Top 5 Skills a Data Scientist must have.

-

Data Science is a competitive field, and people are quickly building more and more skills and experience. This has given rise to the booming job description of Machine Learning Engineer, and therefore, my advice for 2020 is that all Data Scientists need to be developers as well.
To stay competitive, make sure to prepare yourself for new ways of working that come with new tools.

1. Agile


Agile is a method of organizing work that is already much used by dev teams. Data Science roles are filled more and more by people who’s original skillset is pure software development, and this gives rise to the role of Machine Learning Engineer.

More and more, Data Scientists/Machine Learning Engineers are managed as developers: continuously making improvements to Machine Learning elements in an existing code base.
For this type of role, Data Scientists have to know the Agile way of working based on the Scrum method. It defines several roles for different people, and this role definition makes sure that continuous improvement and be implemented smoothly.

2. Github


Git and Github are software for developers that are of great help when managing different versions of software. They track all changes that are made to a code base, and in addition, they add real ease in collaboration when multiple developers make changes to the same project at the same time.
With the role of Data Scientist becoming more dev-heavy, it becomes key to be able to handle those dev tools. Git is becoming a serious job requirement, and it takes time to get used to best practices for using Git. It is easy to start working on Git when you’re alone or when your co-works are new, but when you join a team with Git experts and you’re still a newbie, you might struggle more than you think.

3. Industrialization


What is also changing in Data Science is the way we think about our projects. The Data Scientist is still the person who answers business questions with machine learning, as it has always been. But Data Science projects are more and more often developed for production systems, for example, as a micro-service in a larger software.

At the same time, advanced types of models are getting more and more CPU and RAM intensive to execute, especially when working with Neural Networks and Deep Learning.
In terms of job descriptions of a Data Scientist, it is becoming more important to not only think about the accuracy of your model but also take into account the time of execution or other industrialization aspects of your project.

4. Cloud and Big Data


While industrialization of Machine Learning is becoming a more serious constraint for Data Scientists, it has also become a serious constraint for Data Engineers and IT in general.

Where the Data Scientist can work on reducing the time needed by a model, the IT people can contribute by changing to faster compute services that are generally obtained in one or both of the following:
  • Cloud: moving compute resources to external vendors like AWS, Microsoft Azure, or Google Cloud makes it very easy to set up a very fast Machine Learning environment that can be accessed from a distance. This asks from Data Scientists to have a basic understanding of Cloud functioning, for example: working with servers at distance instead of your computer, or working on Linux rather than on Windows / Mac.

  • Big Data: a second aspect of faster IT is using Hadoop and Spark, which are tools that allow for the parallelization of tasks on many computers at the same time (worker nodes). This asks for using a different approach to implementing models as a Data Scientist because your code must allow for parallel execution.

5. NLP, Neural Networks, and Deep Learning


Recently, it has still been accepted for a Data Scientist to consider that NLP and image recognition as mere specializations of Data Science that not all have to master.

But the use cases for image classification and NLP get more and more frequent even in ‘regular’ business. At current times, it has become unacceptable to not have at least basic knowledge of such models.
Even if you do not have direct applications of such models in your job, a hands-on project is easy to find and will allow you to understand the steps needed in image and text projects.

Comments

Post a Comment

Popular posts from this blog

Scenario based interview questions for Data science!

-
Image
  Q1. You are given a train data set having 1000 columns and 1 million rows. The data set is based on a classification problem. Your manager has asked you to reduce the dimension of this data so that model computation time can be reduced. Your machine has memory constraints. What would you do? (You are free to make practical assumptions.) Answer: Processing a high dimensional data on a limited memory machine is a strenuous task, your interviewer would be fully aware of that. Following are the methods you can use to tackle such situation: Since we have lower RAM, we should close all other applications in our machine, including the web browser, so that most of the memory can be put to use. We can randomly sample the data set. This means, we can create a smaller data set, let’s say, having 1000 variables and 300000 rows and do the computations. To reduce dimensionality, we can separate the numerical and categorical variables and remove the correlated variables. For numerical variable...
Read more

What is Data Preprocessing?

-
Image
  Data preprocessing  is a step in the data mining and data analysis process that takes raw data and transforms it into a format that can be understood and analyzed by computers and machine learning. When we talk about  data , we usually think of some  large datasets  with some rows and columns. While that is a likely scenario, it is not always the case — data could be in so many different forms:  Structured Tables, Images, Audio files, Videos , etc. Machines don’t understand text, image, or video data as it is, they understand 1s and 0s. Real-world data also contains noises, missing values, etc. which cannot be directly used for ML models. Hence,  data preprocessing  is required for cleaning the data and making it suitable for an ML model which increases the accuracy and efficiency of the model. It involves the following steps: Getting the Dataset Importing Libraries Importing Dataset Data Quality Assessment: i) Finding and Processing Missing/Inc...
Read more