Journal Reflection #5: Insights into your FINAL project

Image from: https://imageio.forbes.com/specials-images/imageserve/5f151e9faa78e50007ce7c76/
The-10-Best-AI-And-Data-Science-Master-s-Courses-For-2021/960×0.jpg

Please use your reply to this blog post to detail the following:

  1. Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?
  2. What was your desired learning outcome of your choice of final project?
  3. What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
  4. Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
  5. Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
  6. Include your Github repo URL so your classmates can look at your code.
  7. Of all of the projects you or your classmates created this semester, which one was your favorite, and why?

Take the time to look through the project posts of your classmates. If you saw any project or project descriptions that pique your interest, please reply or respond to their post with feedback. Constructive criticism is allowed, but please keep your comments civil.

THANKS FOR TAKING THE COURSE!!!

This entry was posted in Uncategorized. Bookmark the permalink.

12 Responses to Journal Reflection #5: Insights into your FINAL project

  1. Ori Moore says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?
    I really enjoyed my A* project and wanted to take it to the next level. I ended up having to completely start from scratch to make this project though. It was helpful to already understand the concepts of an A* algorithm but this was a much more complex implementation of it, that included features like diagonal lines, higher resolution, and Polygon objects. It was fun to make a more practical version of my pathfinder with more robust PyGame interfaces for both creating a map and pathfinding on it.

    What was your desired learning outcome of your choice of final project?
    I wanted to tackle advancements to my A* pathfinder and understand how it could be used to model real-life environments instead of a more coarse grid-like representation of real-life. I think my implementation of the Polygon class and my transformation of user-created lines to representations of real-life objects successfully executed that goal for three independent modes that each followed different rules: regular, accessible, and trailblazer. The biggest challenge was tackling “neighbor” nodes that aren’t directly connected by lines, such as walking across a quad or a grassy area.

    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
    The most useful aspect of this class has been being able to combine working alone and working collaboratively to experience coding more like the real world—being able to build our own class structures and working with a combination of structures already in place and structures we can create. I also got a lot out of learning useful Machine Learning techniques, like my introduction to neural networks and pathfinders. I also had never done anything involving UI before and while PyGame is not necessarily the best UI out there, it was helpful to be able to get into visual representations of the algorithms I employed.

    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
    Yes, I thought I got a lot out of introductory projects to a lot of different types of machine learning algorithms. Getting to do a combination of statistical modelling, neural networks, self-play, and pathfinding allowed me to get a start on a variety of algorithms that I can hopefully go deeper into in the future.

    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    I’m not sure exactly what my major will be but it will likely be statistics or something similar (depending on the school). So the short answer is yes. And I do think my background in machine learning and modelling will be helpful in understanding and applying concepts from other disciplines.

    Include your Github repo URL so your classmates can look at your code.
    https://github.com/ori-moore/Campus_Pathfinder.git

    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?
    I really enjoyed the Hnefatafl bot because it was fun to have a joint project with Ryan and get to learn a new concept in a friendly collaboration + rivalry. I do think it would be fun if one project had been required to be a pair project, just to make sure we got a chance to work together.

  2. Ryan Bauroth says:

    [1, 2] My project is a neural network (NN) built from scratch, using no libraries (ex: tensorflow, pytorch, etc). I have successfully trained two models with this NN: data defined by the function y > sin(6x) and on handwritten digit data. I decided on this project because I really wanted to understand the small details that actually make a NN work. I have learned about activation functions, gradient descent, and backpropagation through this project.

    [3] The most useful thing about this class has been the time I have been able to dedicate to semi-structured personal projects. Throughout most school years, any CS projects I do are the second priority behind schoolwork (for good reason), which means I rarely end up doing projects. Also, without the structure of a class, I struggle to actually complete projects. I often run into a bug that kills my interest due to the amount of work required amidst a busy school schedule. Having this class has allowed me to actually pursue these projects and let me make some pretty cool things (I think).

