Timeline for How does a simple logistic regression model achieve a 92% classification accuracy on MNIST?
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 15, 2019 at 2:29 | comment | added | chrylis -cautiouslyoptimistic- | For someone who's interested in but not particularly familiar with this sort of processing, this answer provides a fantastic intuitive example of the mechanics. | |
| Sep 13, 2019 at 18:52 | vote | accept | Nitish Agarwal | ||
| Sep 13, 2019 at 16:02 | comment | added | sintax | @NitishAgarwal, If you think that this answer is the Answer to your Question, consider marking it as such. | |
| Sep 13, 2019 at 14:03 | comment | added | Djib2011 | @EricDuminil I added a commend on the script with your suggestion. Thanks a lot for the input! :D | |
| Sep 13, 2019 at 14:02 | history | edited | Djib2011 | CC BY-SA 4.0 |
updated script
|
| Sep 13, 2019 at 13:13 | comment | added | Djib2011 | @EricDuminil you're right, two lines were missing from the script. I added them. | |
| Sep 13, 2019 at 13:13 | history | edited | Djib2011 | CC BY-SA 4.0 |
Added missing lines in code.
|
| Sep 12, 2019 at 18:41 | comment | added | hobbs | Of course it helps that MNIST samples are centered, scaled, and contrast-normalized before the classifier ever sees them. You don't have to address questions like "what if the edge of the zero actually goes through the middle of the box?" because the pre-processor has already gone a long way towards making all zeroes look the same. | |
| Sep 12, 2019 at 0:27 | comment | added | Nitish Agarwal | Thanks for the illustration. These weight images make it more clear as how the accuracy is so high. Dot multiplication of a handwritten digit image with the weight image corresponding to the true label of the image does 'seem' to be the highest in comparison to the dot product with other weight labels for most (still 92% look like a lot to me) of the images in MNIST. Still, it's a little surprising that $2$ and $3$ or $7$ and $8$ are seldom misclassified as each other upon examining the confusion matrix. Anyways, this is what it is. The data never lies. :) | |
| Sep 11, 2019 at 23:23 | history | answered | Djib2011 | CC BY-SA 4.0 |