You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hivemall.apache.org by my...@apache.org on 2019/12/25 09:30:38 UTC
[incubator-hivemall] branch master updated: Fixed docs for UDF
preparation
This is an automated email from the ASF dual-hosted git repository.
myui pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-hivemall.git
The following commit(s) were added to refs/heads/master by this push:
new 2d9050e Fixed docs for UDF preparation
2d9050e is described below
commit 2d9050e9b17c54b39e9a99b222f0d9f1beb44ea3
Author: Makoto Yui <my...@apache.org>
AuthorDate: Wed Dec 25 18:30:06 2019 +0900
Fixed docs for UDF preparation
---
docs/gitbook/binaryclass/kdd2010a_dataset.md | 3 ---
docs/gitbook/binaryclass/kdd2010b_dataset.md | 3 ---
docs/gitbook/binaryclass/webspam_dataset.md | 6 +----
docs/gitbook/binaryclass/webspam_scw.md | 12 ++-------
docs/gitbook/multiclass/news20_ensemble.md | 30 +++-------------------
docs/gitbook/multiclass/news20_one-vs-the-rest.md | 18 +------------
.../multiclass/news20_one-vs-the-rest_dataset.md | 8 ------
docs/gitbook/recommend/news20_jaccard.md | 10 +-------
docs/gitbook/tips/ensemble_learning.md | 22 ----------------
9 files changed, 8 insertions(+), 104 deletions(-)
diff --git a/docs/gitbook/binaryclass/kdd2010a_dataset.md b/docs/gitbook/binaryclass/kdd2010a_dataset.md
index b5dcadf..46939d0 100644
--- a/docs/gitbook/binaryclass/kdd2010a_dataset.md
+++ b/docs/gitbook/binaryclass/kdd2010a_dataset.md
@@ -26,9 +26,6 @@
---
# Define training/testing tables
```sql
-add jar ./tmp/hivemall.jar;
-source ./tmp/define-all.hive;
-
create database kdd2010;
use kdd2010;
diff --git a/docs/gitbook/binaryclass/kdd2010b_dataset.md b/docs/gitbook/binaryclass/kdd2010b_dataset.md
index 6ad71af..7313b13 100644
--- a/docs/gitbook/binaryclass/kdd2010b_dataset.md
+++ b/docs/gitbook/binaryclass/kdd2010b_dataset.md
@@ -26,9 +26,6 @@
---
# Define training/testing tables
```sql
-add jar ./tmp/hivemall.jar;
-source ./tmp/define-all.hive;
-
create database kdd2010;
use kdd2010;
diff --git a/docs/gitbook/binaryclass/webspam_dataset.md b/docs/gitbook/binaryclass/webspam_dataset.md
index a0777c7..34ccfba 100644
--- a/docs/gitbook/binaryclass/webspam_dataset.md
+++ b/docs/gitbook/binaryclass/webspam_dataset.md
@@ -16,7 +16,7 @@
specific language governing permissions and limitations
under the License.
-->
-
+
Get the dataset from
https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#webspam
@@ -33,10 +33,6 @@ hadoop fs -put - /dataset/webspam/raw/
create database webspam;
use webspam;
-delete jar ./tmp/hivemall.jar;
-add jar ./tmp/hivemall.jar;
-source ./tmp/define-all.hive;
-
create external table webspam_raw (
rowid int,
label int,
diff --git a/docs/gitbook/binaryclass/webspam_scw.md b/docs/gitbook/binaryclass/webspam_scw.md
index 067e8f2..bd7f4c5 100644
--- a/docs/gitbook/binaryclass/webspam_scw.md
+++ b/docs/gitbook/binaryclass/webspam_scw.md
@@ -16,20 +16,12 @@
specific language governing permissions and limitations
under the License.
-->
-
-# Preparation
-
-```
-use webspam;
-
-delete jar ./tmp/hivemall.jar;
-add jar ./tmp/hivemall.jar;
-source ./tmp/define-all.hive;
-```
# PA1
```sql
+use webspam;
+
drop table webspam_pa1_model1;
create table webspam_pa1_model1 as
select
diff --git a/docs/gitbook/multiclass/news20_ensemble.md b/docs/gitbook/multiclass/news20_ensemble.md
index 7389a47..200fd68 100644
--- a/docs/gitbook/multiclass/news20_ensemble.md
+++ b/docs/gitbook/multiclass/news20_ensemble.md
@@ -16,22 +16,11 @@
specific language governing permissions and limitations
under the License.
-->
-
+
This example explains how to run ensemble learning in Hivemall.
Two heads are better than one? Let's verify it by ensemble learning.