    [4] My work this semester most definitely has helped me understand ML and AI much better. Going into this class, AI was like a black box to me: I had no idea how it worked. Now, I can understand exactly what sort of data would result in a good AI/ML project, how to tune activation functions, learning rates, etc to improve training time and results, and the limitations of such results. I no longer look at AI/ML as a sort of magic.

    [5] I imagine I will pursue similar CS coursework in college as I have really enjoyed AI/ML and I imagine this class will set me up well for that endeavor.

    [6] https://github.com/Ryan-Bauroth/NeuralNetwork

    [7] My favorite amongst my projects has been this most recent one. I have really delved into the details in a way I really enjoyed. While losing to Ori’s Hnefatafl bot was devastating, that friendly rivalry was also a major highlight of this semester. Out of the class’ really cool projects, I think Dani’s flappy bird AI was amazing. A lot of projects (mine especially) fall into this place of ‘they kind of work,’ which is probably to be expected given how complicated AI/ML is. Therefore, it really was shocking when it didn’t take all that many training iterations for her model to be unbeatable at the game.

  3. Matthew Guo says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?

    My project is a proof-of-concept for a machine learning model that can run on embedded or very low-end processors. This machine learning model is meant to take a 3-lead ECG data feed and try to diagnose the condition by classification. To be honest, I choose this project because I already made a set of ECG hardware for the physics class final project, so this project would be complementing the hardware, leaving only firmware to do.

    What was your desired learning outcome of your choice of final project?
    My desired learning outcome was to more deeply understand how heart conditions are diagnosed from ECG readings and how the heart works from a neurological standpoint.
    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
    The most useful aspect of this class has been learning about how easy ML is with python.

    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
    I think my work this semester helped me better understand the foundations of ML and AI because I got to implement many different cases of ML in many different hardware requirement environments.

    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    I don’t see myself pursuing this type of coursework but I see concepts I learned in this class as very useful in aiding work, for example, using chatgot to do homework.

    Include your Github repo URL so your classmates can look at your code.
    https://github.com/a-me-lia/ecgproject
    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?
    My favorite project this semester was the flappy bird project because it was very complex but was veru successful.

  4. Aarav Prakash says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?
    I really enjoyed the second project, as I had never been experienced with Reinforcement learning before. I feel that it was the most complex project and I didn’t understand a lot of it, so I decided to do more reinforcement learning. My final project is an AI that can play a 2 dimensional swerve drive simulator game that is FRC based. There is an offensive(blue) robot that tries to get past a defensive(red) robot. I tried to use reinforcement learning and self play to write a defensive and offensive AI that plays this game.
    What was your desired learning outcome of your choice of final project?
    I wanted to learn more about utilizing reinforcement learning for real life scenarios. Of course, a pygame game is not real life, however, it is a simulation of a real life situation. Making a reinforcement learning algorithm that could drive the robot in FRC would represent a much more advanced system of play, making teleoperated play much more interesting. It also would help gain strategies for human drivers. Say a robot in this simulation uses a strategy that human drivers haven’t thought of, a human driver can utilize that strategy.
    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
    In this class, I feel that learning AI concepts and learning how to learn AI concepts was incredibly important. The most useful aspect was learning about learning, because learning how to research and implement certain algorithms with the current tools at our disposal such as chatgpt or the internet is perhaps the most important part of implementing any AI algorithm.
    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
    I believe that it has. I never understood reinforcement learning or the A* pathfinding algorithm before this class, and while a lot of it was quite complex for my math base, I still learned about basic algorithm implementation.
    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    I do, and I do anticipate being ahead because I will know how to learn to implement algorithms without having a professor teach the concepts to me.
    Include your Github repo URL so your classmates can look at your code.
    https://github.com/A0Prakash/Project_04_SDS.git
    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?
    I liked the second and the fourth, because I really enjoyed reinforcement learning. I believe that out of all the projects and concepts that we learned about, reinforcement learning was the most powerful and has the most future potential, which is why I liked learning about it.

