Key takeaway on how to start your ML journey

I recently gave a talk on ML and that has inspired me to tread down the path and explore more.

I can't call it a recent post but it is a recent post from Jason Brownlee that I read which has given me better direction to add some seriousness to this field

Some key takeaway on how to start mine/your ML journey:

  1. DO NOT go through ML tutorials , Get onto hands-on realtime problems 
  2. Get onto Kaggle
  3. Follow the likes of Triskelion 
  4. Reproduce what others have done, DO NOT reinvent the wheel
  5. What other tools ? Vowpal Wabbit Scikit-learn
  6. R or Weka or Python or DO NOT have a weapon of choice
  7. "Competing consistently is the key to getting good."
  8. New to Data Science? Get started with a tutorial on our most popular competition for beginners, Titanic: Machine Learning from Disaster.
  9. Machine learning is the hottest field in data science, and this track will get you started quickly. https://www.kaggle.com/learn/machine-learning

  • Practice a lot: Do as many challenges as you can, incremental improvements.
  • Study evaluation metrics: Really understand AUC, etc. (see a list of metrics)
  • Study the domain: Business cases, papers, state of the art, feature engineering
  • Team up: Top 10 finish is hard, but he need to team up to achieve it.
  • Read the forums: Post to competition threads, understand winning solutions.
  • Share on forums: Lots of angles on a given problem, don’t share too much.
  • Use ensembles: They always improves results, can give you a top 10 with simple models.
  • Experiment: Try out ideas rather than living in thought
  • Creativity: Think outside of the box
  • Tools: Find and use good algorithms.
  • Tuning: Use cross-validation, tune all model parameters.

Comments

Post a Comment

Popular posts from this blog

Software Testing @ Microsoft

Image
In continuation from my previous  series of posts  targeting testing in top few companies from the technology space. After  Software Testing @Facebook  and @Google  we shall now look at Microsoft: Testing@Microsoft Software Development Engineers in Test (SDETs) are usually just called Test and sometimes Software Testing. SDETs are responsible for maintaining high testing and qualityassurance standards for all Microsoft products. Software Development Engineers (SDEs) are often referred to as Software Development. SDEs write the code that drives Microsoft products and upgrades. Very few teams in Microsoft are still using SDETs to do significant amount of manual testing. Manual testing is mainly outsourced. Actually, as early as in 2004, for example, most of the manual testing of MSN Explorer was outsourced. Think about it this way: it just doesn't make sense for test manager to spend their headcount (SDETs) on something can be easily outsourced.  Until 2005, Microso
Read more

Trim / Remove spaces in Xpath?

Image
If you are wondering how you can trim the spaces in your xpath, this is how: The normalize-space function strips leading and trailing white-space from a string, replaces sequences of white-space characters by a single space, and returns the resulting string. Example : Which one is better? //td[starts-with(normalize-space(),'Text to Trim')] //td[starts-with(normalize-space(text()),' Text to Trim ')] The second option is better as it targets at the specific nodes More string functions in XPATH: Functions Description starts-with( string1 ,  string2 ) Returns true if the first string starts with the second string. contains( string1 ,  string2 ) Returns true if the first string contains the second string. substring( string offset length ) Returns a section of the string. The section starts at  offset (which is a number), and is as long as the value provided at  length. substring-before( string1 ,  string2 ) Returns the part of  string1  up unti
Read more