want to learn Python

[kum5758]

bro's

emaina whatsgroups unte pm on python or pyspark

[BetaCat]

8 minutes ago, kum5758 said:

different things different use cases generate aithay and each can try any use case

But what's the point of learning other cases when you are not going to use them for work. Python is just a tool, it's not like a fancy science that you will miss out on if you didn't know or something that's remotely interesting. 

[mettastar]

16 minutes ago, kum5758 said:

konni use cases create chesi daily status and code share cheskovadam better anpisthandi

few use cases

1. take a .csv file read it using python, add a new column and convert into parquet or avro file

2. connect to databases (SQL server, oracle) using python

3. take a sample api- extract the data

4. take a JSON and convert into .CSV

Good idea

[jlo]

Connecting sql etc all are outdated.  Don’t trap in that.  
 

do like this, few ideas:
 

Get stock data from yahoo

parse it and do some data analysis. I can give more details if you need. 

[BetaCat]

I can't follow this thread, as in I can't understand what people are talking about. I need help. 

[macha]

Multiprocessing on dataframes I have 4 million rows with 2 columns ...evarikina idea untey zarra help karo... google not helping me in this case...

[meri_zindagi]

4 minutes ago, macha said:

Multiprocessing on dataframes I have 4 million rows with 2 columns ...evarikina idea untey zarra help karo... google not helping me in this case...

What you are trying to do

[macha]

3 hours ago, meri_zindagi said:

What you are trying to do

I have 32 processors I want create 30 process assign a chunk of data frame to one process and join resulting data frame to one single data frame from all 30 process and write it out to text file 

[ekunadam_enkanna]

There are some strategies that work independent of any programming language. For instance, rotating an array by k times: the typical way is to call k times another function rotateArray. The other way is to append the array or in terms of strings: AA, then just call AA[k:len(A)]. This trick is also used in string algorithms: how to check whether string A is a rotation of string B in linear time? Construct a string: A$BB, and use Z-algorithm to check whether A matches the string BB.

Another example: mechanics of mergeable heaps (melds) such as Fibonacci, binomial heaps is same as that of LSM (log structured mergeable) trees. Over the time, you have to learn to see patterns at a deeper level. That way, you can retain some insights.

 

[pachimirchi]

16 hours ago, cosmopolitan said:

Data structures and algorithms.. python tho try cheyi .. u will get used to python easily 

+1 worked for me

[GullyBoy]

9 hours ago, BetaCat said:

I can't follow this thread, as in I can't understand what people are talking about. I need help. 

dial thommidondalapadakondu 

[pachimirchi]

4 hours ago, macha said:

I have 32 processors I want create 30 process assign a chunk of data frame to one process and join resulting data frame to one single data frame from all 30 process and write it out to text file 

Good use case for spark. If you just want to use cython try multiprocessing module

[JUST444FUN]

9 hours ago, jlo said:

Connecting sql etc all are outdated.  Don’t trap in that.  
 

do like this, few ideas:
 

Get stock data from yahoo

parse it and do some data analysis. I can give more details if you need. 

please share some light bro.. any usefull api's or references

[macha]

3 hours ago, pachimirchi said:

Good use case for spark. If you just want to use cython try multiprocessing module

I did Multiprocessing module try no improvement in the performance there a method which gets response from a url and processes it I feel that’s the bottle neck will multiprocessing dose take care of url request response also ?