  5. Shreya Rao says:

    1) Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?

    My final project is a Book Review Analysis System that predicts whether I would enjoy reading a book based on its ending and overall journey sentiment. The system uses web scraping to gather book reviews (e.g., from Goodreads), sentiment analysis to classify reviews into categories like “Ending” and “Journey,” and a machine learning model to predict a personalized score for the book. The project combines multiple components, including scraping, natural language processing (NLP), machine learning (ML), and data visualization, presented through a user-friendly graphical user interface (GUI). Throughout the semester, I have explored a lot of different concepts but wanted to try something new. I enjoy reading books, but sometimes the ending or overall plot progression determines how much I like or dislike a book. This project allowed me to merge my interest in books with my technical knowledge. It also challenged me to build a system that automates scraping, analyzing, and predicting—skills I wanted to refine.

    2) What was your desired learning outcome of your choice of final project?

    My goal with this project was to deepen my understanding of integrating machine learning models with real-world data pipelines, such as scraping and NLP. I also aimed to improve my ability to build end-to-end solutions and create meaningful visualizations. I wanted to learn how AI/ML tools can solve personal and practical problems while also mastering how to structure a Python project for readability and modularity.

    3) What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.

    The most useful aspect of this class has been learning data cleaning and preprocessing. I learned to remove errors, handle missing values, standardize formats, filter out irrelevant data, and categorize information. I also practiced transforming raw data into structured datasets ready for machine learning models, ensuring the data is reliable, consistent, and usable for accurate predictions.

    4) Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?

    All the work I’ve done this semester has given me a solid foundation in machine learning and AI. From building basic models to fine-tuning and integrating them into larger systems, I have a better understanding of how data-driven decisions are made. I also learned how to work with real-world data, handle errors, and optimize ML workflows, which are necessary skills for understanding the foundations of AI.

    5) Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    Yes, I definitely see myself pursuing coursework in data science, analytics, and AI/ML in college. I believe these experiences have made me confident and prepared for advanced studies.

    6) Include your Github repo URL so your classmates can look at your code.

    https://github.com/Srao2020/Book_AI

    7) Of all of the projects you or your classmates created this semester, which one was your favorite, and why?

    My favorite of mine is definitely this one because it is so useful to me. And I had the most fun with it.

  6. Tyler Slomianyj says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?

    This project implements a simple voice assistant with speech-to-text capabilities, natural language processing (NLP), and integration with Spotify to play songs and a timer feature. The assistant can understand spoken commands, interpret them, and execute actions based on user intent.

    What was your desired learning outcome of your choice of final project?

    I an AI Assistant that I could use in everyday life or at least in some circumstances that would actually be useful to complete certain tasks.

    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.

    The most useful aspect of this class is definitely the self-learning. I have a lot more freedom to pursue my interests and have fun with my projects. I also think learning specific AI concepts and learning more about AI in general is really important, especially in today’s society. I feel like I have learned a lot about today’s current situation with AI and a little bit how big models such as ChatGPT and others work.

    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?

    Yes, I feel this work has helped me understand some foundations of ML and AI. Before taking this class, AI sort of seemed like pure magic putting a few words and sprinkling magic dust on it and out comes a perfect response. After taking this class, I realize its a lot more advanced than that and it actually just comes down to math and algorithms. For example, the A* pathfinding and DDQNs.

    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?

    I’m not really sure. I think it would be interesting to pursue but also at the same time the concepts and mathematical background required for this type of path is very complex. I do anticipate being ahead in some of my classmates thanks to the things I studied this semester.

    Include your Github repo URL so your classmates can look at your code.

    https://github.com/tslom/AIAssistant

    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?

    Of all the projects I did this semester, this final project was pretty cool. I think it was something that was closest to actually being used in real life. The product I created was functional and I could think of some use cases (if siri didn’t exist). I might look to improve it and make it actually viable in real life circumstances.

