scilens 0.4.7__py3-none-any.whl → 0.4.9__py3-none-any.whl

scilens/config/models/reader_metrics.py

@@ -3,4 +3,4 @@ _B='forbid'
 _A=None
 from pydantic import BaseModel,Field
 class ReaderTxtMetricsConfig(BaseModel,extra=_B):name:str|_A=Field(default=_A,description=_C);pattern:str=Field(default=_A,description='Expression régulière pour identifier la métrique.');number_position:int=Field(default=1,description='Position du nombre dans le tableau de nombres de la ligne.')
-class ReaderColsMetricsConfig(BaseModel,extra=_B):name:str|_A=Field(default=_A,description=_C);col:int|str|_A=Field(default=_A,description="Index de la colonne pour la métrique (index `1` pour la première colonne). Si `String`, nom de la colonne pour la métrique. Si la colonne n'existe pas, il n'y aura pas de parsing. (Valide seulement si la première ligne est les en-têtes)");aggregation:str=Field(default='sum',description="Méthode d'agrégation pour la métrique. Peut être `mean`, `sum`, `min`, `max`.")
+class ReaderColsMetricsConfig(BaseModel,extra=_B):name:str|_A=Field(default=_A,description=_C);aggregation:str=Field(default='sum',description="Méthode d'agrégation du vecteur `col` ou du vecteur résultat de la `function`. Peut être `mean`, `sum`, `min`, `max`.");col:int|str|_A=Field(default=_A,description='Colonnes (index ou noms) de la métrique.');function:str|_A=Field(default=_A,description="Fonction norme de l'espace vectoriel . Peut être `euclidean_norm`.");components:list[int]|list[str]|_A=Field(default=_A,description='Colonnes (index ou noms) de la `function`.')