----
-
-## UDF preparation
-```sql
-delete jar /home/myui/tmp/hivemall.jar;
-add jar /home/myui/tmp/hivemall.jar;
-
-source /home/myui/tmp/define-all.hive;
-```
-
-[Case1] Model ensemble/mixing
-=======================
+# [Case1] Model ensemble/mixing
## training
```sql
@@ -117,15 +106,6 @@ where actual == predicted;
> 0.8494866015527173
-## Cleaning
-
-```sql
-drop table news20mc_ensemble_model1;
-drop view news20mc_ensemble_predict1;
-drop view news20mc_ensemble_submit1;
-```
----
-
Unfortunately, too many cooks spoil the broth in this case :-(
| Algorithm | Accuracy |
@@ -135,11 +115,7 @@ Unfortunately, too many cooks spoil the broth in this case :-(
| Ensemble(model) | 0.8494866015527173 |
| CW | 0.850488354620586 |
-
----
-
-[Case2] Prediction ensemble
-=================
+# [Case2] Prediction ensemble
## prediction
```sql
diff --git a/docs/gitbook/multiclass/news20_one-vs-the-rest.md b/docs/gitbook/multiclass/news20_one-vs-the-rest.md
index d98329f..75640e7 100644
--- a/docs/gitbook/multiclass/news20_one-vs-the-rest.md
+++ b/docs/gitbook/multiclass/news20_one-vs-the-rest.md
@@ -16,16 +16,8 @@
specific language governing permissions and limitations
under the License.
-->
-
-A one-vs-the-rest classifier use the binary classifier for each class.
-
-## UDF preparation
-```sql
-delete jar /home/myui/tmp/hivemall.jar;
-add jar /home/myui/tmp/hivemall.jar;
-source /home/myui/tmp/define-all.hive;
-```
+A one-vs-the-rest classifier use the binary classifier for each class.
## training
```sql
@@ -334,14 +326,6 @@ where actual == predicted;
> 0.8567493112947658
-## Cleaning
-
-```sql
-drop table news20_onevsrest_arow_model1;
-drop view news20_onevsrest_arow_predict1;
-drop view news20_onevsrest_arow_submit1;
-```
-
| Algorithm | Accuracy |
|:-----------|------------:|
| AROW(multi-class) | 0.8474830954169797 |
diff --git a/docs/gitbook/multiclass/news20_one-vs-the-rest_dataset.md b/docs/gitbook/multiclass/news20_one-vs-the-rest_dataset.md
index 62bc397..639dcf3 100644
--- a/docs/gitbook/multiclass/news20_one-vs-the-rest_dataset.md
+++ b/docs/gitbook/multiclass/news20_one-vs-the-rest_dataset.md
@@ -20,14 +20,6 @@
*One-vs-the-rest* is a multiclass classification method that uses binary classifiers independently for each class.
http://en.wikipedia.org/wiki/Multiclass_classification#one_vs_all
-## UDF preparation
-```sql
-delete jar /home/myui/tmp/hivemall.jar;
-add jar /home/myui/tmp/hivemall.jar;
-
-source /home/myui/tmp/define-all.hive;
-```
-
## Dataset preparation for one-vs-the-rest classifiers
```sql
diff --git a/docs/gitbook/recommend/news20_jaccard.md b/docs/gitbook/recommend/news20_jaccard.md
index 0166ed5..ad80b6e 100644
--- a/docs/gitbook/recommend/news20_jaccard.md
+++ b/docs/gitbook/recommend/news20_jaccard.md
@@ -19,18 +19,10 @@
List related (similar) articles for each article.
-# Preparation
+# Extract clusters
```sql
use news20;
-delete jar /home/myui/tmp/hivemall.jar;
-add jar /home/myui/tmp/hivemall.jar;
-
-source /home/myui/tmp/define-all.hive;
-```
-
-# Extract clusters
-```sql
set hivevar:hashes=100; -- Generate N sets of minhash values for each row (DEFAULT: 5)
set hivevar:keygroups=2; -- Use K minhash value for generating a resulting value (DEFAULT: 2)
diff --git a/docs/gitbook/tips/ensemble_learning.md b/docs/gitbook/tips/ensemble_learning.md
index 2157a5b..54503b9 100644
--- a/docs/gitbook/tips/ensemble_learning.md
+++ b/docs/gitbook/tips/ensemble_learning.md
@@ -22,16 +22,6 @@ Two heads are better than one? Let's verify it by ensemble learning.
<!-- toc -->
----
-
-## UDF preparation
-```sql
-delete jar /home/myui/tmp/hivemall.jar;
-add jar /home/myui/tmp/hivemall.jar;
-
-source /home/myui/tmp/define-all.hive;
-```
-
# [Case1] Model ensemble/mixing
## training
@@ -118,15 +108,6 @@ where actual == predicted;
> 0.8494866015527173
-## Cleaning
-
-```sql
-drop table news20mc_ensemble_model1;
-drop view news20mc_ensemble_predict1;
-drop view news20mc_ensemble_submit1;
-```
----
-
Unfortunately, too many cooks spoil the broth in this case :-(
| Algorithm | Accuracy |
@@ -136,9 +117,6 @@ Unfortunately, too many cooks spoil the broth in this case :-(
| Ensemble(model) | 0.8494866015527173 |
| CW | 0.850488354620586 |
-
----
-
# [Case2] Prediction ensemble
## prediction