  7. Cameron Morris says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?
    I decided to make a full football game simulator, I choose this project, because I thought it was the project I made with the most possible improvement. I felt like it was a project that I wasn’t fully able to complete because of how early it was in the year and how little I knew about ML. In addition, it was a project I was/am hoping to work on in the future, and thus this was the perfect opportunity to start working on it.

    What was your desired learning outcome of your choice of final project?
    Experience a more realistic way the people could use ML in the real world and just to see a way forward through an interesting project using machine learning.

    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
    Learning about ML and AI has been the most helpful, because it was the area that I knew the least about coming in to the class. It also showed me what they are able to do and how I can use them in the future and what the future could look like.

    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
    I think it has helped me understand the basics of ML and AI and realize how it is utilized in the world what are the best applications, how it could be threatening, or how it could be harmless, and what types of things it excels at and why.

    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    I think that I will, and I think I will be slightly ahead because of this, but rather I will be farther ahead because of the projects this class has inspired me to take on in the coming years.

    Include your Github repo URL so your classmates can look at your code.
    https://github.com/CameronJMorris/Project04_ML_Footbal.git

    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?
    Ryan’s project of making his own neural network, because it was cool and allowed him to learn about how the methods and libraries we use actually work.

  8. Preston Swigart says:

    My final project was a ML model built to predict the max temperature on a given day in Raleigh. I got my data for this from https://www.ncdc.noaa.gov/cdo-web/search, and I picked this project due to it being a dataset I knew I could make predictions on. Back in my first project, I really didn’t know what data would work and what wouldn’t, and I ended up picking a dataset without any real correlation because it looked like a fun dataset, so I wanted to rebuild this project with an actual working dataset in a jupyter notebook rather than a regular python file.

    The desired learning outcome for this project was making up what I hadn’t learned back during the first project. During the first project, I did not use a jupyter notebook and my model didn’t work very well due to a lack of correlation in my dataset. Having seen classmates using jupyter notebooks for their projects, I thought it would be a good skill to learn as well as actually building a working model.

    The most useful aspect of this was learning AI and ML. I have a decent understanding of github, and at least a base level of python knowledge, but I didn’t really know anything about how AI and ML works as seen by my first project being something that wouldn’t work.

    Absolutely, I think the linear regression project helped teach me how predictive algorithms, but also generative algorithms, function and how they work rather than being the mystery machine that just magically spits out answers. Even though it wasn’t related to my final project, I think the A* project also helped a lot with this, as it explains the foundations of many different applications that I regularly use such as google maps.

    I don’t know if I’ll take any data/analytical courses in college, but I think I will have a small advantage over others. My projects weren’t as complicated as others were, so I would not have the advantage that many others have, but I still think I will have an advantage.

    https://github.com/PrestonSwigart/MLProject_04

    Quite frankly, I don’t really know if I can pick one single project because so many of them were so cool. Ryan and Ori’s Hnefatafl (hope i spelled that right) AI, the AI music, so many projects were just super interesting. I really liked these two in particular because you could see the results super easily and it was really fun to watch it work.

  9. Joshua Yoon says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?
    My final project is a voice recognition software, a simplified version of voice biometric authentication where it intakes a voice file and then determines whether the voice is a “reliable” voice to grant access or not. It uses mfcc extraction, which are essentially voice prints, that provide unique characteristic for every signal. For my first and second project, I worked with a lot of neural networks and I thought voice recognition could also be a great extension of that, where every value of the wave can be transposed to an array, similar to an image, and then apply neural networks with a layer of sequential data analysis on them.

    What was your desired learning outcome of your choice of final project?
    Preferrably I would use markov chains to perform speaker identification, that is, the model being able to tell which voice belongs to whom. However, I had to divert from that idea a bit and simplify it into a binary classification of authorized speaker and not authorized speaker. Also, I would have liked for it to be a bit more accurate, which hopefully can be fixed with more training voice samples.

