MultiClass Classification using Dense Layers

[jeremy (Jeremy Howard)]

Have you watched the relevant lessons? Have a look at lessons 4-6 and look at the notebooks and spreadsheets, and then let us know any specific questions you have based on that material.

[tursun]

@jeremy AND janardhanp22,
You have had very interesting conversation.

"I experimented normal machine learning tasks like random forest and getting f1 score (Train)= 0.8602 and fi (Test) = 8138 for the same dataset."
well, with Fandom Forest, I get F1=0.967 for my data (another dataset similar to tursun_deep_p6.csv).I used RF in Orange.
For my data @janardhanp22 said “since it sounds like you have structured data. In general, deep learning isn’t the best tool for structured data”

Did I find contadict here:
place one:
here, @Jeremy said “do you have structured data. In general, deep learning isn’t the best tool for structured data”

place two:
@sibnick in this webpage:

Multi class classification problem

I suppose your dataset is to small for effective using with CNN, RNN. Also it is not clear nature of parameters. CNN, RNN can be effectively used when parameters have structure (e.g. time series, geometric distribution like pixels in image)

said this to my dataset:
“CNN, RNN can be effectively used when parameters have structure (e.g. time series, geometric distribution like pixels in image)”

my data “tursun_deep_p6.csv” in this webpage:

Multi class classification problem

@jeremy, I am Tursun, a student in Malaysia. I have seen your posts in another thread. But I think this thread is right place. I have this multi-class classification problem. my data : tursun.csv , available to download from here : size of my data set : 512*16, last column is 21 classes, they are digits 1-21 (so maybe OneHotEncoding is not needed) note: number of samples (rows in my data) for each class is different. mostly 20 rows, but sometimes 17 or 31 rows my network has: first l…

[sibnick (Nikolay Tolstokulakov)]

I believe Jeremy means structured data like user profile or something like this. CNN, RNN work with correlated data - e.g. pixel’s are correlated with neighbor pixels on real images.

[pietz]

Switch Softmax to Sigmoid

[rforgione (Rob Forgione)]

@jeremy (or anyone else familiar with the topic) – why does DL seem to struggle with structured data compared to other methods (like rf’s)? is there any interesting published research on the topic? I work on the analytics team for an ecommerce company, so I can think of many interesting problems that use structured data. I’d love to find a way to get strong performance using neural networks to minimize the “guesswork” around feature engineering.

[sibnick (Nikolay Tolstokulakov)]

From my point of view it is nature of neural networks. They are universal approximators for continuous functions. But if you have no continuous function for initial model or for loss function then NN are not strong.