https://burakhimmetoglu.com/blogs/2020-12-06T16:13:24+00:00weekly0.6https://burakhimmetoglu.com/deterministic-randomness/https://burakhimmetoglu.com/wp-content/uploads/2016/11/brain-2062057_1920.jpgbrain-2062057_1920https://burakhimmetoglu.com/wp-content/uploads/2016/11/abacus-358569_1920.jpgabacus-358569_1920https://burakhimmetoglu.com/wp-content/uploads/2016/11/meandlev.jpegmeandlev2020-03-19T06:25:03+00:00weekly0.6https://burakhimmetoglu.com/2016/12/01/stacking-models-for-improved-predictions/https://burakhimmetoglu.com/wp-content/uploads/2016/12/table.pngtablehttps://burakhimmetoglu.com/wp-content/uploads/2016/12/stackplot.pngstackplothttps://burakhimmetoglu.com/wp-content/uploads/2016/12/models.pngmodelshttps://burakhimmetoglu.com/wp-content/uploads/2016/12/workflow.pngworkflowhttps://burakhimmetoglu.com/wp-content/uploads/2016/12/stacking.pngstacking2019-01-07T17:23:12+00:00monthlyhttps://burakhimmetoglu.com/2019/01/07/pancake/https://burakhimmetoglu.com/wp-content/uploads/2018/12/pancake-640869_1280-e1546039519258.jpgpancake-640869_12802019-01-07T07:09:26+00:00monthlyhttps://burakhimmetoglu.com/2018/09/19/an-overview-of-feature-selection-strategies/https://burakhimmetoglu.com/wp-content/uploads/2018/09/lsvt_table2.pnglsvt_tableResults for LSVT datahttps://burakhimmetoglu.com/wp-content/uploads/2018/09/small_table1.pngsmall_tableComparison of performance metric for the two synthetic datasets.https://burakhimmetoglu.com/wp-content/uploads/2018/09/feats_comparison-e1536552812596.pngfeats_comparisonFeature weights (rescaled) from each selection algorithmhttps://burakhimmetoglu.com/wp-content/uploads/2018/09/overlap.pngoverlapThe red vertical dotted line is the weighted average of the means of + and - classes. Higher overlap corresponds to a case when the + and - distributions are close to each other.https://burakhimmetoglu.com/wp-content/uploads/2018/09/lsvt_feats1.pngLSVT_featsP-value and overlap for each feature in the LSVT datasethttps://burakhimmetoglu.com/wp-content/uploads/2018/09/fs_algo-e1536443490231.jpegFS_algoPseudo-code forward feature selection algorithm.https://burakhimmetoglu.com/wp-content/uploads/2018/09/ffs1.pngFFSFig.2 Forward feature selection procedure. The desired number of features is obtained when CV score is maximized.https://burakhimmetoglu.com/wp-content/uploads/2018/09/lasso_and_ridge-e1536377993337.jpeglasso_and_ridgeFig.1 LASSO (left) and Ridge (right) regression feature weight shrinkage.https://burakhimmetoglu.com/wp-content/uploads/2018/09/volcano_dressed-e1536296426709.jpegvolcano_dressedhttps://burakhimmetoglu.com/wp-content/uploads/2018/09/volcano_dressed-e1536296303133.pngvolcano_dressed2019-06-05T17:27:41+00:00monthlyhttps://burakhimmetoglu.com/2017/08/22/time-series-classification-with-tensorflow/https://burakhimmetoglu.com/wp-content/uploads/2017/08/har_lstm.pngHAR_lstmhttps://burakhimmetoglu.com/wp-content/uploads/2017/08/training.pngtraininghttps://burakhimmetoglu.com/wp-content/uploads/2017/08/har_cnn.pngHAR_cnnhttps://burakhimmetoglu.com/wp-content/uploads/2017/08/ekg-158177_640.pngekg-158177_6402019-07-30T16:47:15+00:00monthlyhttps://burakhimmetoglu.com/portfolio/machine-learning-in-medicine/https://burakhimmetoglu.com/wp-content/uploads/2017/08/pulse-trace-163708.jpgpulse-trace-1637082017-09-18T08:12:41+00:00monthlyhttps://burakhimmetoglu.com/portfolio/robobohr-a-machine-learning-tool-for-molecular-discovery/https://burakhimmetoglu.com/wp-content/uploads/2017/08/molecules-1818492.jpgmolecules-18184922017-08-17T06:23:35+00:00monthlyhttps://burakhimmetoglu.com/2017/08/14/machine-learning-for-alchemy/https://burakhimmetoglu.com/wp-content/uploads/2017/08/hand-898232_1920.jpghand-898232_19202017-08-14T16:10:23+00:00monthlyhttps://burakhimmetoglu.com/2017/02/07/yet-another-introduction-to-neural-networks/https://burakhimmetoglu.com/wp-content/uploads/2017/02/neuron.pngneuron2017-04-25T21:46:55+00:00monthlyhttps://burakhimmetoglu.com/2017/04/07/feature-engineering-with-tidyverse/https://burakhimmetoglu.com/wp-content/uploads/2017/04/rac.pngRachttps://burakhimmetoglu.com/wp-content/uploads/2017/04/importance.pngimportancehttps://burakhimmetoglu.com/wp-content/uploads/2017/04/sf_crimes.jpegsf_crimes2017-04-08T02:02:46+00:00monthlyhttps://burakhimmetoglu.com/2016/12/16/deciphering-the-neural-language-model/https://burakhimmetoglu.com/wp-content/uploads/2016/12/negloglikelihood1.pngnegloglikelihoodhttps://burakhimmetoglu.com/wp-content/uploads/2016/12/probs.pngprobshttps://burakhimmetoglu.com/wp-content/uploads/2016/12/embed.pngembedhttps://burakhimmetoglu.com/wp-content/uploads/2016/12/input_batch.pnginput_batchhttps://burakhimmetoglu.com/wp-content/uploads/2016/12/c.pngchttps://burakhimmetoglu.com/wp-content/uploads/2016/12/a1.pnga1https://burakhimmetoglu.com/wp-content/uploads/2016/12/condprob.pngcondprobhttps://burakhimmetoglu.com/wp-content/uploads/2016/12/langnet.pnglangnet2017-02-07T23:29:43+00:00monthlyhttps://burakhimmetoglu.com/list-of-publications/2016-11-28T20:22:22+00:00weekly0.6https://burakhimmetoglu.com/machine-learning-meets-quantum-mechanics/https://burakhimmetoglu.com/wp-content/uploads/2016/11/rbpca.pngrbpca2016-11-22T20:33:16+00:00weekly0.6https://burakhimmetoglu.comdaily1.02020-12-06T16:13:24+00:00