    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
    I think the underlying mathematic concepts about machine learning theories was very interesting while I was learning about them. Another useful takeaway was just bolting down those python programming paradigms in general such as using os or numpy. Furthermore, I think I learned how to navigate through material myself, that is, learning how to solve problems independently and finding the correct sources to learn from. As a purely project driven course, I think this course really helped me out in terms of allowing me to learn on my own and exploring things without guidance.

    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
    Definitely. I think I came into this class knowing little to none about how machine learning actually works. The only knowledge I had would have been a bit of linear regression. This course really opened my eye into the diverse branches of machine learning itself such as neural networks or reinforcement learning. But in a much deeper level, I was able to experience firsthand that AI was not just some magical, mystery tech but really just math, logic, and linguistics, that can totally be understood.

    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    I really enjoyed the mathematical analysis portion of exploring AI using calculus or statistics( also a class I am taking right now), and is also something that I would be interested in pursuing. Right now, even though I really didn’t get to do it this course, I want to connect the concepts I learned in this class to chemical analytics, which is also something I am passionate about. I’m not sure if I would be ahead per se, but I know I would have a solid understanding on introductory ml concepts.

    Include your Github repo URL so your classmates can look at your code.
    https://github.com/dbstjrgus/final.git

    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?
    I think my favorite was Cameron’s othello reinforcement learning project, mainly because it beat me in the game. After seeing a lot of my projects fail and kind of show some success it was really interesting seeing a model perform excellently.

  10. Alex Ru says:

    My project is an AI that plays Poker, and specifically Leduc Hold’em currently. Leduc Hold’em is a two-player variant of Poker where there are only 6 cards: King, Queen, and Jack of two suits. You are dealt one card at the start and can choose to call, raise or fold. After the first round, one community card is revealed, and you can call, raise, or fold. There are a maximum of two raises per round. I picked this as my final project because I wanted to do a game AI for a game with imperfect information. Also, Poker is a very classic game, so I thought it would be fun. The reason I chose Leduc Hold’em is that it is significantly simpler than No Limit Hold’em. My main model uses CFR (counterfactual regret minimization) and self-play to develop a winning strategy.

    I wanted to learn more about game theory and I was also curious about different optimizations for these algorithms. Poker is a classic game, so I thought it would be a good learning experience for me.

    The most useful aspect was learning about how to implement AI and models into different topics. I mainly explored game AI’s, but it was cool to also see how it could be applied to pathfinding. In general, I also know more about how AI works and some of the code behind the scenes.

    I think my work this semester has greatly helped me understand concepts about AI and ML. Trying out multiple stuff, from CNNs to PPO to A* to CFR, I have gained a wider range of knowledge regarding AI and ML.

    I see myself doing more stuff related to AI, ML, or data analysis in the future. I am likely to go into some information sciences major, so the stuff I have learned this class will probably be really helpful.

    https://github.com/alexru26/Project04_ML.git

    My favorite project was probably Ryan’s final project where he made his own Neural Network. That’s legit really really impressive. Massive props to him.

  11. Andrew Lim says:

    Please give a full description of your final project. Based on your prior work this semester, what made you pick this as your project?
    This project generates classical music. It trains off of every known song written and composed by Bach, then uses an LSTM to train off of the songs. Then, I take the model and run it through a music-generating code that makes music in the form of a MIDI file. Based on my prior work, it is similar to my first project because it uses neural networks to train off of a large dataset, but the nature of music generation is more complex.

    What was your desired learning outcome of your choice of final project?
    I wanted to learn more about how neural networks function and dive deeper into AI applications in music. In my first project, I had no idea what I was doing, so I just went off of a Kaggle repo that I found associated with my dataset and made tweaks to it. This time, I wanted to try making the model architecture myself and understand how changing inputs, architecture (i.e. layers), and other things would change how well my model performs.

