-SVM for regression
 +
 
 +
== Thursday ==
 +
* Remote desktop into a better computer to use more training data
 +
* Started working on a programmatic rule-based ensemble of differently trained models
 +
* REU working lunch with Dr. Corliss
 
<br>
 
<br>
  
 
= '''Week 7 (9 Jul - 13 Jul)''' =
 
= '''Week 7 (9 Jul - 13 Jul)''' =
Code and test algorithm
+
 
 +
== Monday ==
 +
* Meeting with Dr. Povinelli to discuss project and his and graduate student's visit to the customer company
 +
* Created different figure of LS-SVM to only include the 30 minute forecast value every minute
 +
* Another meeting with Dr. Povinelli to go over his and the graduate student's trip to the customer company
 +
 
 +
== Tuesday ==
 +
* Meeting with Dr. Povinelli to go over revised material for his trip
 +
* Meeting with Dr. Corliss to get his opinion on prepared items
 +
* Worked on background of research paper
 +
* Created figures to explain SVM and SVR
 +
 
 +
== Wednesday ==
 +
* Studied SVMs and SVRs heavily to be able to explain them in the research paper
 +
* Worked on the background for research paper
 +
* Tweaked SVM and SVR figures
 +
* Worked on Methods section for research paper
 +
 
 +
== Thursday ==
 +
* Work on research paper
 +
* Creation of naive model and its RMSE and MAE
 +
* Direct forecast with several different time horizons
 +
* GasDay seminar
 +
* REU Working Lunch
 +
 
 +
== Friday ==
 +
* Worked on research paper
 +
* Went over very rough draft of paper with Dr. Corliss
 +
* Ensembled new direct forecast models
 +
* Direct forecast testing and tuning
 +
* Reorganized git repository
 +
 
 
<br>
 
<br>
  
 
= '''Week 8 (16 Jul - 20 Jul)''' =
 
= '''Week 8 (16 Jul - 20 Jul)''' =
More experiments with algorithm
+
== Monday ==
 +
* Scrutinized ensemble of direct forecasts script for bugs
 +
* Looked for a target publication venue
 +
* Worked on method section in research paper
 +
 
 +
== Tuesday ==
 +
* More scrutinizing of code, determined inaccuracies are caused by a bias in the algorithm
 +
* "Cleaned" some anomalous spikes in data
 +
* Created a training function
 +
* Tried to make the poster a little prettier
 +
* Generated some figures for poster
 +
* Meeting with Dr. Povinelli
 +
 
 +
== Wednesday ==
 +
* Tested out adding a new feature to see if prediction is a little more accurate
 +
* Created MATLAB objects to store models and model parameters
 +
* Created an alarm forecasting script
 +
* Poster work
 +
 
 +
== Thursday ==
 +
* Poster work
 +
* GasDay Seminar
 +
* REU Working Lunch
 +
* Project meeting
 +
 
 +
== Friday ==
 +
* Finishing touches on poster
 +
* Created sensitivity figures for multiple time horizons
 +
* Left a script running for the weekend to produce figures
 +
 
 
<br>
 
<br>
  
 
= '''Week 9 (23 Jul - 27 Jul)''' =
 
= '''Week 9 (23 Jul - 27 Jul)''' =
Work on final draft of research paper
+
 
<br>Work on poster
+
== Monday ==
 +
* Operator error on time horizon stop, 3 instead of 30
 +
* Ran script on Dr. Corliss's machine
 +
* Converted paper to Elsevier format
 +
* Worked on paper
 +
* Worked on GasDay presentation for seminar
 +
 
 +
== Tuesday ==
 +
* Worked on paper
 +
* Project meeting
 +
* Worked on GasDay presentation for seminar
 +
* Ran co-researcher through code since I'm forsaking him soon
 +
 
 +
== Wednesday ==
 +
* Paper work
 +
* Worked on GasDay presentation for seminar
 +
 
 +
== Thursday ==
 +
* Finishing touches on GasDay presentation
 +
* Presentation at GasDay
 +
* GasDay seminar
 +
* Weekly REU lunch meeting
 +
* Project meeting
 +
* Worked on paper
 +
 
 +
== Friday ==
 +
* Paper work
 +
 
 
<br>
 
<br>
  
 
= '''Week 10 (30 Jul - 3 Aug)''' =
 
= '''Week 10 (30 Jul - 3 Aug)''' =
Prepare final presentation
+
 
 +
== Monday ==
 +
* Poster forum at AMU
 +
* Talked to random people about my poster
 +
* Worked on paper
 +
 
 +
== Tuesday ==
 +
* REU presentation preparation
 +
* Worked on paper
 +
* Went through code with co-researcher
 +
 
 +
== Wednesday ==
 +
* REU presentations and lunch
 +
* Final project meeting
 +
* Worked on paper more
 +
* Ran a final experiment
 +
 
 +
== Thursday ==
 +
* Worked on paper
 +
* Wrote readmes
 +
* Created a directory with my work for GasDay
 +
 
 +
== Friday ==
 +
* Take me home
 +
* country roads
 +
* to the place
 +
* I belong

Latest revision as of 14:41, 2 August 2018

I am Jaired Collins, a student at Missouri Southern State University in Joplin, MO, currently majoring in computational mathematics. Outside of academia, I enjoy cycling, weights, table tennis, tennis, and cooking.

