Timeline for What is the name of this chart showing false and true positive rates and how is it generated?
Current License: CC BY-SA 3.0
18 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 21, 2016 at 18:48 | history | edited | amoeba | CC BY-SA 3.0 |
removed second title
|
| Jan 21, 2016 at 18:22 | comment | added | Brandon Amos | @OrangeDog 0.9977 ± 0.0006 is Baidu's accuracy from the LFW results page, not AUC. This is confusing because the LFW results page doesn't have a heading for the column this is from. However, their v4 arxiv paper presents this number as the accuracy. I have computed the AUC on their curve in this function. Even though the AUC of 1.000 is confusing, I believe my technique is valid. | |
| Jan 21, 2016 at 16:23 | comment | added | Axoren | @OrangeDog So I did, it seems. I was so overtaken by my chronic failure to identify this brand of curve, having known the name before, that I stopped reading as soon as I saw it and remembered my frustration. | |
| Jan 21, 2016 at 2:12 | history | edited | Jeromy Anglim | CC BY-SA 3.0 |
edited title
|
| Jan 20, 2016 at 14:04 | comment | added | OrangeDog | Incidentally, you missed that your source had the answer: "See Wikipedia for more details about reading the ROC curve." | |
| Jan 20, 2016 at 14:04 | comment | added | OrangeDog | OK, they mixed results of different experiments, and rounded their source data incorrectly. Baidu should be 0.9977 ± 0.0006 | |
| Jan 20, 2016 at 2:53 | answer | added | polarise | timeline score: 1 | |
| S Jan 20, 2016 at 1:39 | history | suggested | Amir |
Added a relevant tag
|
|
| Jan 20, 2016 at 0:46 | review | Suggested edits | |||
| S Jan 20, 2016 at 1:39 | |||||
| Jan 19, 2016 at 17:29 | comment | added | Axoren | @OrangeDog bamos.github.io/2016/01/19/openface-0.2.0 | |
| Jan 19, 2016 at 17:23 | history | tweeted | twitter.com/StackStats/status/689498441597071360 | ||
| Jan 19, 2016 at 16:48 | comment | added | OrangeDog | I'd be interested as to where this came from. It appears to be claiming that Baidu is 100% perfect (and better than people) at identifying/recognising faces. Either that or it's using Baidu's results as ground truth rather than human categorisation, which is also really weird. | |
| Jan 19, 2016 at 16:11 | answer | added | Franck Dernoncourt | timeline score: 21 | |
| Jan 19, 2016 at 15:35 | answer | added | Sycorax♦ | timeline score: 9 | |
| Jan 19, 2016 at 15:14 | history | edited | Tim |
edited tags
|
|
| Jan 19, 2016 at 15:10 | vote | accept | Axoren | ||
| Jan 19, 2016 at 14:35 | answer | added | Morten | timeline score: 28 | |
| Jan 19, 2016 at 14:25 | history | asked | Axoren | CC BY-SA 3.0 |