147 Commits

Author SHA1 Message Date
tqchen
a569bf2698 change gitignore 2014-12-06 09:19:08 -08:00
tqchen
dc12958fc7 rename master to tracker, to emphasie rabit is p2p in computing 2014-12-06 09:15:31 -08:00
nachocano
67b68ceae6 adding timing 2014-12-05 16:00:47 -08:00
nachocano
54eb5623cb worked on my machine !!! finally 2014-12-05 15:24:00 -08:00
nachocano
d9c22e54de closer, but still does not work... stays in map 100%. I think an exception is being thrown 2014-12-05 13:28:42 -08:00
tqchen
7765e2dc55 add status report 2014-12-05 09:49:26 -08:00
tqchen
ab278513ab ok 2014-12-05 09:39:51 -08:00
Tianqi Chen
e7a22792ac Update submit_job_hadoop.py 2014-12-05 09:14:44 -08:00
Tianqi Chen
e05098cacb Update submit_job_hadoop.py 2014-12-05 09:10:26 -08:00
Tianqi Chen
f9e95ab522 Update submit_job_hadoop.py 2014-12-05 09:09:20 -08:00
nachocano
bb7d6814a7 creating initial version of hadoop submit script. Not working.
Not sure how to get the master uri and port. I believe I cannot do it before I launch the job.

Updating the name from submit_job to submit_job_mpi
2014-12-05 03:27:02 -08:00
nachocano
e00fb99e7b cosmetic 2014-12-04 19:02:11 -08:00
nachocano
e9a3f5169e cosmetic changes 2014-12-04 18:02:07 -08:00
tqchen
1af3e81ada chg robust to reliable 2014-12-04 17:32:22 -08:00
tqchen
7cd5474f1a chg interface 2014-12-04 17:31:40 -08:00
tqchen
821eb21ae2 before make rabit public 2014-12-04 17:30:58 -08:00
tqchen
cc410b8c90 add local model in checkpoint interface, a new goal 2014-12-04 11:09:15 -08:00
tqchen
79e7862583 change note 2014-12-04 09:09:56 -08:00
tqchen
f9d634ce06 change notes 2014-12-04 09:09:29 -08:00
tqchen
65a1cdf8e5 remove doc from main repo 2014-12-04 09:07:36 -08:00
tqchen
67229fd7a9 change model 2014-12-04 09:05:48 -08:00
tqchen
3033177e9e ok 2014-12-03 22:36:16 -08:00
tqchen
656a8fa3a2 ok 2014-12-03 22:32:30 -08:00
tqchen
0e9b64649a ok 2014-12-03 22:30:23 -08:00
tqchen
9da3c6c573 Merge branch 'master' of ssh://github.com/tqchen/rabit 2014-12-03 22:28:59 -08:00
tqchen
09a1305628 chg readme 2014-12-03 22:27:52 -08:00
nachocano
7d314fef78 open for writing 2014-12-03 21:58:58 -08:00
nachocano
dece767084 Revert "open for writing"
This reverts commit 63bf9c799506a27b40125a68a763622395516432.
2014-12-03 21:58:33 -08:00
nachocano
63bf9c7995 open for writing 2014-12-03 21:58:17 -08:00
tqchen
1c76483b4b ok 2014-12-03 21:53:34 -08:00
tqchen
9abe6ad4d8 checkin makefile 2014-12-03 21:30:11 -08:00
tqchen
8175df1002 bug fix in kmeans 2014-12-03 20:05:16 -08:00
tqchen
a1a1a8895e add kmeans 2014-12-03 18:23:58 -08:00
tqchen
69af79d45d sparse kmeans 2014-12-03 18:15:28 -08:00
nachocano
e3a95b2d1a Merge branch 'master' of https://github.com/tqchen/allreduce 2014-12-03 15:39:05 -08:00
nachocano
5c23b94069 updating kmeans based on Tianqi feedback. More efficient now 2014-12-03 15:38:58 -08:00
tqchen
85bb6cd027 Merge branch 'master' of ssh://github.com/tqchen/rabit 2014-12-03 15:13:09 -08:00
tqchen
90b9f1a98a add keepalive script 2014-12-03 15:04:30 -08:00
nachocano
55c2a5dc83 Merge branch 'master' of https://github.com/tqchen/allreduce 2014-12-03 14:21:42 -08:00
nachocano
1d0d5bb141 kmeans seems to be working.. not restarting anything though 2014-12-03 14:21:10 -08:00
tqchen
7a983a4079 add keepalive 2014-12-03 13:21:30 -08:00
tqchen
2523288509 basic recovery works 2014-12-03 12:19:08 -08:00
tqchen
8a6768763d bug fixed ver 2014-12-03 11:51:39 -08:00
tqchen
a186f8c3aa ok 2014-12-03 11:19:43 -08:00
tqchen
ceeb6f0690 bug version, check in and rollback 2014-12-03 11:17:39 -08:00
tqchen
f3e5b6e13c ok 2014-12-03 10:00:47 -08:00
tqchen
34f2f887b1 add more broadcast and basic broadcast 2014-12-03 09:59:13 -08:00
nachocano
20b51cc9ce cleaner 2014-12-03 01:44:34 -08:00
nachocano
56aad86231 adding incomplete kmeans.
I'm having a problem with the broadcast, and still need to implement the logic
2014-12-03 01:16:13 -08:00
tqchen
ed1de6df80 change AllReduce to Allreduce 2014-12-02 21:11:48 -08:00