DEV Community: williamjfermo

Categorical data and a hungry girlfriend.

williamjfermo — Mon, 16 Mar 2020 03:40:53 +0000

You can get data anywhere. But what better way then get data from personal life experiences. Here I gathered data regarding on what my girlfriend would want to eat. Seems like it would be a simple problem but believe me it is not. I gathered the data and had to deal with categorical data. Project is still in progress but would hope to be able to make a machine learning model that would accurately predict what my girlfriend would want to eat.

Check out my github project Here

Life after a coding bootcamp

williamjfermo — Mon, 09 Mar 2020 05:20:35 +0000

Will be posting more about live after a coding bootcamp.
What I do to keep myself motivated.
New Technologies I learn to add to my data scientist tool kit.
Attend as many events as possible to build up my network.

Follow here to see what I'm up to.

Networking with Benefits

williamjfermo — Fri, 15 Nov 2019 20:25:41 +0000

I was not born to talk to people. I was super introverted all the way through college. I was okay with studying then go home and play videogames all day. It didn't take till I went to graduate school that I started to open up more and took to the challenge of wanting to meet other people. So out of self perseverance I did something I was uncomfortable with and that was to talk to people. I was starting a new graduate program and new no one. I attended meetups and started to meet people. Before I knew it I made friends.

That same philosophy I am applying to being a data scientist I know nothing about the field other then what I've been studying but I want to know more about the industry. One way to do that is through networking.

Every person I network with I get new knowledge that may help me reach my goal. They give me tips and even motivation that what I want is within reach I just need to keep fighting for it. Sometimes you need that extra pair of eyes to show you something about yourself you didn't know you had.

To get myself more entrenched in learning about the industry I have been joining meetup groups that are in the lines of what I am interested such as data science, machine learning, python, etc... Meetups are not only a great way to network but they also have lectures and lightning talks helping you gain more knowledge.

Now that I have went over why to network now I will give you some tips on how I do it and how you can apply it when you need to network.

You have to approach networking like you are trying to get a beautiful woman's/man's number. If you can do that. You can network. Now if you can't then here are some tips how I do it and you can do it too.

This is a picture of me in a gorilla costume I ran a 5k in. It was not comfortable and after the race I took off my mask and still looked like a mess. I talked to a female runner after the race still in gorilla gear and got her number. How that happened wasn't because how I looked it was because of confidence. Don't go up to anyone and be timid, be confident. I've tried a bunch different lines before but the best line to use is a simple "Hello".

Learn to talk to people in groups. It will be rare you see someone just standing by themselves and waiting for you to talk to them. The people you will be wanting to talk to will be the same people others want to talk to. Waiting for them to be alone will often not happen, they will always be talking to someone. You have to somehow insert yourself into the conversation. Like when joining different dataframes you need to find that key. I try to over hear what people may be talking about then just start talking about what they are talking about to get in there.

Now that the hard part is over what do you talk about? I always tailor my conversation depending on where I am at if you are networking after a meetup or at a bar you can't approach them the same way. Don't go straight into give me a job. Make it conversational and ease into it.

Questions I sometimes ask:

What are highly sought after skills in your industry?
How do you like your job?
Are you hiring?

Questions I get asked:

Why data science?
Where did you work before?
Do you have a kaggle account?
What are your plans after?

If you think this contact is networkable get their card, linkedin, or even a phone number. Never forget to do this. Always close!

After linkedin connect send a follow up message saying it was nice meeting them and personalize it with something you may have talked about.

Now if you follow all these simple steps and still get a no. Just do it again. Rejection is part of the process. The important part is being able to brush it off and being able to not stop trying.

If you follow these simple things you should be able to network professionally.

Quick forecasting with tableau

williamjfermo — Tue, 05 Nov 2019 19:58:29 +0000

I have been teaching myself tableau and found it a great way to visualize data with little effort. While doing my last project I found you can do forecasting with tableau. This quick tutorial will show you how to get forecasting with your data with just a couple of clicks.

As a warning if you have a data set with over 15 million rows Tableau Public will not work with it. I believe Tableau Desktop removes these caps so if you have a data set larger then 15 million I'd recommend using something other then Tableau Public or pay for Tableau Desktop.

Import time series data. One thing I noticed with tableau you don't necessarily need to do any encoding to the time series data it will instantly recognize it as time series data and you can quickly do analysis over it.

