Hi Debu,
On 27/12/16 08:18, Andrew Howe wrote:
> 5. I got a prediction result with True Positive Rate (TPR) as 10-12
> % on probability thresholds above 0.5
Getting a high True Positive Rate (recall) is not a sufficient condition
for a well behaved model. Though 0.1 recall is still p
Your model is overfit to the training data. Not to say that it's
necessarily possible to get a better fit. The default settings for trees
lean towards a tight fit, so you might modify their parameters to increase
regularisation. Still, you should not expect that evaluating a model's
performance on
Dear Joel, Andrew and Roman,
Thank you very much for
your individual feedback ! It's very helpful indeed ! A few more points
related to my model execution:
1. By the term "scoring" I meant the process of executing the model once
again without ret
On 27 December 2016 at 18:17, Debabrata Ghosh wrote:
> Dear Joel, Andrew and Roman,
> Thank you very much
> for your individual feedback ! It's very helpful indeed ! A few more points
> related to my model execution:
>
> 1. By the term "scoring"
Thanks Guillaume for your quick feedback ! Appreciate it a lot.
I will definitely try out the links you have given. Another quick one
please. My objective is to execute the model without retraining it. Let me
get you an example here to elaborate this - I train my model on a huge set
of data (histo
On 27 December 2016 at 19:38, Debabrata Ghosh wrote:
> Thanks Guillaume for your quick feedback ! Appreciate it a lot.
>
> I will definitely try out the links you have given. Another quick one
> please. My objective is to execute the model without retraining it. Let me
> get you an example here t
Hi Guillaume,
And when I say that I have been able to run my models
using joblib.load, I meant that I have run using joblib.load on a completely
different dataset compared to the one I used for model training. And I got very
similar result to joblib.load run as compared
I am not sure to understand your terminology. While calling joblib.load,
you actually load the RandomForestClassifier. Therefore, calling predict
from the estimator loaded with joblib is identical as using the
RandomForestClassifier which you trained at the first place.
I think that it would be mu
