Search for a command to run...
I am a researcher, writer, engineer, and storyteller. I started writing to help me express myself through storytelling. I made the website to build my portfolio and show the different projects and activities I am involved in.
Hi Kaniz! Thank you for asking. I do not have any recommendations for any specific courses or materials for learning Python or machine learning. In my opinion, the best way to learn these is by doing projects. You'll find many ML projects on Kaggle. I do plan to write more about these topics in future posts. So please keep checking back or subscribe to be notified automatically!
Thank you Reem! 🙏
Why Should You Produce Creative Works?
If you are a researcher in STEM, you might come across a few colleagues on campus who are LaTeX fanatics. Well, I am not a fanatic. I do use it when the time calls for it. Otherwise, I would stick with the good old Word. LaTeX came out in the 1980s. ...
During my graduate studies, I had the opportunity to work on a few research projects. I have used Python extensively in these projects. In this post, I will discuss six areas where Python has helped me. If you are doing research and love programming, these reasons may convince to start learning and using Python right away!
One of the key activities in research is collecting and analyzing data. Python has great libraries for processing and analyzing data. Whether you want to remove the anomalous data points, normalize the dataset, or fill in the missing points, Python has you covered. Two libraries often used for data processing and analysis are NumPy and Pandas. They are powerful and efficient. They come with a lot of built-in functions that can make your data analysis tasks less daunting!
Quantitative research often relies on statistical methods for drawing important conclusions. statsmodels is a Python library for statistics. Whether you want the vanilla descriptive statistics or a fancy multivariate regression model, this library can help you out!
Presentation is everything. Once you have analyzed your data, you want to present them in different types of plots: histograms, bar plots, pie charts or others. Matplotlib is the most popular plotting library for Python. It even supports 3D plots. Furthermore, for statistical data visualization there is the seaborn library. It can help you create heat maps, correlation plots and so on.
Python’s greatest strength is in the area of machine learning. There are plenty of libraries that support Python in machine learning. If you want classical machine learning, you can start with scikit-learn. If you want deep learning, you can go with Pytorch, Tensorflow, or Keras. The application areas for machine learning are far too wide to cover here.
Python has great libraries for scientific image analysis and manipulation. Two libraries often used for image analysis and computer vision are OpenCV-Python and scikit-image. You can combine these libraries with deep learning models to do object recognition.
Finally, Python can be used as a scripting language to automate repetitive tasks. For example, if you need to do the same type of analysis on 1,000 images, a for-loop or a while-loop will save you many hours of frustration. Although it may take hours to setup the first image, the rest will be quicker!
Here, I have given you a quick overview about how you can start using Python as a researcher. There are plenty of Python libraries out there which are hosted on PyPI, Anaconda and other repositories. You can check out Github for many open-source projects based on Python. Great thing about using Python is that you do not have to re-invent the wheel. All you need is to start learning and using it right away!