taxi.git - Kaggle Taxi Destination Prediction Challnge

	Commit message (Collapse)	Author	Age	Files	Lines
*	Remove useless n_valid in configs.	Étienne Simon	2015-05-27	17	-34/+0
\|
*	Add parametrizability for how the training data is presented	Alex Auvolat	2015-05-22	17	-1/+21
\|
*	Fix embeddings sizes.	Étienne Simon	2015-05-22	8	-16/+16
\|
*	Remove useless n_end_pts from configs.	Étienne Simon	2015-05-21	17	-17/+0
\|
*	Fix new models.	Alex Auvolat	2015-05-21	3	-9/+12
\|
*	Merge branch 'alex' into new	Alex Auvolat	2015-05-21	3	-0/+170
\|\ \| \| \| \| \| \| \| \| \| \|	Conflicts: model/joint_simple_mlp_tgtcls.py train.py
\| *	Model changes	Alex Auvolat	2015-05-21	3	-0/+170
\| \|
* \|	Use os.path.join and close files	Étienne Simon	2015-05-18	8	-8/+16
\| \|
* \|	Refactor models, clean the code and separate training from testing.	Étienne Simon	2015-05-18	14	-21/+83
\|/
*	Add model with hidden layers specific to time/dest prediction.	Alex Auvolat	2015-05-08	1	-0/+55
\|
*	Add scaling factor between two costs in joint model	Alex Auvolat	2015-05-08	1	-0/+3
\|
*	Add model for a network that predicts both time and destination.	Alex Auvolat	2015-05-08	1	-0/+52
\|
*	Fusion AddFirstK and AddLastK transformers, and add 'input_time' field.	Alex Auvolat	2015-05-08	1	-2/+0
\|
*	Add target class based model for time prediction (seems to work)	Alex Auvolat	2015-05-07	1	-0/+41
\|
*	Move weights init to config files ; fix s/time/travel_time	Alex Auvolat	2015-05-06	10	-1/+69
\|
*	Adapt Alexandre's model to the new interface	Étienne Simon	2015-05-05	1	-3/+5
\|
*	Merge branch 'master' of github.com:adbrebs/taxi	AdeB	2015-05-05	7	-16/+30
\|\
\| *	Clean data module and generalize use of hdf5.	Étienne Simon	2015-05-05	7	-16/+30
\| \|
* \|	New hyperparameters. Training error is monitored.	AdeB	2015-05-05	1	-0/+30
\|/
*	Add models for time predictioAdd models for time prediction	Alex Auvolat	2015-05-05	9	-7/+52
\|
*	Add day type and taxi id	Alex Auvolat	2015-05-05	3	-2/+58
\|
*	Add date/time transformer and new model that uses it	Alex Auvolat	2015-05-05	5	-16/+28
\|
*	Refactor architecture so that embedding sizes can be easily changed.	Alex Auvolat	2015-05-05	4	-6/+42
\|
*	New config (added a hidden layer), small changes to train.py	Alex Auvolat	2015-05-05	1	-0/+25
\|
*	Fix it	Alex Auvolat	2015-05-04	1	-1/+1
\|
*	Add model with predefined target classes	Alex Auvolat	2015-05-04	1	-0/+25
\|
*	Restructure model & config	Alex Auvolat	2015-05-04	1	-0/+2
\|
*	Mew data analysis tool: clustering of arrival points.	Alex Auvolat	2015-05-04	1	-2/+2
\|
*	Fixes	Alex Auvolat	2015-05-04	1	-1/+1
\|
*	Fix	Alex Auvolat	2015-04-29	1	-1/+1
\|
*	Fix CSV import (partially)	Alex Auvolat	2015-04-29	1	-5/+1
\|
*	Adapt model to hdf5 dataset. WIP	Étienne Simon	2015-04-29	2	-0/+21