May be in machin learning where is black box..but linear regression already providing contribution like intercept is your base value..and multiply the predicted coefficients with data points of that variable..if coefficients is negative then contribution is on decreasing side
Hi Aman... It was good to know about this library. But I still have the question about which you touched a little. When we can get weights to see the coef and bias, how does this make it different ?
Lets say my prediction for first records is 20000. I want to know what makes it 20000. Can you tell me 14000 of this 20000 is made of feature1, another 2000 from column 2 and another 2000 from feature 3 using coef anf bias?
May be in machin learning where is black box..but linear regression already providing contribution like intercept is your base value..and multiply the predicted coefficients with data points of that variable..if coefficients is negative then contribution is on decreasing side
Thanks
Could you please let us know how the base value is calculated?
Thanks!
Hi Aman... It was good to know about this library. But I still have the question about which you touched a little. When we can get weights to see the coef and bias, how does this make it different ?
I am bit unclear on that. Pl help
Lets say my prediction for first records is 20000. I want to know what makes it 20000. Can you tell me 14000 of this 20000 is made of feature1, another 2000 from column 2 and another 2000 from feature 3 using coef anf bias?
Can you explain PSO please?
how to minimize its computation time
It is not slow always - I tried in VS code and local jupyter as well. I was not very dissapointed.
Hi Aman. Can you pls show, the data in this model. y = x1+x2+x3...+xn. ie. 1000 = 300+250+250+...+150+50, some thing like that.pls