    What has been the most useful aspect of this class? Learning more about Python, GitHub, PyCharm, AI, ML, or …? You decide and please explain why.
    I’ll answer two aspects for this question, one related to CS and one not. The most useful aspect of this class related to CS is learning how to code AI. That’s mostly the basis of this entire class, and that document we watched at the beginning shows the growing importance of AI in the world. The other aspect is learning how to tackle projects without guidance. This is a very difficult class if you don’t have any self-motivation, so this class helped a lot.

    Do you feel your work this semester, as summarized by your choice of final project, has helped you better understand some of the foundations of ML and AI?
    Yes, my work this semester has helped me better understand the foundations of ML and AI. The flexibility we had for each project allowed me to dive deep into many different aspects of ML/AI, such as pathfinding or game theory, and these foundations will be very helpful in the future.

    Do you see yourself pursuing data/analytical sciences coursework once you get to college? Do you anticipate being ahead of some of your classmates thanks to the things you studied this semester?
    I’m still not really sure if I will pursue data/analytical sciences in college, but I’m sure that if I were to, then I would be set up nicely thanks to this class.

    Include your Github repo URL so your classmates can look at your code.
    https://github.com/AndrewLim0314/Project04_MusicGenerator

    Of all of the projects you or your classmates created this semester, which one was your favorite, and why?
    I think my favorite project from this semester is Ryan’s Spotify Daylist recreation because I also enjoy listening to the Spotify Daylist quite often, and seeing one of my classmates make a very successful and well-made project as the introductory, starter project was very impressive.

  12. Danielle Li says:

    1. I really enjoyed making my first linear regression model, so I wanted to do something similar. For my NBA model, in the beginning, it started off as a prediction model that instead of predicting the number of points, it would predict the win percentage. I then looked into what I could predict with accuracy, and I had wanted to do a bot at the time, so I ended up choosing to do a bot that sorted emails. This was also a modern problem because of the amount of college spam emails I receive on a daily basis, so I thought being able to improve this would be fun. My final project first trains on 80% of a csv where I have 2000 regular emails that I generated, 1000 spam ones that I generated, and 1000 college spam emails that I copied and pasted. It ran a TF-IDF on all of the emails and found the words that were the most unique to each class of email and put high weights on them. There were 3 separate logistic regression models that trained for each model, and when it asks for user input, it puts the email in each logistic regression model to get a weighted sum, which is then put in a sigmoid function that turns it into a probability of the email being that class. Then the algorithm prints the class with the highest probability.

    2. I wanted to do more with data training and statistics because I find that the most applicable to my life. If I were to do future projects, it would be on a similar topic within machine learning and not something like game design. I also find it fun to see how the accuracy changes depending on the data and any trends with that.

    3. Learning about Machine Learning has been the most useful aspect because I find something completely new to present about each class like I would’ve never imagined that I would learn about how linear or logistic regression models work or concepts like the NEAT neural networks. I think writing the exact code is still hard, but learning these concepts have been the most useful aspects because then I can write code based on my knowledge.

    4. Yes, I think I definitely better understand the foundations of ML especially with the previous projects because I had very little to no knowledge of ML or AI before this class, and being able to implement it in projects that I’m interested in has helped me a lot.

    5. I do see myself pursuing data/analytical sciences when I get to college because I want to learn more about integrating data science and biological sciences like biochemistry. I do think I’m ahead of some of my classmates because of statistical analysis methods that I can now code like Linear or Logistic Regression models.

    6. https://github.com/dani0621/Project_04_EmailSort.git

    7. My favorite is either theFlappy Bird Game or the NBA Predictor model because both had way higher accuracy than I had predicted, especially the NBA predictor model. The Flappy Bird Game was just fun because of learning how NEAT neural networks work and seeing the bird play for hours at a time.

Leave a Reply