-
Notifications
You must be signed in to change notification settings - Fork 124
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
pyspark scripts to preprocess amazon-google data
- Loading branch information
1 parent
54a41be
commit f14e405
Showing
5 changed files
with
514 additions
and
51 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,72 +1,82 @@ | ||
| { | ||
| "fieldDefinition":[ | ||
| { | ||
| "fieldDefinition": [ | ||
| { | ||
| "fieldName" : "title", | ||
| "matchType" : "text", | ||
| "fields" : "title", | ||
| "dataType": "\"string\"" | ||
| "fieldName": "id", | ||
| "matchType": "DONT_USE", | ||
| "fields": "id", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "manufacturer", | ||
| "fieldName": "title", | ||
| "matchType": "text", | ||
| "fields": "title", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName": "manufacturer", | ||
| "matchType": "fuzzy", | ||
| "fields" : "manufacturer", | ||
| "dataType": "\"string\"" | ||
| "fields": "manufacturer", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "price", | ||
| "fieldName": "price", | ||
| "matchType": "fuzzy", | ||
| "fields" : "price", | ||
| "dataType": "\"double\"" | ||
| }], | ||
| "output" : [{ | ||
| "name":"output", | ||
| "format":"csv", | ||
| "fields": "price", | ||
| "dataType": "\"double\"" | ||
| } | ||
| ], | ||
| "output": [ | ||
| { | ||
| "name": "output", | ||
| "format": "csv", | ||
| "props": { | ||
| "location": "/tmp/zinggOutput", | ||
| "delimiter": ",", | ||
| "header":false | ||
| "header": false | ||
| } | ||
| }], | ||
| "data" : [{ | ||
| "name":"test1", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "examples/Amazon-Google/dataset_orig/tableA.csv", | ||
| "delimiter": ",", | ||
| "header":false | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| }, | ||
| } | ||
| ], | ||
| "data": [ | ||
| { | ||
| "name":"test2", | ||
| "format":"csv", | ||
| "name": "test1", | ||
| "format": "csv", | ||
| "props": { | ||
| "location": "examples/Amazon-Google/dataset_orig/tableB.csv", | ||
| "location": "examples/amazon-google/dataset_orig/tableA.csv", | ||
| "delimiter": ",", | ||
| "header":false | ||
| "header": false | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true}, | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| } | ||
| ], | ||
| "labelDataSampleSize" : 0.4, | ||
| "numPartitions":4, | ||
| "modelId": 1100, | ||
| "zinggDir": "models", | ||
| "collectMetrics": false | ||
| } | ||
| ] | ||
| }" | ||
| }, | ||
| { | ||
| "name": "test2", | ||
| "format": "csv", | ||
| "props": { | ||
| "location": "examples/amazon-google/dataset_orig/tableB.csv", | ||
| "delimiter": ",", | ||
| "header": false | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true}, | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| } | ||
| ], | ||
| "labelDataSampleSize": 0.4, | ||
| "numPartitions": 4, | ||
| "modelId": 1101, | ||
| "zinggDir": "models", | ||
| "collectMetrics": false | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,78 @@ | ||
| { | ||
| "fieldDefinition":[ | ||
| { | ||
| "fieldName" : "id", | ||
| "matchType" : "DONT_USE", | ||
| "fields" : "id", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "title", | ||
| "matchType" : "text", | ||
| "fields" : "title", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "manufacturer", | ||
| "matchType": "fuzzy", | ||
| "fields" : "manufacturer", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "price", | ||
| "matchType": "fuzzy", | ||
| "fields" : "price", | ||
| "dataType": "\"double\"" | ||
| }], | ||
| "output" : [{ | ||
| "name":"output", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "/tmp/zinggOutput", | ||
| "delimiter": ",", | ||
| "header":false | ||
| } | ||
| }], | ||
| "data" : [{ | ||
| "name":"testA", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "examples/amazon-google/testA.csv", | ||
| "delimiter": ",", | ||
| "header":true | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| }, | ||
| { | ||
| "name":"testB", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "examples/amazon-google/testB.csv", | ||
| "delimiter": ",", | ||
| "header":true | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| } | ||
| ], | ||
| "labelDataSampleSize" : 0.4, | ||
| "numPartitions":4, | ||
| "modelId": 1101, | ||
| "zinggDir": "models", | ||
| "collectMetrics": false | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,102 @@ | ||
| { | ||
| "trainingSamples" : [{ | ||
| "name":"trainingPos", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "examples/amazon-google/training.csv", | ||
| "delimiter": ",", | ||
| "header":false, | ||
| "badRecordsPath":"/tmp/bad" | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
|
|
||
| {\"name\" : \"z_cluster\",\"type\" : \"string\",\"nullable\" : false, \"metadata\" : {}}, | ||
| {\"name\":\"z_ismatch\",\"type\":\"integer\",\"nullable\":true,\"metadata\" : {}}, | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
|
|
||
| ] | ||
| }" | ||
|
|
||
| }], | ||
| "fieldDefinition":[ | ||
| { | ||
| "fieldName" : "id", | ||
| "matchType" : "DONT_USE", | ||
| "fields" : "id", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "title", | ||
| "matchType" : "text", | ||
| "fields" : "title", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "manufacturer", | ||
| "matchType": "fuzzy", | ||
| "fields" : "manufacturer", | ||
| "dataType": "\"string\"" | ||
| }, | ||
| { | ||
| "fieldName" : "price", | ||
| "matchType": "exact", | ||
| "fields" : "price", | ||
| "dataType": "\"double\"" | ||
| }], | ||
| "output" : [{ | ||
| "name":"output", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "/tmp/zinggOutput", | ||
| "delimiter": ",", | ||
| "header":false | ||
| } | ||
| }], | ||
| "data" : [{ | ||
| "name":"test1", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "examples/amazon-google/dataset_orig/tableA.csv", | ||
| "delimiter": ",", | ||
| "header":true | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| }, | ||
| { | ||
| "name":"test2", | ||
| "format":"csv", | ||
| "props": { | ||
| "location": "examples/amazon-google/dataset_orig/tableB.csv", | ||
| "delimiter": ",", | ||
| "header":true | ||
| }, | ||
| "schema": | ||
| "{\"type\" : \"struct\", | ||
| \"fields\" : [ | ||
| {\"name\":\"id\", \"type\":\"string\", \"nullable\":false}, | ||
| {\"name\":\"title\", \"type\":\"string\", \"nullable\":true}, | ||
| {\"name\":\"manufacturer\",\"type\":\"string\",\"nullable\":true} , | ||
| {\"name\":\"price\", \"type\":\"double\", \"nullable\":true} | ||
| ] | ||
| }" | ||
| } | ||
| ], | ||
| "labelDataSampleSize" : 0.4, | ||
| "numPartitions":4, | ||
| "modelId": 1101, | ||
| "zinggDir": "models", | ||
| "collectMetrics": false | ||
| } |
Oops, something went wrong.