Aarit-Gupta

[aaritg]

Week 1: First week of semester. Add/drop so nothing happened. Week 2: First real meeting of the semester. Dr. Kastner gave a presentation of what this VIP team will be doing and how it will operate by explaining the different possible sub teams. Later we filled out a form on what subteams we're interested in. Week 3: Subteams and their members are shown as well as a more in dept explanation of what each is.

Weekly Notebook Entry — Week 4

1/29 - 2/4: Presented our intro slides about what our topic is about. I did some of the literature review (Westermann and Evins paper).

Tasks for this week:

• Task 1 (Completed): Read the papers recommended by Dr. Kastner from the subteam meeting launch (Week 3). This topic is already well researched, so we just have to first get a good understanding of the problem at hand. Outcome: I have a good understanding of what this project is about and we've taken notes along the way.
• Task 2 (Ongoing): Find other papers that align with what we're doing. Outcome: Found the NREL paper linked below.

Challenges:

• Research papers are dense, so it can be challenging for unexperienced individuals to find the important bits. This was definitely the case for the first paper because I don't know much about this topic.

Personal Reflection:

• I feel that I have a good general understanding of the whole topic and generally what's expected of us going forward.

Team Meetings:

• During the week, met with Dr. Kastner and discussed another paper from Illinois that's doing something similar and even has a GitHub folder that he liked.

Plans for next week:

• Continue looking at that new paper I found and potentially find some more. I will get my teammates to look at its GitHub and see what can be done with it.

[aaritg]

Weekly Notebook Entry — Week 5

Overview

• Week Span: [2/5] to [2/11]

Tasks for This Week

• [ ] Task 1: Now that Dr. Kastner has looked over and approved of the NREL paper, I began to dig deeper into its specifics. This paper is very similar to what we are looking for (modeling hour data on residential buildings) and it even has a GitHub that he thinks looks good.

Tasks

Task: Find relevant papers - (completed)

• Because we know this work is already well researched, we want to find more papers that can help guide us:
• Link to Work: [Link](https://docs.google.com/document/d/1tao5_4oLTmIoPKCjA62Nh1SMRc3dTThy5MaDDVmbHwM/edit)
• Outcome & Reflection: As Dr. Kastner explained, this paper is quite good because it was produced by NREL, a large player in this space. We know that this information is vetted an reliable, but we're still figuring out what parts we need and how to make good use of this information.

Task: [Continue looking this paper/GitHub] - (ongoing)

• Hope to have this done by next week.

Challenges & Solutions

• Again, this document is quite long so it'll take some time to fully understand what's going on and how to act on it.

Team Meetings

• Meeting Highlights: Met with Dr. Kastner for our weekly office hours and explained that we're working on getting through all the details.

Personal Reflection

• Each meeting, our vision for this project get clearer and generally moves in the right direction. With this research, along with that of my teammates, I think we're on the right track to get something done.

Plans for Next Week

• [x] Next Week's Task 1: Try to get the GitHub running because it has a step-by-step documentation.

[aaritg]

Weekly Notebook Entry — Week [6]

Overview

• Week Span: [2/12] to [2/18]

Tasks for This Week

• [ ] Task 1: Try to get the NREL GitHub running. As mentioned earlier, this work is similar to ours and because it has documentation, we may be able to get it going.

Tasks

Task: [Run GitHub] - (completed)

• Description: I didn't quite understand it all, so I asked one my teammates who eventually got it running (thanks Anubha). Unfortunately it doesn't work because their AWS integration is down though we don't have access to it either way.
• Outcome & Reflection: Essentially we won't be able to use this, but other teammates are looking into other existing models that we can use.

Challenges & Solutions

• While the research paper was good, the GitHub unfortunately is of little use to us.

General Takeaway, Skills & Knowledge Acquired

• We'll have to continue looking at other resources, but at least this showed us part of the work that's out there.

Team Meetings

• Meeting Highlights: In class, we explained the situation to Dr. Kastner who was very understanding of the issue at hand. We explained what we're going to do instead (look at other resources).
• Peer Feedback:

Plans for Next Week

• [ ] Next Week's Task 1: Starting working on a poster to present later at the VIP competition.