scilens/readers/cols_dataset.py

@@ -9,40 +9,59 @@ from scilens.components.compare_models import CompareGroup
 from scilens.components.compare_floats import CompareFloats
 from scilens.config.models.reader_format_cols import ReaderCurveParserNameConfig
 from scilens.config.models.reader_metrics import ReaderColsMetricsConfig
-def get_index_col_index(index_col,numeric_col_indexes,names):
-	C=names;A=index_col;B=_A
-	if isinstance(A,int):B=A-1
-	if isinstance(A,str):A=[A]
-	if isinstance(A,list):
+import math
+def get_col_indexes(source,numeric_col_indexes,names):
+	C=names;A=source;B=[]
+	if not isinstance(A,list):A=[A]
+	if isinstance(A[0],int):
+		for F in A:B.append(F-1)
+	else:
 		for D in A:
-			if D in C:B=C.index(D);break
-	if B:
-		if B not in numeric_col_indexes:raise ValueError(f"Index column index {A} is not a numeric column.")
+			if D in C:B.append(C.index(D))
+	for E in B:
+		if E not in numeric_col_indexes:raise ValueError(f"Index column index {E} is not a numeric column.")
 	return B
 @dataclass
 class ColsDataset:
 	cols_count:int=0;rows_count:int=0;names:list[str]=field(default_factory=lambda:[]);numeric_col_indexes:list[int]=field(default_factory=lambda:[]);data:list[list[float]]=field(default_factory=lambda:[]);origin_line_nb:list[int]=field(default_factory=lambda:[])
-	def get_curves_col_x(I,col_x):
-		H='title';D=col_x;A=I;F={};C=get_index_col_index(D,A.numeric_col_indexes,A.names);F[_D]=C;J=[B for(A,B)in enumerate(A.numeric_col_indexes)if A!=C];E=[];G=[]
-		for B in J:D=A.data[C];K=A.data[B];L={H:A.names[B],'short_title':A.names[B],'series':[[D[A],K[A]]for A in range(A.rows_count)],_C:B};E+=[L];M={H:A.names[B],'type':'simple','xaxis':A.names[C],'yaxis':A.names[B],_B:[len(E)-1]};G+=[M]
-		return{_B:E,_E:G},F
-	def compute_metrics(I,config):
-		D=I;H={}
-		for A in config:
-			E=A.col;B=_A
-			if isinstance(E,int):B=D.numeric_col_indexes.index(E-1)
-			if isinstance(E,str):B=D.names.index(E)
-			if B is _A or B<0 or B>=D.cols_count:raise ValueError(f"Metric '{A.name}' has an invalid column: {E}. Skipping.")
-			G=A.name
-			if not G:G=f"{D.names[B]} {A.aggregation}"
-			F=D.data[B];C=_A
-			if A.aggregation=='mean':C=sum(F)/len(F)
-			elif A.aggregation=='sum':C=sum(F)
-			elif A.aggregation=='min':C=min(F)
-			elif A.aggregation=='max':C=max(F)
-			if C is _A:raise ValueError(f"Metric '{A.name}' has an invalid aggregation: {A.aggregation}.")
-			H[G]=C
-		return H
+	def get_col_indexes(A,col_x):return get_col_indexes(col_x,A.numeric_col_indexes,A.names)
+	def get_curves_col_x(F,col_x):
+		J='title';C=col_x;A=F;G={};H=F.get_col_indexes(C)
+		if not H:raise ValueError(f"get_curves_col_x Column {C} not found in numeric columns.")
+		D=H[0];G[_D]=D;K=[B for(A,B)in enumerate(A.numeric_col_indexes)if A!=D];E=[];I=[]
+		for B in K:C=A.data[D];L=A.data[B];M={J:A.names[B],'short_title':A.names[B],'series':[[C[A],L[A]]for A in range(A.rows_count)],_C:B};E+=[M];N={J:A.names[B],'type':'simple','xaxis':A.names[D],'yaxis':A.names[B],_B:[len(E)-1]};I+=[N]
+		return{_B:E,_E:I},G
+	def compute_metrics(K,config):
+		C=K;L={}
+		for(H,F)in enumerate(config):
+			I=F.name;B=F.col;J=F.function;D=F.aggregation
+			if not B and not J:raise ValueError(f"Metric #{H} must have either a column or a function.")
+			if B and J:raise ValueError(f"Metric #{H} cannot have both a column and a function.")
+			if B:
+				E=_A
+				if isinstance(B,int):E=C.numeric_col_indexes.index(B-1)
+				if isinstance(B,str):E=C.names.index(B)
+				if E is _A or E<0 or E>=C.cols_count:raise ValueError(f"Metric #{H} has an invalid column: {B}.")
+				if not I:I=f"{C.names[E]} {D}"
+				A=C.data[E]
+			elif J:
+				A=[0 for A in range(C.rows_count)]
+				if J=='euclidean_norm':
+					if not I:I=f"Euclidean Norm {F.components} {D}"
+					M=K.get_col_indexes(F.components)
+					if not M:continue
+					for N in M:
+						for(O,P)in enumerate(C.data[N]):A[O]+=P**2
+					A=[math.sqrt(A)for A in A]
+				else:raise ValueError(f"Metric #{H} has an invalid function: {J}.")
+			G=_A
+			if D=='mean':G=sum(A)/len(A)
+			elif D=='sum':G=sum(A)
+			elif D=='min':G=min(A)
+			elif D=='max':G=max(A)
+			if G is _A:raise ValueError(f"Metric #{H} has an invalid aggregation: {D}.")
+			L[I]=G
+		return L
 @dataclass
 class ColsCurves:type:str;info:dict;curves:dict
 def compare(group,compare_floats,reader_test,reader_ref,cols_curve):

