remove pandas dependency

Author: Debabrata Acharya

csv_handler.py

@@ -1,9 +1,10 @@
+import copy
+import os
+from typing import List, Union
+
 import file
 import model as m
 import util as u
-from typing import List, Union
-import copy
-import os
 
 
 def readcsv(filename: str, separator: str = ',', rstrip: bool = True) -> List[List[str]]:

data.py

@@ -1,12 +1,11 @@
-import csv_handler as csv
-import model as m
-from typing import List, Union, Tuple
 import copy
+import os
+from typing import List
 import anndata as ad
 import scanpy as sc
-import pandas as pd
+import csv_handler as csv
+import model as m
 import util as u
-import os
 
 
 def build_model_from_csv(filename: str, separator: str = ',', rstrip: bool = True) -> m.DataMatrix:
@@ -48,7 +47,7 @@ def read_as_anndata(list_of_list: List[List[float]], roundoff_decimal: int = 5,
 	list_of_list = [[u.roundoff(value, roundoff_decimal) for value in row] for row in list_of_list]
 
 	u.create_path_if_not_exists(temp_folder)
-	pd.DataFrame(list_of_list).to_csv(complete_file_path, index=False, index_label=False, header=False)
+	csv.writecsv(filename, list_of_list, directory=temp_folder)
 
 	return sc.read_csv(complete_file_path)
 

datamatrix.py

@@ -1,6 +1,7 @@
+import copy
 from typing import List, Union, Tuple
+
 from sample import Sample
-import copy
 
 
 class DataMatrix:

preprocess.py

@@ -1,15 +1,12 @@
+import copy
+import statistics as st
 from typing import List, Tuple, Union
-import model
+import scanpy as sc
+import csv_handler as csv
 import data
+import model
 import normalization as nz
 import util as u
-import statistics as st
-import copy
-import pandas as pd
-import scanpy as sc
-import anndata as ad
-import csv_handler as csv
-import os
 
 
 def filter_csv_by_sd(filename: str, attr_count: int, separator: str = ',', rstrip: bool = True) -> model.DataMatrix:
@@ -205,8 +202,10 @@ def filter_singlecells(datamatrix: model.DataMatrix, min_cells: int, min_genes:
 
 	datamatrix = filter_genes(datamatrix, min_genes=min_genes, roundoff_decimal=roundoff_decimal, filehash=filehash)
 
+	u.clear_temp()
+
 	return datamatrix
 
 
-def fsc(datamatrix: model.DataMatrix, min_cells: int, min_genes: int, rd: int = 5) -> model.DataMatrix:
-	return filter_singlecells(datamatrix, min_cells, min_genes=min_genes, roundoff_decimal=rd)
+def fsc(datamatrix: model.DataMatrix, mc: int, mg: int, rd: int = 5) -> model.DataMatrix:
+	return filter_singlecells(datamatrix, mc, mg, roundoff_decimal=rd)

sample.py

@@ -1,6 +1,7 @@
+import copy
 from typing import List, Union
+
 from datapoint import DataPoint
-import copy
 
 
 class Sample:

sample_datasets/bmmcaml-5898-1000.csv

@@ -1,10 +1,12 @@
-from typing import List, Union
 import copy
-import model as m
-import string
-import random
-import os
 import math
+import os
+import random
+import string
+import glob
+from typing import List, Union
+
+import model as m
 
 
 def hash(length=10) -> str:
@@ -103,3 +105,17 @@ def get_classlabeled_list_of_list_from_datamatrix(datamatrix: m.DataMatrix) -> L
 
 def gclld(datamatrix: m.DataMatrix) -> List[List[Union[float, str]]]:
 	return get_classlabeled_list_of_list_from_datamatrix(datamatrix)
+
+
+def clear_temp():
+
+	temp_folder: str = '__temp__'
+
+	if not is_valid_path(temp_folder):
+		return
+
+	filelist = glob.glob(os.path.join(temp_folder, '*'))
+
+	for f in filelist:
+		if is_valid_path(os.path.join(f)):
+			os.remove(f)