GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/7
Matrix: Fix multiple input/output issues
JIRA: MADLIB-932
This commit contains fixes for following issues:
- 'inf' or 'infinity' was read by Python as a float value. This has
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/4#discussion_r45804084
--- Diff: src/ports/postgres/modules/svm/svm.py_in ---
@@ -66,211 +253,118 @@ def svm(schema_madlib, source_table, model_table
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/4#discussion_r45806336
--- Diff: src/ports/postgres/modules/svm/svm.py_in ---
@@ -440,55 +595,68 @@ def _process_epsilon(is_svc, args):
def _extract_params(schema_madlib
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/4#discussion_r45806649
--- Diff: src/ports/postgres/modules/svm/svm.py_in ---
@@ -440,55 +595,68 @@ def _process_epsilon(is_svc, args):
def _extract_params(schema_madlib
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/4#discussion_r45920065
--- Diff: src/ports/postgres/modules/svm/svm.py_in ---
@@ -389,11 +521,20 @@ def _process_epsilon(is_svc, args):
as_rel_source = '_src
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/4#discussion_r45919749
--- Diff: src/ports/postgres/modules/svm/svm.py_in ---
@@ -66,211 +253,118 @@ def svm(schema_madlib, source_table, model_table
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/42#discussion_r65259215
--- Diff: src/ports/postgres/modules/stats/pred_metrics.py_in ---
@@ -0,0 +1,562 @@
+# coding=utf-8
+#
+# Licensed to the Apache Software
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/42#discussion_r65259392
--- Diff: src/ports/postgres/modules/stats/pred_metrics.py_in ---
@@ -0,0 +1,562 @@
+# coding=utf-8
+#
+# Licensed to the Apache Software
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/42
Along with casting the columns to int in binary classification, we also
need to change docs/online-help/tests to reflect that boolean columns allowed
for observation columns
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/42#issuecomment-23823
I have made some changes and added validation and online help functions (in
my [private fork
branch](https://github.com/iyerr3/incubator-madlib/commits/feature
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/47#discussion_r67946909
--- Diff: src/ports/postgres/modules/utilities/pivot.sql_in ---
@@ -0,0 +1,202
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/49#discussion_r67954685
--- Diff: src/ports/postgres/modules/utilities/sessionize.py_in ---
@@ -35,41 +36,83 @@ def sessionize(schema_madlib, source_table,
output_table
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/47#discussion_r68128267
--- Diff: src/ports/postgres/modules/utilities/pivot.py_in ---
@@ -0,0 +1,201 @@
+# coding=utf-8
+#
+# Licensed to the Apache Software
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/47#discussion_r67259935
--- Diff: doc/mainpage.dox.in ---
@@ -1,12 +1,12 @@
/**
@mainpage
-Apache MADlib (incubating) is an open-source library for scalable
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/23#issuecomment-190345079
Merged (0545cdfc499354820af670b627a2769ae6af75b1)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/22#discussion_r54282752
--- Diff: methods/array_ops/src/pg_gp/array_ops.c ---
@@ -702,8 +702,14 @@ array_sub(PG_FUNCTION_ARGS){
PG_FUNCTION_INFO_V1(array_mult
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/22#discussion_r54290851
--- Diff: methods/array_ops/src/pg_gp/array_ops.c ---
@@ -702,8 +702,14 @@ array_sub(PG_FUNCTION_ARGS){
PG_FUNCTION_INFO_V1(array_mult
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/29#issuecomment-195136869
@fmcquillan99 Please pull this branch to review/update documentation.
@mktal Please give this a review.
