From b5ca2bd7001a385594af8dc4b9206399c679f8c5 Mon Sep 17 00:00:00 2001
From: Chris Dyer <cdyer@cs.cmu.edu>
Date: Wed, 22 Dec 2010 08:58:07 -0600
Subject: remove report

---
 report/vest_mira_hope_direction/README             | 23 -------------
 .../vest_mira_hope_direction/dev.bleu.baseline.txt | 23 -------------
 .../dev.bleu.oracle_mira_direction.txt             | 39 ----------------------
 3 files changed, 85 deletions(-)
 delete mode 100644 report/vest_mira_hope_direction/README
 delete mode 100644 report/vest_mira_hope_direction/dev.bleu.baseline.txt
 delete mode 100644 report/vest_mira_hope_direction/dev.bleu.oracle_mira_direction.txt

(limited to 'report/vest_mira_hope_direction')

diff --git a/report/vest_mira_hope_direction/README b/report/vest_mira_hope_direction/README
deleted file mode 100644
index 44dccc6f..00000000
--- a/report/vest_mira_hope_direction/README
+++ /dev/null
@@ -1,23 +0,0 @@
-First column = tune iteration, second column = tune BLEU
-
-Choosing the average of a few random sentence's model->hope direction, With Vlad's default --bleu_weight=1, nothing improved, and convergenced seemed to be slower, which is the opposite of what we intended.  5 random directions and the orthogonal directions were still used (fewer than normal random directions; the oracle directions replaced some of them).
-
-dev.bleu.baseline.txt - usual search directions (orth+random)
-
-dev.bleu.oracle_mira_direction.txt - as above, but fewer random directions, 10 model->hope directions
-
-note that the search directions are recomputed after each MERT iteration.
-
-/home/jgraehl/tune/urdu/vest-dth-oracle*
-
-~/tune/urdu$ ls -d vest-dth*
-vest-dth  vest-dth-oracle  vest-dth-oracle-bleuwt=0.1  vest-dth-oracle-bleuwt=10  vest-dth-oracle-bleuwt=100
-
-oraclewt devtest-estimated-BLEU
-0   22.01 (orth. dirs, 10 random dirs)
-0.1 22.13 (orth. dirs, 10+10 model->hope and fear->hope random directions, 5 purely random dirs)
-1   22.06 "
-10  21.85 "
-100 22.08 "
-
-no evidence of any advantage using oracle directions (TODO: decode mt09 as test set, check for any advantage when using many features)
diff --git a/report/vest_mira_hope_direction/dev.bleu.baseline.txt b/report/vest_mira_hope_direction/dev.bleu.baseline.txt
deleted file mode 100644
index 2cf5a023..00000000
--- a/report/vest_mira_hope_direction/dev.bleu.baseline.txt
+++ /dev/null
@@ -1,23 +0,0 @@
-1	0.18892
-2	0.192939
-3	0.193087
-4	0.19316
-5	0.211582
-6	0.213164
-7	0.214736
-8	0.2164
-9	0.217853
-10	0.21972
-11	0.221133
-12	0.221733
-13	0.222067
-14	0.222347
-15	0.220126
-16	0.220623
-17	0.220766
-18	0.220915
-19	0.220979
-20	0.22119
-21	0.221739
-22	0.22174
-23	0.221802
diff --git a/report/vest_mira_hope_direction/dev.bleu.oracle_mira_direction.txt b/report/vest_mira_hope_direction/dev.bleu.oracle_mira_direction.txt
deleted file mode 100644
index f6cd4972..00000000
--- a/report/vest_mira_hope_direction/dev.bleu.oracle_mira_direction.txt
+++ /dev/null
@@ -1,39 +0,0 @@
-1	0.189294
-2	0.193352
-3	0.194648
-4	0.194931
-5	0.195066
-6	0.21277
-7	0.214142
-8	0.214968
-9	0.215563
-10	0.215798
-11	0.216671
-12	0.217189
-13	0.217404
-14	0.21743
-15	0.217222
-16	0.217487
-17	0.217598
-18	0.217633
-19	0.216885
-20	0.217254
-21	0.218321
-22	0.219699
-23	0.2204
-24	0.220668
-25	0.220817
-26	0.220831
-27	0.220091
-28	0.220761
-29	0.220872
-30	0.220923
-31	0.220602
-32	0.220661
-33	0.221046
-34	0.221234
-35	0.221252
-36	0.220876
-37	0.220985
-38	0.220989
-39	0.221083
-- 
cgit v1.2.3