Timeline for Why is the accuracy of a LinearSVC not the same as the SDGClassifier?
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 22, 2020 at 21:17 | comment | added | Anna | I understand, thank you, I didn't realise it! | |
| Sep 22, 2020 at 20:57 | comment | added | Multivac | The "squared_hinge" is the default on the LinearSVM that's why I'm using it on SGDClassifier | |
| Sep 22, 2020 at 20:56 | comment | added | Multivac | Ok, so I see there is no much more I can do for helping you, but If you want to undestand why the results are so different the answers lies here..."LinearSVM uses the full data and solve a convex optimization problem with respect to these data points. SGDClassifier can treat the data in batches and performs a gradient descent aiming to minimize expected loss with respect to the sample distribution, assuming that the examples are iid samples of that distribution" | |
| Sep 22, 2020 at 12:45 | comment | added | Anna | I also tried increasing the number of iterations, but to no avail | |
| Sep 22, 2020 at 9:51 | vote | accept | Anna | ||
| Sep 22, 2020 at 10:13 | |||||
| Sep 22, 2020 at 8:44 | comment | added | Anna | Thanks, but why the squared hinge in the SDGClassifier? Also, C parameters etc. need to be fine tuned, default ones are of no use | |
| Sep 22, 2020 at 2:14 | history | answered | Multivac | CC BY-SA 4.0 |