---
If your project is set up for it, you can
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/29#issuecomment-198561189
@fmcquillan99 - Added a new commit to ensure that the symbol order is
maintained. The expected behavior is "if a row matches with two symbols, the
s
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/29#discussion_r56901766
--- Diff: src/ports/postgres/modules/utilities/path.py_in ---
@@ -387,11 +395,16 @@ def _parse_symbol_str(symbol_expr, pattern_expr
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/32#discussion_r57494868
--- Diff:
src/ports/postgres/modules/recursive_partitioning/decision_tree.py_in ---
@@ -669,6 +669,8 @@ def _get_n_and_deplist(training_table_name
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/33#discussion_r57504858
--- Diff: src/ports/postgres/modules/summary/summary.sql_in ---
@@ -285,6 +294,16 @@ string should be double-quoted; in this case the input
would
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/33#issuecomment-201615733
LGTM pending some of the spelling corrections
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/33#discussion_r57504745
--- Diff: src/ports/postgres/modules/pca/pca.sql_in ---
@@ -324,13 +341,20 @@ string should be double-quoted; in this case the
input would
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/33#discussion_r57503961
--- Diff: src/ports/postgres/modules/elastic_net/elastic_net.sql_in ---
@@ -130,7 +130,13 @@ the dependent variable expression to the
excluded string
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/29#discussion_r57402068
--- Diff: src/ports/postgres/modules/utilities/path.py_in ---
@@ -108,8 +107,11 @@ def path(schema_madlib, source_table, output_table,
partition_expr
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/29#discussion_r57402003
--- Diff: src/ports/postgres/modules/utilities/path.py_in ---
@@ -118,140 +120,175 @@ def path(schema_madlib, source_table, output_table
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/29#discussion_r57401060
--- Diff: src/ports/postgres/modules/utilities/path.py_in ---
@@ -118,140 +120,175 @@ def path(schema_madlib, source_table, output_table
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/27#issuecomment-193432184
Thanks, Roman. I can take care of the merge.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/43
SVM: Add class weights for use with unbalanced data
JIRA: MADLIB-998
Added 'class_weight' in the 'params' argument. It can either be a string or
a
dictionary-like mapping
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/43#issuecomment-221046547
Testing with scikit learn revealed odd behavior. See figures below.
1) With equal class size the results are similar between scikit and madlib.
2
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/pull/43#issuecomment-221083387
Things are better when we don't use the default step size.
Using init_stepsize=0.1 instead of the default (0.01):
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/76
Merged with 0e00a27. Need to close this PR (either by @lifepillar or via a
future commit message)
---
If your project is set up for it, you can reply to this email and have your
reply
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/89#discussion_r98118503
--- Diff: src/ports/postgres/modules/kmeans/kmeans.py_in ---
@@ -34,6 +38,34 @@ def kmeans_validate_src(schema_madlib, rel_source,
**kwargs
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/103
Looks good to me. I was able to run through the steps. Only comment would
be to update documentation to reflect the `madlib` docker hub account.
---
If your project is set up for it, you
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/106
Build: Fix module sort order for PGXN installation
JIRA: MADLIB-1024
PGXN installation involves creating a single extension sql file that
contains all the SQL commands run
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/101
Multiple: Add casting to allow compilation with GCC 6+
JIRA: MADLIB-1025
GCC 6+ introduced stricter rules for implicit casting where loss of
information is possible
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/102
Build: Add error for missing server includedir
JIRA: MADLIB-1065
Missing server headers is a common error with new contributors. This
leads to a non-related error with CMake
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/105#discussion_r103327658
--- Diff: doc/design/modules/graph.tex ---
@@ -86,3 +131,49 @@ \section{Single Source Shortest Path}
\label{sec:graph:sssp
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/105#discussion_r103325238
--- Diff: doc/design/modules/graph.tex ---
@@ -33,9 +34,53 @@
This module implements various graph algorithms that are used in a number
Github user iyerr3 commented on the pull request:
https://github.com/apache/incubator-madlib/commit/d62bfbe5234d9e0185cd8acda711fe6dabfd7574#commitcomment-19326675
I don't expect any specific performance or usage changes. We should do a
full round of performance tests to confirm
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/70
Apologies for the rebased branch. New changes include:
```
Changes:
- Fix madpack to assume revision string can be any valid Semantic
versioning scheme.
- Change
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/74
Merged with commit 82e56a4
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/70
Need final approval to close this PR.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/70
See recent update to enable versions of the form "1.9.1-rc". Per semantic
versioning, all identifier labels should be a suffix starting with a hyphen.
This is considered a val
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/74
@njayaram2: Could you please close this PR (forgot to close it with the
commit).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/77
Cross validation in Elastic Net
Note: Currently this is rebased on top of the grouping PR (#69) and has the
commits from that PR. When that PR is merged, this should be corrected to only
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/70
Thanks for the comment, @stephendotcarter.
I thought I had changed that to int last week but clearly I didn't push to
remote. Please check the updated code - I've also added some
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/69#discussion_r86223466
--- Diff: src/ports/postgres/modules/utilities/in_mem_group_control.py_in
---
@@ -54,6 +54,18 @@ class BaseState(object
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/70
Thanks! Updated code.
On Sun, Oct 30, 2016 at 9:39 AM, Stephen Carter <notificati...@github.com>
wrote:
> There is another line where we need to convert ba
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/70
Build: Fix version parsing in madpack
Primary change is line 359. Other changes are whitespace fixes.
