dbsc | upgrading databases using SQL scripts stored in source | SQL Database library

I use drop_duplicates() to delete the duplicate data, and then there is no error. I think the memory error is caused by too much data.

you should take care of the null when it return from (first()) method:

If I understand correctly, you are looking for the precise y value of the inflection point appearing in your ε(x) plot (it should be around 2.0), right?

If this is correct, being ε(x) your curve, the problem is reduced to:

1. Compute the second derivative of your curve: ε''(x).
2. Find the zero (or zeroes) of such second derivative: x0.
3. Recover the optimized ε value, just by plugging the zero into your curve: ε(x0).

Here I attach my answer, based in this two other Stack Overflow answers: https://stackoverflow.com/a/26042315/10489040 (Compute derivative of an array) https://stackoverflow.com/a/3843124/10489040 (Find zero in array)

So the answer to this problem is simple: when you call exit_json or fail_json you can't have sys.stdout with any sort of output. The reason is also simple: after your code is ran the result is processed by this Ansible code. Which parses res.get('stdout', u'') and if there are any parsing errors then you end up with that massive STDOUT dump I mentioned in my question.

The solution is not that simple. I tried various hacks to empty STDOUT before calling Module exit methods but it didn't really worked. So since I have a full control over what I'm calling I simply changed my logging to temp-files based in the library I call and did few other things to ensure that I'm not using sys.stdout in any way. Which is really silly and I hope that someone can show me a better way

You could try a different approach and let WMI determine if below your minimum values.

Store the labels in an additional column in the original DataFrame