Week 1 (29 May - 1 Jun)

Tuesday

  • Initial meeting to get introduced into the program.
  • Lunch meeting with mentors.
  • Obtained MU and MSCS account logins and ID card
  • Paperwork

Wednesday

  • Met with Dr. Corliss to get acquainted with the GasDay lab and receive a key.
  • Met with Dr. Povinelli and another student to set up initial parameters of the project.
  • Literature review.

Thursday

  • More literature review
  • Seminar with GasDay
  • Meeting with Dr. Povinelli and student to establish a weekly plan and talk about the project.

Friday

  • Literature review
  • Planned for a presentation at a GasDay Seminar on the 21st of June
  • Data cleaning


Week 2 (4 Jun - 8 Jun)

Monday

  • Meeting with Dr. Corliss
  • Meeting with Dr. Povinelli and Colin
  • Literature search
  • Cleaned up more data

Tuesday

  • Meeting with Dr. Corliss
  • Data cleaning with bash and Matlab
  • Played around with data and SVMs in MatLab
  • Meeting with research consultant
  • Literature review

Wednesday

  • Cross validation with k-Fold
  • GasDay Camp
  • Literature search

Thursday

  • More built-in SVM work in MatLab
  • GasDay camp
  • REU working lunch meeting, bad presentation

Friday

  • Meeting with Dr. Corliss
  • Creation of ShareLaTeX project
  • Converted temporal pressure data into nonuniform time series data
  • Tried resampling time series data into uniform data


Week 3 (11 Jun - 15 Jun)

Monday

  • Meeting with Dr. Corliss
  • Meeting with Dr. Povinelli
  • Resampled nonuniform data
  • Forecasting with ARIMA
  • Left computer running to finish ARIMA models

Tuesday

  • Examined and saved ARIMA models
  • Dissimilarity-based approach to prediction, where correlation coefficients between sensors are used as features
  • SVM and k-fold cross validation at k = 5, 10, and 20

Wednesday

  • Meeting with Dr. Corliss to elucidate LS-SVM for regression
  • Coding LS-SVM
  • RCR lunch meeting

Thursday

  • Meeting with Dr. Corliss, further explained LS-SVM
  • GasDay seminar
  • Coding of LS-SVM for regression
  • Meeting with Dr. Povinelli
  • Called contact to clarify threshold values

Friday

  • Out for wedding


Week 4 (18 Jun - 22 Jun)

Monday

  • Out for wedding

Tuesday

  • Wrote working abstract
  • Wrote summaries for related works section
  • Read Anomaly Detection in Streaming Non-stationary Temporal Data
  • Read Chapter 5 of Operations & Maintenance Best Practices Release 3.0
  • Tested different features in LS-SVM; residuals are very low

Wednesday

  • Meeting with Dr. Corliss, showed LS-SVM results
  • Work on conference paper with graduate student
  • LS-SVM Tuning
  • Work on PPT for presentation

Thursday

  • Practice presentation at GasDay Seminar
  • Working REU lunch
  • Knowledge and Information Discovery with Dr. Povinelli
  • Project meeting with Dr. Povinelli and graduate student

Friday

  • Recursive nonlinear autoregressive model with LS-SVM
  • Clarification on building different models with different time horizons instead of recursive approach
  • Coding nonlinear autoregressive model with LS-SVM with multiple time horizons
  • Work on abstract
  • Work on background


Week 5 (25 Jun - 29 Jun)

Monday

  • Worked on slides for presentation on Thursday
  • Tried to debug direct multiple time horizon least squares support vector machine nonlinear autoregressive model (LS-SVM NAR)
  • Showed abstract to Dr. Povinelli, got feedback
  • Resampled data to 1 minute and wrote several MATLAB scripts to automate the LS-SVM NAR process
  • Came to the conclusion that LS-SVM NAR is not predicting well at all. Plotted values were erroneous and coincidentally showed promising results

Tuesday

  • Meeting with Dr. Corliss
  • Read more into SVMs and SVRs to help the LS-SVM NAR algorithm
  • Implementation of radial basis function kernel for LS-SVM NAR

