chengtao

4 ms·

1.
by chengtao 12y ago
To add on top of that, even with great data analysis skill, I had another blog-post talking about it requires all the product, data, and engineering skills together to make a good data science team. http://ml.posthaven.com/w
2.
by chengtao 12y ago
As you pointed out the transforming features is powerful, I believe that's the exact reason which makes SVM powerful. Though the way features can be combined with SVM is limited, the limitation makes SVM training fast in the dual space
3.
by chengtao 12y ago
great comment and +1
4.
by chengtao 12y ago
Yes, and IMO, most of the time, the insight behind the data is far more important than the modeling algorithms to achieve high performance with few exceptions (say computer vision, NLP, etc which really requires A LOT OF data). Even in some
5.
by chengtao 12y ago
I personally love the topic of bayesian optimization over all the possible parameters including model choice. My point was more about given the resource is always constrained, it typically pays off long term for practitioners to analyze the
6.
by chengtao 12y ago
All the other comments are great. Just bear in mind that it's important to really understand the mechanics behind each importance measurement. Some can use information gain, some can use the t-test on coefficient, while some use random
7.
by chengtao 12y ago
Great question and my main point is less about up sampling the rare cases but more about the default loss function used in the model training might not directly align with the final business metric (which is the metric practitioners should
8.
by chengtao 12y ago
Also, thank you all for reading the post. I'm the author and I'll be happy to clarify any of the points in the blog~
9.
by chengtao 12y ago
Not really, it was actually more inspired by Statistics Done Wrong.
10.
Code quality: speed + comprehensibility(ml.posthaven.com)
2 points by chengtao 12y ago|0 comments
11.
A/B testing – statistical hypothesis testing vs. multi-armed bandit(ml.posthaven.com)
2 points by chengtao 12y ago|0 comments
12.
Common mistakes when building machine learning models(ml.posthaven.com)
11 points by chengtao 12y ago|1 comments
13.
Bridging the gap between lean startup in theory and in practice (ml.posthaven.com)
1 points by chengtao 12y ago|0 comments
14.
by chengtao 12y ago
great point and this is the exact rationale behind the architectural design
15.
by chengtao 12y ago
if you are interested, there is also a separate blog post, http://www.codecademy.com/blog/143-eventhub-open-sourced-fun... , in which we talk about some high level architecture consideration
16.
EventHub: Open-sourced Funnel analysis, Cohort analysis and A/B testing tool(codecademy.com)
7 points by chengtao 12y ago|0 comments
17.
Why building a data science team is deceptively hard(codecademy.com)
7 points by chengtao 12y ago|0 comments