You can merge this pull request into a Git repository by running:
$ git pull
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/70#discussion_r83716925
--- Diff: src/madpack/madpack.py ---
@@ -357,7 +356,7 @@ def _get_rev_num(rev):
@param rev version text
""&quo
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/82
Encode categorical variables
JIRA: MADLIB-1038
Major overhaul of the dummy/one-hot encoding of categorical variables
with new name and updated arguments. Older function has been
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/77#discussion_r93540854
--- Diff:
src/ports/postgres/modules/validation/internal/cross_validation.py_in ---
@@ -336,22 +339,31 @@ class CrossValidator(object
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/110
retest this please
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/110
retest this please
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled
Github user iyerr3 closed the pull request at:
https://github.com/apache/incubator-madlib/pull/107
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/107
Merged with 7be68936f2cf09e44fd9a8ae3e893db73dc99b26
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/108
Pivot: Add support for array output
JIRA: MADLIB-1066
When total pivoted columns exceed 1600, an array output becomes
essential. This commit adds support to get each pivoted
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/123#discussion_r112568270
--- Diff: licenses/MADlib.txt ---
@@ -1,10 +0,0 @@
-Portions of this software Copyright (c) 2010-2013 by EMC Corporation. All
rights reserved
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/116
Jenkins, OK to test.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/123#discussion_r112544856
--- Diff: licenses/MADlib.txt ---
@@ -1,10 +0,0 @@
-Portions of this software Copyright (c) 2010-2013 by EMC Corporation. All
rights reserved
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/119#discussion_r112535164
--- Diff: src/ports/postgres/modules/graph/sssp.py_in ---
@@ -432,11 +433,17 @@ def graph_sssp(schema_madlib, vertex_table,
vertex_id, edge_table
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/119#discussion_r112534924
--- Diff: src/ports/postgres/modules/graph/sssp.py_in ---
@@ -314,9 +314,13 @@ def graph_sssp(schema_madlib, vertex_table, vertex_id,
edge_table
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/119#discussion_r112535321
--- Diff:
src/ports/postgres/modules/elastic_net/test/elastic_net_install_check.sql_in ---
@@ -840,27 +840,27 @@ SELECT elastic_net_train
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/124
Bugfix/jenkins xml report
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/iyerr3/incubator-madlib
bugfix/jenkins_xml_report
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/124
Note: Only two files have changed with this commit. For some reason
github/master has not updated to upstream (apache).
---
If your project is set up for it, you can reply to this email
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/117
Decision Tree: Update defaults for max_depth, num_splits
Reduce the defaults for max_depth to 7 and num_splits to 20 to ensure we
don't run out of memory when initializing tree
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/124
retest this please.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/107
Build: Avoid downloading mathjax during make doc
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/iyerr3/incubator-madlib infra
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/115
jenkins, ok to test
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/115
+1
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/117
Reading the commit description again, I would rephrase it as
"Reduce the defaults for max_depth to 7 and num_splits to 20 to **minimize
the chances of running out of m
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/116#discussion_r112040453
--- Diff: methods/array_ops/src/pg_gp/array_ops.sql_in ---
@@ -636,3 +663,30 @@ CREATE OR REPLACE FUNCTION
MADLIB_SCHEMA.array_cum_prod(x anyarray
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/117
No, that's a separate JIRA: MADLIB-1057
<https://issues.apache.org/jira/browse/MADLIB-1057>. This one is just about
setting the defaults to a more reasonable value consi
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/119
OK to test
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/120
DT: Assign memory only for reachable nodes
JIRA: MADLIB-1057
TreeAccumulator assigns a matrix to track the statistics of rows
reaching the last layer of nodes. This matrix
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/117
The docs/latest corresponds to latest release (1.10) and won't be updated
till the next release. We also have
[docs/master](http://madlib.incubator.apache.org/docs/master/), which can
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/120
@ivannovick Those are good questions.
Short answer is it's problem (data) dependent - the memory reduction
depends on how sparse the tree is.
I can run some experiments
GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/156
DT: Add option to treat NULL as category
This commit adds an option to treat NULL as a level in the categorical
feature. The level is added as a string (instead of a NULL value
Github user iyerr3 commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/156#discussion_r131048326
--- Diff:
src/ports/postgres/modules/recursive_partitioning/decision_tree.py_in ---
@@ -825,22 +855,34 @@ def _get_bins(schema_madlib
Github user iyerr3 commented on the issue:
https://github.com/apache/incubator-madlib/pull/157
Added more details in the 2nd commit message which will be used in the
final merge.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub
1 - 100 of 147 matches
Mail list logo