[aaritg]

Weekly Notebook Entry — Week [7]

Overview

• Week Span: [2/19] to [2/25]

Tasks for This Week

• [ ] Task 1: Try to get the other models running.

Tasks

Task: [ Help get another model running ] - (In progress)

• Description: Because the other model wasn't working, we'll need to use a different one. Currently, we're trying out the
• Link to Work:
• Outcome & Reflection: The Optimization of Building Energy Consumption model reveals some data, as shown in the PowerPoint.

Challenges & Solutions

• The challenge is figuring out how to get all of these models to run. Since they are quite large, it takes some understanding of what they're trying to achieve and what tools and all are needed to get it going.

General Takeaway, Skills & Knowledge Acquired

• Currently we're trying to make sense of these models and their data and how it works.

Team Meetings

• Meeting Highlights: Met with Dr. Kastner and he discussed the source of data input, such as the Fulton County Assessor's Office website that has property records.
• Peer Feedback:

Personal Reflection

• We just have to keep trying the different models that we can find. We're making progress each week and remain hopeful we'll get something meaningful done by the end.

Visual Documentation (optional)

Additional Notes & Resources (optional)

• Resource Name - Brief description.

---

Plans for Next Week

• [ ] Next Week's Task 1: Get models running and get data.

[aaritg]

Weekly Notebook Entry — Week [8]

Overview

• Week Span: [2/26] to [3/3]

Tasks for This Week

• [ ] Task 1: Find the actual sources of data that were used by demand.ninja. This a very good source, and Abi is making progress on getting it to run. Once we run it though, we'll still need real energy data.
• [ ] Task 2: If that doesn't work, we'll have to seek out other data sources. This could potentially mean finding new models or contacting the government.

Tasks

Task: [Find demand.ninja datasets] - (completed)

• Description: demand.ninja, the creator of renewables.ninja, reveals that their sources of data on their website, which was very important to getting their tool to work. It'd be very helpful if we could get that data.
• Link to Work: renewables.ninja Sources
• Outcome & Reflection: I dug through those sources, but with little luck. I wasn't able to find real datasets, but there was some information about processes and math. Those were of little use because they didn't really give us any data, just the complex statistics of it.

Task: [Find other sources of data] - (ongoing)

• Description: There's the tax assessor's website, which reveals size and heating sources, but not any energy usage data that we're looking for. Dr. Kastner has posted other resources and my team member's have recommended looking into them.
• Expected Completion: End of next week

Challenges & Solutions

• Again, finding data is a challenge because most of it's kept private. Other sources are often convoluted and getting actual data (vs. ideas or processes) is hard to come by.

General Takeaway, Skills & Knowledge Acquired

• For this part of the project, finding good data is hard, but we believe it can be done.

Team Meetings

• Meeting Highlights: Didn't meet with Dr. Kastner but met amongst ourselves to get the models working.
• Peer Feedback: Progress is slowly but surely happening on the project.

Personal Reflection

• Right now, we're at a point where we're making progress on getting them to run, but we still don't know where the actual data came from. We're still going at it though, with good process being made.

Visual Documentation (optional)

Our weekly slides

Additional Notes & Resources (optional)

• Resource Name - Brief description.

---

Plans for Next Week

• [ ] Next Week's Task 1: Look into the new data sources.

[aaritg]

Weekly Notebook Entry — Week [9]

Overview

• Week Span: [3/4] to [3/10]

Tasks for This Week

• [ ] Task 1: Look into BAIT article shared by Dr. Kastner

Tasks

Task: [BAIT Article] - (completed)

• Description: This article talks directly about the demand.ninja model that we are trying to understand and run. Building Adjusted Internal Temperature (BAIT) is a way to account for the external climate's effect on a building. For example, sunny weather will reduce the need for heating and thus lower energy usage.
• Link to Work:
• Outcome & Reflection: The last paragraph of the article summarizes the issue at hand. For demand.ninja to actually be accurate, large datasets on retrofitted buildings are needed. This data is generally kept private, so it's questionable how accurate demand.ninja really is.

