The Author Online Book Forums are Moving

The Author Online Book Forums will soon redirect to Manning's liveBook and liveVideo. All book forum content will migrate to liveBook's discussion forum and all video forum content will migrate to liveVideo. Log in to liveBook or liveVideo with your Manning credentials to join the discussion!

Thank you for your engagement in the AoF over the years! We look forward to offering you a more enhanced forum experience.

Nelson Silva (1) [Avatar] Offline
#1
Hello,

I am interested in HMMs and Tensorflow, mainly:

- I usually have inputs like the following data:

timestamp; AreaID; watch_time_in_milliseconds

A typical file would be a sequence of choices, like:
0, 3, 5, 7, ..., 8, 24

I would have the timestamp and the amount of time spent on each choice, e.g., all choices were less than 100ms, except for 5 and 24 (final state) where the user spent more than 200ms

I would like to train my model and be able to predict later on a similar sequence, that the user would jump to 5 (when the state was 3) and that the user will jump to 24 (when the user choice was smilie

I would like to have a probability of the next choices and present the top 3 most probable next choices based on the current state and the times spent on each previous state (maybe).
I've looked into Tensorflow, however, the examples seemed complex and too far away from what I need here in my example.


Can you somehow help by providing a similar type of example?

Thank you so much in advance.

--
best,

Nelson Silva