Is there any good way to understand the book, "deep learning" by Ian Goodfellow ? The book is quite math heavy, and I don't understand several things. Like how one equation leads to other. Is there a guide book of sorts to help me? Or is there any other book which can hold my hand while going through math? Any math book I could read first which will help me through this?
It takes a lot of efforts to share answers here and some idiots just downvote everything because you commented on their answer but didn't replied back to their re-commemt
Not really. There is a lot to learn about how all the options and hyper-parameters work. Probably most important is normalise your inputs. Each input should be a fixed length vector, and each element of the vector should have mean 0, standard deviation 1.0 across the dataset. SKlearn's StandardScaler will do that for you.
Since there doesn't seem to be any demand for this chat room, may I suggest users who come here looking for chat to come to the cross validated room instead?
I posted an answer to that meta question. Not a topic that is going away any time soon, as any "clean" split between sites would be very difficult, especially in machine learning topics. Confused question posters, and different opinions of long-standing community members will keep this running for a while yet.
Anyway, votes on Meta don't count for anything, and everyone was well behaved. Kudos for you in keeping your cool when being criticised so heavily for asking .
This may or may not be currently true. Also, I have not been here very long, I mostly gathered my opinion by reading through a bunch of old meta posts, and getting barked at in the occasional post of my own. Lots of strong feeling on the subject.
Input Data:
$X$ -> features of t shirt (colour,logo,etc)
$Y$ -> profit margin
I have trained a random forest on the above $X$ and $Y$ and have achieved reasonable accuracy on a test data.
So, I have
$P(Y|X)$.
Now, I would like to find $P(X|Y)$ i.e probability distribution of $X$ features giv...
@Cowthulhu: Good evening in fact. It's not so busy here. Depends on details, but questions about details of small code segments in TensorFlow should be welcome here and SO. Longer questions about making your TensorFlow NN match some standard feature - such as dropout, batch normalisation etc, or why your NN implementation runs without error, but has performance issues - are IMO more suitable on Data Science where we can mix and match the code and theory more freely.
As @NeilSlater mentioned, the answer has to include the practical aspects of Data Science. My take on this would be that questions are particularly well suited for Data Science SE if they help you iterate (faster/more informed) through the data science project lifecycle or help you in some way to 'apply' the scientific method. That would include the above debugging questions of NN, data preparation, selecting the right tools for the job, understanding algorithm complexity and so on.
We probably need a more positive spin on what the site is actively for, as opposed to "we'll handle questions on topics that SO/CV cover, but that have the wrong angle for them"
On Stack Overflow, the debugging work requires too much NN theory. On stats.stackexchange the mood seems to switch between these being acceptable or off-topic depending on who is answering.
Input Data:
$X$ -> features of t shirt (colour,logo,etc)
$Y$ -> profit margin
I have trained a random forest on the above $X$ and $Y$ and have achieved reasonable accuracy on a test data.
So, I have
$P(Y|X)$.
Now, I would like to find $P(X|Y)$ i.e probability distribution of $X$ features giv...