2013-05-28 07:52:46 -0600 | commented question | adaboost features selection from large pool i want create a ranking of most important features of pedestrian , the pool is very large because i use a features simil to haar, and i can generate a large pool of features, the lenght of vector features descriptor is 10^6 byte |
2013-05-28 04:20:22 -0600 | received badge | ● Student (source) |
2013-05-28 03:43:01 -0600 | received badge | ● Editor (source) |
2013-05-28 03:41:59 -0600 | asked a question | adaboost features selection from large pool Hi, I want to train adaboost with 10^9 features but it can't stay all to gheter in memory , i procedeed with this approch:
for(i=1 to 1000)
{
-load 10^6 features from the 10^9 pool
-select the best 1 features and store it i think the first approches is more exaustive then the second but the second is better on validation set, someone can tell me anything about this anomaly? |