Timeline for Updated: Correct Formatting of Dataset (labeled columns and rows) for Predict v NetTrain?
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 30, 2016 at 17:50 | history | bounty ended | SumNeuron | ||
| Sep 30, 2016 at 17:50 | vote | accept | SumNeuron | ||
| Sep 30, 2016 at 17:50 | |||||
| Sep 28, 2016 at 19:06 | comment | added | Taliesin Beynon |
@SumNeuron so is there anything remaining that prevents you marking this question as answered? I've checked that your example works with rf = Predict[Values[ds] -> "e", Method -> "RandomForest"]
|
|
| Sep 27, 2016 at 14:13 | comment | added | Taliesin Beynon |
@SumNeuron that's fine, but the labeled rows are irrelevant to Classify and Predict, so they must be dropped. Yes ds[Values] is equivalent to Values[ds]. I've verified that both work on the dataset you give in your actual question.
|
|
| Sep 27, 2016 at 3:10 | comment | added | SumNeuron | I have noticed that and hence have tired using ds[Values] (which is equivalent to Values[ds]?), which yielded similar errors. As to why one may wish to have labeled rows: depending on what the data is, each record may be of importance to identify later. | |
| Sep 26, 2016 at 21:14 | comment | added | Taliesin Beynon | @SumNeuron I have updated the answer. | |
| Sep 26, 2016 at 21:13 | history | edited | Taliesin Beynon | CC BY-SA 3.0 |
added 203 characters in body
|
| Sep 23, 2016 at 10:45 | comment | added | SumNeuron | here is a follow-up edit to further my question | |
| Sep 2, 2016 at 10:59 | comment | added | Taliesin Beynon |
@SumNeuron because Predict doesn't know which column is the output, whereas with networks there are named ports, and "Output" is the default name for a single layer or NetChain's output port, so that's why NetTrain knows that "Output" is the output (but there could be others depending on the network). It's trivial to make Predict use "Output" as the output by just writing Predict[dataset -> "Output", ...].
|
|
| Sep 2, 2016 at 4:04 | comment | added | SumNeuron | that does clarify my quesiton thank you. I guess this comes back to part two as to why Predict would not work with the Input -> #, Output -> # format if the classes were encoded as numbers. | |
| Sep 1, 2016 at 22:25 | history | answered | Taliesin Beynon | CC BY-SA 3.0 |