Transfer the datetime dimension you want to analyze and drag to columns section.

Transfer the measure you want to analyze for your target in the rows section.

If the dimension you selected includes date and time you can quickly seperate that to analyze quarterly, monthly, weekly, or daily. It will start of with year then just click the plus and you can select which specific datetime you want to analyze. Here I will be doing months.

Click on analytics and more options will open up. Under models select Forecast and drag that to the main sheet.

Quick forecast with a couple of clicks.

Will go more into detail about the forecast and other options in a future post.

Day in the life of a Data Science Bootcamp

williamjfermo — Fri, 27 Sep 2019 15:31:27 +0000

Day in the life of a Data Science Bootcamp.

Having been in the program for a month my process has changed from the first day I arrived. That first day I was eager to start the Data Science program with wide eyed optimism and excited to change my life . A month later I’m still somewhat optimistic but things are a lot harder. I’ll be giving you my honest experience of what a day in the life of a coding bootcamp is like.

My bootcamp is 15 weeks and by the end of that I should be able to have the knowledge I need to get a job in Data Science. Schedule is from 9 am to 5 pm. I remember the first week I’d come in around 9 and have a decent night's sleep but i’d be fighting traffic for an hour and that time was wasted when I could be studying. Now I come in around 6:30 am this gives me an extra couple of hours to study, which I need. I turn on my laptop, grab a cup of coffee ,and settle in for the long day.

I use this morning before classes start to brush up on my Python. If you plan on getting into any bootcamp I suggest really do a month or more pre work before starting. This way when the material comes you won’t feel as overwhelmed. I don’t have a tech background and my Python skills were non existent. I feel like I'm in a rpg where I’m still battling bugs for experience while my classmates are battling monsters.

Two hours pass and now it’s 9 am. We check in for the day and often our data science coach will greet us and ask us if there are any questions about any of the material. Material is a lot. Our current module is statistics and we should take 3 weeks to go through it. There are brief lessons and labs. I try my best to do the labs but end up looking at the solution and looking back. We will have a lecture in a couple hours and I’ll go make myself a caffeine free hot tea to trick my mind into being awake even though I’m already tired. We have one low key lecture in the morning and one project based lecture in the afternoon before lunch.

A lot of us just bring our own lunch to save money since none of us are working and this is a full time course. That break for lunch isn’t really scheduled, sometimes I’ll just take a moment for myself to walk around or skip lunch to get some extra studying done. There really isn’t any breaks and any are all self imposed. Only times I get a break is when I go grab something to drink or use the restroom. All that time is used for studying.

There is an afternoon lecture and we can attend it or choose to focus on other materials. At this point it’s around 3 pm and the hot tea isn’t working for me. I go up to our 10th floor where the good coffee is and get myself an espresso shot. Sleeping isn’t a luxury I can say I have. I try during this afternoon session to study up on lectures we will be having the next day so I can understand the concepts better when it is presented to us. When it reaches 5 pm I go over the notes I made for the day and look over it to review what I learned for the day. By 6 pm I’ve been studying for almost 11 hours. I take a break and go up to the 10th floor where it is more casual and go over my python again for another hour.

Normally by 7 pm I go home or later depending how traffic is. I go home, eat dinner, set out my clothes for the next day, try to sleep before 11, and then do it all over again. I knew a bootcamp wasn’t going to be easy and I would have to work for it. I didn’t know it was going to be this hard. I think most days I have doubts I have what it takes to make it in this field but I have to fight those thoughts to keep pushing through. When I'm not studying, I feel like I should be studying. Bootcamp can be physically and mentally tough but you need to prepare yourself for that. It’s all about how bad you want it. I want it bad. Maybe it’s easier for others but this is my experience.

My friend sent me this quote that help me get through one of my rough days and I hope it will help you too when times are tough during your bootcamp.

“If you really want something, you will find a way. If you don’t you will find an Excuse”

I attended my first hackathon.

williamjfermo — Thu, 05 Sep 2019 12:44:33 +0000

The hackathon was organized by Yuuvis a content manager api. Yuuvis allows businesses to effectively obtain, organize, store, and deliver critical information. This was the platform all the groups would need to use for their projects. There was a prize pool of 10,000 plus a trip to Berlin at stake for the grand prize. So teams could win at minimum 2,500 dollars if they won one of the categories.