Wednesday

  • Debugging LS-LSVM NAR with RBF kernel with test cases
  • Tuning LS-LSVM NAR hyperparameters
  • Presentation preparation

Thursday

  • More preparation for REU Midterm
  • GasDay Seminar
  • Midterm presentations
  • KID Seminar
  • Meeting with Dr. Povinelli to discuss project

Friday

  • Put a lot of effort into making LS-SVM work
  • Meeting with Dr. Povinelli to help with LS-SVM and things to do while he is on vacation


Week 6 (2 Jul - 6 Jul)

Monday

  • Confirmed some math with Dr. Corliss
  • Messing around with LS-SVM to get it to work
  • Created normalized time series
  • Extracted correlated information from multiple sensors

Tuesday

  • Implemented long short-term memory recurrent neural network, peaks too often
  • Tried several configurations of LS-SVM NAR
  • Created "artificial" data with extra troughs in it to improve accuracy of troughed data with LS-SVM
  • Literature search on time series/forecast*/machine learning

Wednesday

  • Created "artificial" data with extra peaks in it to improve accuracy of peaked data with LS-SVM
  • Manual ensemble of peaked, trough, and normal LS-SVM models for forecasting
  • Created several figures to show LS-SVM for regression

Thursday

  • Remote desktop into a better computer to use more training data
  • Started working on a programmatic rule-based ensemble of differently trained models
  • REU working lunch with Dr. Corliss


Week 7 (9 Jul - 13 Jul)

Monday

  • Meeting with Dr. Povinelli to discuss project and his and graduate student's visit to the customer company
  • Created different figure of LS-SVM to only include the 30 minute forecast value every minute
  • Another meeting with Dr. Povinelli to go over his and the graduate student's trip to the customer company

Tuesday

  • Meeting with Dr. Povinelli to go over revised material for his trip
  • Meeting with Dr. Corliss to get his opinion on prepared items
  • Worked on background of research paper
  • Created figures to explain SVM and SVR

Wednesday

  • Studied SVMs and SVRs heavily to be able to explain them in the research paper
  • Worked on the background for research paper
  • Tweaked SVM and SVR figures
  • Worked on Methods section for research paper

Thursday

  • Work on research paper
  • Creation of naive model and its RMSE and MAE
  • Direct forecast with several different time horizons
  • GasDay seminar
  • REU Working Lunch

Friday

  • Worked on research paper
  • Went over very rough draft of paper with Dr. Corliss
  • Ensembled new direct forecast models
  • Direct forecast testing and tuning
  • Reorganized git repository


Week 8 (16 Jul - 20 Jul)

Monday

  • Scrutinized ensemble of direct forecasts script for bugs
  • Looked for a target publication venue
  • Worked on method section in research paper

Tuesday

  • More scrutinizing of code, determined inaccuracies are caused by a bias in the algorithm
  • "Cleaned" some anomalous spikes in data
  • Created a training function
  • Tried to make the poster a little prettier
  • Generated some figures for poster
  • Meeting with Dr. Povinelli

Wednesday

  • Tested out adding a new feature to see if prediction is a little more accurate
  • Created MATLAB objects to store models and model parameters
  • Created an alarm forecasting script
  • Poster work

Thursday

  • Poster work
  • GasDay Seminar
  • REU Working Lunch
  • Project meeting

Friday

  • Finishing touches on poster
  • Created sensitivity figures for multiple time horizons
  • Left a script running for the weekend to produce figures


Week 9 (23 Jul - 27 Jul)

Monday

  • Operator error on time horizon stop, 3 instead of 30
  • Ran script on Dr. Corliss's machine
  • Converted paper to Elsevier format
  • Worked on paper
  • Worked on GasDay presentation for seminar

Tuesday

  • Worked on paper
  • Project meeting
  • Worked on GasDay presentation for seminar
  • Ran co-researcher through code since I'm forsaking him soon

Wednesday

  • Paper work
  • Worked on GasDay presentation for seminar

Thursday

  • Finishing touches on GasDay presentation
  • Presentation at GasDay
  • GasDay seminar
  • Weekly REU lunch meeting
  • Project meeting
  • Worked on paper

Friday

  • Paper work


Week 10 (30 Jul - 3 Aug)

Monday

  • Poster forum at AMU
  • Talked to random people about my poster
  • Worked on paper

Tuesday

  • REU presentation preparation
  • Worked on paper
  • Went through code with co-researcher

Wednesday

  • REU presentations and lunch
  • Final project meeting
  • Worked on paper more
  • Ran a final experiment

Thursday

  • Worked on paper
  • Wrote readmes
  • Created a directory with my work for GasDay

Friday

  • Take me home
  • country roads
  • to the place
  • I belong