func TestLNSmote(t *testing.T) {
// Read sample dataset.
dataset := tabula.Claset{}
_, e := dsv.SimpleRead(fcfg, &dataset)
if nil != e {
t.Fatal(e)
}
fmt.Println("[lnsmote_test] Total samples:", dataset.GetNRow())
// Write original samples.
writer, e := dsv.NewWriter("")
if nil != e {
t.Fatal(e)
}
e = writer.OpenOutput("phoneme_lnsmote.csv")
if e != nil {
t.Fatal(e)
}
sep := dsv.DefSeparator
_, e = writer.WriteRawRows(dataset.GetRows(), &sep)
if e != nil {
t.Fatal(e)
}
// Initialize LN-SMOTE.
lnsmoteRun := lnsmote.New(100, 5, 5, "1", "lnsmote.outliers")
e = lnsmoteRun.Resampling(&dataset)
fmt.Println("[lnsmote_test] # synthetic:", lnsmoteRun.Synthetics.Len())
sep = dsv.DefSeparator
_, e = writer.WriteRawRows(lnsmoteRun.Synthetics.GetRows(), &sep)
if e != nil {
t.Fatal(e)
}
e = writer.Close()
if e != nil {
t.Fatal(e)
}
}
/*
InitReadWriter initialize reader and writer.
*/
func InitReadWriter(finput string) (reader *wvcgen.Reader,
writer *dsv.Writer) {
reader, e := wvcgen.NewReader(finput)
if e != nil {
panic(e)
}
writer, e = dsv.NewWriter(finput)
if e != nil {
panic(e)
}
initReader(reader)
return reader, writer
}
// writeOutliers will save the `outliers` to file specified by
// `OutliersFile`.
func (in *Runtime) writeOutliers() (e error) {
writer, e := dsv.NewWriter("")
if nil != e {
return
}
e = writer.OpenOutput(in.OutliersFile)
if e != nil {
return
}
sep := dsv.DefSeparator
_, e = writer.WriteRawRows(&in.outliers, &sep)
if e != nil {
return
}
return writer.Close()
}
//
// WriteSynthetics will write synthetic samples in resampling module `ri` into
// `file`.
//
func WriteSynthetics(ri Interface, file string) (e error) {
writer, e := dsv.NewWriter("")
if nil != e {
return
}
e = writer.OpenOutput(file)
if e != nil {
return
}
sep := dsv.DefSeparator
_, e = writer.WriteRawDataset(ri.GetSynthetics(), &sep)
if e != nil {
return
}
return writer.Close()
}
// runMerge will append original dataset to synthetic file.
func runMerge(smote *smote.Runtime, dataset *tabula.Claset) (e error) {
writer, e := dsv.NewWriter("")
if e != nil {
return
}
e = writer.ReopenOutput(smote.SyntheticFile)
if e != nil {
return
}
sep := dsv.DefSeparator
n, e := writer.WriteRawDataset(dataset, &sep)
if e != nil {
return
}
if DEBUG >= 1 {
fmt.Println("[smote] # appended:", n)
}
return writer.Close()
}
func TestWriteRawColumns(t *testing.T) {
dataset := tabula.NewDataset(tabula.DatasetModeColumns, nil, nil)
populateWithColumns(t, dataset)
writer, e := dsv.NewWriter("")
if e != nil {
t.Fatal(e)
}
outfile := "testdata/writerawcolumns.out"
expfile := "testdata/writeraw.exp"
e = writer.OpenOutput(outfile)
_, e = writer.WriteRawDataset(dataset, nil)
if e != nil {
t.Fatal(e)
}
assertFile(t, outfile, expfile, true)
}
func TestMergeRows(t *testing.T) {
reader1, e := dsv.NewReader("testdata/config.dsv", nil)
if nil != e {
t.Fatal(e)
}
reader2, e := dsv.NewReader("testdata/config_skip.dsv", nil)
if nil != e {
t.Fatal(e)
}
reader1.SetMaxRows(-1)
reader2.SetMaxRows(-1)
_, e = dsv.Read(reader1)
if e != io.EOF {
t.Fatal(e)
}
_, e = dsv.Read(reader2)
if e != io.EOF {
t.Fatal(e)
}
e = reader1.Close()
if e != nil {
t.Fatal(e)
}
e = reader2.Close()
if e != nil {
t.Fatal(e)
}
reader1.MergeRows(reader2)
// write merged reader
writer, e := dsv.NewWriter("")
if e != nil {
t.Fatal(e)
}
outfile := "testdata/output_merge_rows.dat"
e = writer.OpenOutput(outfile)
if e != nil {
t.Fatal(e)
}
sep := "\t"
ds1 := reader1.GetDataset().(tabula.DatasetInterface)
_, e = writer.WriteRawDataset(ds1, &sep)
if e != nil {
t.Fatal(e)
}
e = writer.Close()
if e != nil {
t.Fatal(e)
}
expfile := "testdata/expected_merge_rows.dat"
assertFile(t, outfile, expfile, true)
}