Challenges & Solutions

• Still trying to get everything running by understanding how it actually works.

General Takeaway, Skills & Knowledge Acquired

• Even with the above takeaways, we'll keep trying to run the models. Whether our data is even accurate is a toss-up but we're still pushing forward.

Team Meetings

• Meeting Highlights: Met up and discussed how to get the other models running
• Peer Feedback:

Personal Reflection

• While I haven't done much of the real coding part (not a CS major), I may have to start reading through the documentation and code comments to start understanding where our issues lay.

Visual Documentation (optional)

• Brief Weekly Presentation

Additional Notes & Resources (optional)

Plans for Next Week

• [ ] Next Week's Task 1: Look into Kaggle datasets to understand demand.ninja model.

[aaritg]

Weekly Notebook Entry — Week [10]

Overview

• Week Span: [3/11] to [3/17]

Tasks for This Week

• [ ] Task 1: Find the Kaggle dataset to run the demand.ninja model.

Tasks

Task: [Find data for the demand.ninja model] - (completed)

• Description: While Abi is working on getting the model running, I'm trying to find the data that they used to run it. I found a few sources on the Building Data Genome 2 Github page
• Link to Work: kaggle.csv
• Outcome & Reflection: It's a large dataset, but there's some potentially good info in there. The meaning of some of the numbers can be found on the creator's Kaggle page.

Challenges & Solutions

• We're going to have to understand what each value in the dataset means. There's some confusion over what building_id means for the Rank 1-5 work model being run by Shivam.

General Takeaway, Skills & Knowledge Acquired

• It's good that we got some data, but we can't do much with it until our models run. It does clear some confusion though.

Team Meetings

• Meeting Highlights: Had a short meeting with Dr. Kastner to discuss what was going on.
• Peer Feedback:

Personal Reflection

• Still moving a fair pace. Progress is being made in the other subgroups, so things are looking fine.

Visual Documentation (optional)

Additional Notes & Resources (optional)

• Resource Name - Brief description.

---

Plans for Next Week

• [ ] Next Week's Task 1: Spring Break, so not going to do any work

[aaritg]

Weekly Notebook Entry — Week [12]

(Week 11 was Spring Break, so no entry)

Overview

• Week Span: [3/25] to [3/31]

Tasks for This Week

• [ ] Task 1: Keep working on fixing issues with demand.ninja
• [ ] Task 2: Look into Fulton County Assessor's website to potentially get building data

Tasks

Task: [Work on running demand.ninja] - (in progress)

• Description: Abi and I are tasked with working on demand.ninja, but she's running into a few errors, despite the code running. I'm looking into why that issue is occurring by reading the code.
• Link to Work: Link
• Current Status: We still haven't fixed the issue simply because of the size of the model. Not sure when it will be complete either as this a busy time of the semester for some of us.

Task: [Look at Fulton Assessor's website] - (completed)

• Description: Essentially a government database that contains data on all buildings/land in Fulton County. The second link allows you to find databases from certain other parts of the country.
• Link: Fulton County and other locations

Challenges & Solutions

• Not sure where our code is erroring, but we're reading the comments and documentation to see what every line does.

General Takeaway, Skills & Knowledge Acquired

• Reading code comments and potentially the example will be helpful.

Team Meetings

• Meeting Highlights: Met with Dr. Kastner and revealed our general progress. He recommend reading an example use of demand.ninja, if provided.
• Peer Feedback: Other team members have their own struggles on the other models.

Personal Reflection

• I think we're still on track to get something running out of this model.

Visual Documentation (optional)

Plans for Next Week

• [ ] Next Week's Task 1: Keep working on demand.ninja

[aaritg]

Weekly Notebook Entry — Week [13]

Overview

• Week Span: [4/1] to [4/7]

Tasks for This Week

• [ ] Task 1: Keep working on demand.ninja model

Tasks

Task: [Understand demand.ninja] - (in progress)

• Description: We've have made more progress in understanding the ins and outs of the demand.ninja model, however, there are still issues with the BAIT values not calculating correctly.
• Outcome & Reflection: Anubha has agreed to help us out and take a look at the problems.

