/
cleaning.go
106 lines (92 loc) · 3.51 KB
/
cleaning.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
//
// Copyright © 2021 Uncharted Software Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package task
import (
"github.com/pkg/errors"
"github.com/uncharted-distil/distil-compute/metadata"
"github.com/uncharted-distil/distil-compute/model"
"github.com/uncharted-distil/distil-compute/primitive/compute"
"github.com/uncharted-distil/distil-compute/primitive/compute/description"
"github.com/uncharted-distil/distil/api/serialization"
)
// Clean will clean bad data for further processing.
func Clean(schemaFile string, dataset string, params *IngestParams, config *IngestTaskConfig) (string, error) {
outputPath := createDatasetPaths(schemaFile, dataset, compute.D3MLearningData)
// load metadata from original schema
meta, err := metadata.LoadMetadataFromOriginalSchema(schemaFile, true)
if err != nil {
return "", errors.Wrap(err, "unable to load original schema file")
}
mainDR := meta.GetMainDataResource()
metaStorage, err := params.MetaCtor()
if err != nil {
return "", errors.Wrap(err, "unable to initialize metadata storage")
}
vars := []*model.Variable{}
exists, _ := metaStorage.DatasetExists(meta.ID)
if exists {
vars, err = metaStorage.FetchVariables(meta.ID, false, false, false)
if err != nil {
return "", err
}
} else if params.DefinitiveTypes != nil {
for _, v := range mainDR.Variables {
if params.DefinitiveTypes[v.Key] != nil {
clone := v.Clone()
clone.Type = params.DefinitiveTypes[v.Key].Type
vars = append(vars, clone)
}
}
}
// create & submit the solution request
pip, err := description.CreateDataCleaningPipeline("Mary Poppins", "", vars, config.ImputeEnabled)
if err != nil {
return "", errors.Wrap(err, "unable to create format pipeline")
}
// pipeline execution assumes datasetDoc.json as schema file
datasetURI, err := submitPipeline([]string{outputPath.sourceFolder}, pip, true)
if err != nil {
return "", errors.Wrap(err, "unable to run format pipeline")
}
// output the header
output := [][]string{}
header := make([]string, len(mainDR.Variables))
for _, v := range mainDR.Variables {
header[v.Index] = v.HeaderName
}
output = append(output, header)
// parse primitive response (raw data from the input dataset)
// first row of the data is the header
// first column of the data is the dataframe index
readStorage := serialization.GetStorage(datasetURI)
csvData, err := readStorage.ReadData(datasetURI)
if err != nil {
return "", errors.Wrap(err, "unable to parse clean result")
}
output = append(output, csvData[1:]...)
// output the data
datasetStorage := serialization.GetStorage(outputPath.outputData)
err = datasetStorage.WriteData(outputPath.outputData, output)
if err != nil {
return "", errors.Wrap(err, "error writing clustered output")
}
mainDR.ResPath = outputPath.outputData
// write the new schema to file
err = datasetStorage.WriteMetadata(outputPath.outputSchema, meta, true, false)
if err != nil {
return "", errors.Wrap(err, "unable to store cluster schema")
}
return outputPath.outputSchema, nil
}