My technical knowledge was basically some pre-work i did for Flatiron Good thing about hackathons is that you don’t necessarily need to be a coder to attend. I think it is important to have quality coders, web developers, and data science but sometimes it also helps to be a person with that great idea. I had to choose a profile before I registered and chose super power, which entailed not being a techie but the ability to help bring projects to life.

Saturday morning there were a lot of people all with their laptops and almost all the tables were occupied. Yuuvis introduced themselves and how excited they were to have this hackathon. I could feel the nerves stir inside me as I didn’t know how I could actually contribute to any project. First order of business was if there were any people without any groups. Maybe less than ten people raised their hands without a group including me. I think many of the groups either knew each other or found groups through slack. I had just installed slack a couple days before the hackathon started so using it was new for me. To help groups needing members every person without a group had to speak to the crowd about their credentials or in my case lack of. After that was done everyone got into their respective groups and got to work. I had to find myself a group. I talked to a couple of groups and they weren’t interested. Finally found a group interested in having me and I decided to join them.

Now we had a couple of coders, a data scientist, a web developer, a guy with an idea, and me. Our project was to be based on collecting police reports and autopsy reports to see trends in the data to find serial killers, the Holmesbot3000. First thing we did was research to see if there was any similar ideas in the market and if this was a viable project. One of the guys was a solution architect it was his first hackathon also but he was joining to get back to coding. He started writing out a plan for us to do and what we needed to do to get it done. Civil justice institutions would import police reports and autopsy reports into our web based program. The program would pop up with an alert that there were similar cases and it could be a possible serial killer. The Holmesbot 3000 was born my first hackathon project. I was in charge of research, slides, and found standard data sets to use for police reports and autopsy reports. The Hackathon felt similar to studying at flatiron but during the day they would have free food, snacks to keep our energy levels up. They would also have little breaks for stretching and a competition playing beat sabre, a vr game. We started at 8 am and finished around 6 pm. Before we left we had assigned things to be done before we would begin again on Sunday. We would communicate through the night on slack.

Sunday came and projects were due by 11 am. Everyone in my group seemed to be rushing to get things done and worse part is one of our team members who’s idea it was for the project skipped out on the final day. He was going to be the one doing the pitch, with him gone I volunteered to do the intro for the pitch. We were about to skip pitch practice to work on the project more but were able to get one of the last slots for pitch practice before submitting our project. Pitch practice involved us doing an informal pitch with some members of Yuuvis. All presentations had to be 4 minutes with questions. They gave us some quick pointers and we went back to work on the submission. Group presentations were interesting as it seemed like there were a variety of different groups with even people solo presenting a project. The room also didn’t seem as packed as Saturday with groups not submitting their projects at all. I was nervous to talk but I wanted to contribute more to the group then just research and making slides. We didn’t practice other than that quick pitch to Yuuvis. Things didn’t run as smoothly as I wanted and could have been avoided if we did an actual dry run of our presentation. When all was said and done we submitted a project and I was happy to be able to say I was a part of that Project.

Things I learned from my first hackathon. You can win prizes by just showing up. Coding knowledge is important but not a necessity to be able to contribute. Someone needs to take a lead to make a plan and assign tasks to get project done. Slack is a great tool to communicate with people and for projects. Great presentation skills help you get your project known. Hackathons are a great place to network there are people who have established teams but there are people like my team that literally just came together at the last minute. You can actually learn from other people. Yuuvis provided mentors with a variety of knowledge people could go to for any help they might need on their projects.
Hackathon's can be exhausting but rewarding. Nothing like seeing what you were able to accomplish after such a short time. Finally they feed you well and who doesn't love free food?

Seeing all the projects and what people could do with information really opened my eyes about why I wanted to get in the field. There were projects like One Chat, a messaging app that allowed you to send and receive messages in the language of your choice without the need to click translate. The winner was, Hire Us, a resume retrieval and filtering project. They won 5,000 and a trip to Berlin. We didn't win but hopefully my next Hackathon I can contribute as a Data Scientist.

Post party hackathon with DJ!
Bonus Post Party pic

LINKS
Yuuvis
Yuuvis Hackathon
Holmesbot 3000 Project.
Flatiron Bootcamp