Challenges & Solutions

• The model is a fair size and we don't entirely understand how some parts work. ChatGPT has been a help, though the code already runs without error, so it's a little unclear why values don't compute.

General Takeaway, Skills & Knowledge Acquired

Team Meetings

• Meeting Highlights: Met with Dr. Kastner, who advised us on our progress and met amongst ourselves.
• Peer Feedback: We'll start consolidating our work and may focus more attention towards the Rank 5 model because it's more promising.

Personal Reflection

• We've made good progress on this model, and we will continue to work on it, but it may be time to set aside for a bit.

Visual Documentation (optional)

Additional Notes & Resources (optional)

• Resource Name - Brief description.

---

Plans for Next Week

• [ ] Next Week's Task 1: Make a plan of what need to happen in the next few weeks. Work on the demand.ninja model will likely continue, though at a slower pace, as we need to prepare for the end of the semester.

[aaritg]

Weekly Notebook Entry — Week [14]

Overview

• Week Span: [4/8] to [4/14]

Tasks for This Week

List your specific objectives for this week, including learning targets and project milestones.

• [ ] Task 1: Figure out plan for the end of semester.

Tasks

Task: [Plan for next week(s)] - (completed)

• Description: Anubha has prepared a web scraper that allows us to collect building data in the City of Atlanta from the Fulton County Assessor's Office. Shivam and Anubha will work a bit more on Rank 5 while Abi and I try a bit more on demand.ninja.
• Link to Work: Link
• Outcome & Reflection: Semester is quickly coming to an end and while we're trying hard to get Rank 5 to entirely work, as it's the most promising right now, we need demand.ninja as a backup.

Task: [Help on demand.ninja] - (ongoing)

• Current Status: Abi is making some progress on getting demand.ninja running. Anubha will try to help us later.
• Expected Completion: Next week

Challenges & Solutions

• The same old, understanding why some numbers come from what source.

General Takeaway, Skills & Knowledge Acquired

• Even if this model isn't fully functionable, good progress has been made overall and will continue.

Team Meetings

• Meeting Highlights: Met amongst ourselves and with Dr. Kastner. We're close to finishing our project, so we're all working hard to get the final piece of work out.
• Peer Feedback: We're all still committed to getting something for the rest of the class to see.

Personal Reflection

• I think we can successfully get the task done.

Visual Documentation (optional)

Additional Notes & Resources (optional)

• Resource Name - Brief description.

---

Plans for Next Week

• [ ] Next Week's Task 1: Work on presentation

[aaritg]

Weekly Notebook Entry — Week [15]

Overview

• Week Span: [4/15] to [4/22]

Tasks for This Week

• [ ] Task 1: Create final presentation

Tasks

Task: [Create Final Presentation] - (completed)

• Description: The last day of class is Monday 4/22, so while the others work on finishing the models, I'm creating the presentation.
• Link to Work: Final Presentation
• Outcome & Reflection: Luckily Anubha was able to figure out why demand.ninja wasn't giving BAIT values and plotted it on the geoJSON map

Challenges & Solutions

• Getting that demand.ninja model was a fair amount of work but luckily it worked out in the end.

General Takeaway, Skills & Knowledge Acquired

• Overall, this was a good project and I learned a lot about a topic I didn't previously know much about.

Team Meetings

• Meeting Highlights: Presented our final project to the class.
• Peer Feedback: Might do some more work to polish up some things, but we're effectively done with the project.

Personal Reflection

• While it wasn't easy, it was an effective semester and we created a good project. I enjoyed learning about the other topics from the other groups, and of course, a big thanks to Dr. Kastner for the class.

Visual Documentation (optional)

Final presentation and map are linked above.

Additional Notes & Resources (optional)

• All models and documentation are on GitHub.