scilens/readers/reader_csv.py

@@ -2,7 +2,7 @@ _B=True
 _A=None
 import logging,csv,re
 from scilens.readers.reader_interface import ReaderInterface
-from scilens.readers.cols_dataset import ColsDataset,ColsCurves,compare as cols_compare,get_index_col_index
+from scilens.readers.cols_dataset import ColsDataset,ColsCurves,compare as cols_compare,get_col_indexes
 from scilens.readers.mat_dataset import MatDataset,from_iterator as mat_from_iterator,compare as mat_compare,get_data
 from scilens.config.models.reader_format_csv import ReaderCsvConfig,ReaderCsvMatrixConfig
 from scilens.config.models.reader_format_cols import ReaderCurveParserNameConfig
@@ -27,10 +27,10 @@ class ReaderCsv(ReaderInterface):
 				if bool(re.match(B,line)):return _B
 		return False
 	def read(A,reader_options):
-		C=reader_options;A.reader_options=C;F,L,U=csv_detect(A.origin.path,A.reader_options.delimiter,A.reader_options.quotechar,encoding=A.encoding);A.has_header=F;A.cols=L;A.numeric_col_indexes=U;A.index_col_index=_A;A.ignore_lines_patterns=_A;G=_A;H=_A;I=_A;J=_A;B=C.cols
+		C=reader_options;A.reader_options=C;F,L,V=csv_detect(A.origin.path,A.reader_options.delimiter,A.reader_options.quotechar,encoding=A.encoding);A.has_header=F;A.cols=L;A.numeric_col_indexes=V;A.index_col_index=_A;A.ignore_lines_patterns=_A;G=_A;H=_A;I=_A;J=_A;B=C.cols
 		if B:
 			if B.index_col:
-				if B.index_col:A.index_col_index=get_index_col_index(B.index_col,A.numeric_col_indexes,L)
+				if B.index_col:R=get_col_indexes(B.index_col,A.numeric_col_indexes,L);A.index_col_index=R[0]if R else _A
 			if B.rows:
 				A.ignore_lines_patterns=B.rows.ignore_patterns;G=B.rows.line_start;H=B.rows.line_end
 				if G and H and H<G:raise ValueError(f"Line end {H} cannot be before line start {G}.")
@@ -39,8 +39,8 @@ class ReaderCsv(ReaderInterface):
 					I=B.rows.index_min_value;J=B.rows.index_max_value
 					if I and J and I>J:raise ValueError(f"Index min value {I} cannot be greater than index max value {J}.")
 		A.raw_lines_number=_A;A.curves=_A;A.report_matrices=_A
-		with open(A.origin.path,'r',encoding=A.encoding)as V:
-			R=V.readlines();M=csv.reader(R,delimiter=A.reader_options.delimiter,quotechar=A.reader_options.quotechar)
+		with open(A.origin.path,'r',encoding=A.encoding)as W:
+			S=W.readlines();M=csv.reader(S,delimiter=A.reader_options.delimiter,quotechar=A.reader_options.quotechar)
 			if C.is_matrix:
 				K=C.matrix or ReaderCsvMatrixConfig();P=mat_from_iterator(x_name=K.x_name,y_name=K.y_name,reader=M,has_header=F,x_value_line=K.x_value_line,has_y=K.has_y)
 				if K.export_report:A.report_matrices=get_data([P],['csv'])
@@ -49,8 +49,8 @@ class ReaderCsv(ReaderInterface):
 				if C.cols and C.cols.ignore_columns:
 					if not F:raise Exception('Ignore columns is not supported without header.')
 					if isinstance(C.cols.ignore_columns[0],str):A.numeric_col_indexes=[B for B in A.numeric_col_indexes if A.cols[B]not in C.cols.ignore_columns]
-					if isinstance(C.cols.ignore_columns[0],int):W=[A-1 for A in B.ignore_columns];A.numeric_col_indexes=[A for A in A.numeric_col_indexes if A not in W]
-				S=len(L);D=ColsDataset(cols_count=S,names=L,numeric_col_indexes=A.numeric_col_indexes,data=[[]for A in range(S)]);E=0
+					if isinstance(C.cols.ignore_columns[0],int):X=[A-1 for A in B.ignore_columns];A.numeric_col_indexes=[A for A in A.numeric_col_indexes if A not in X]
+				T=len(L);D=ColsDataset(cols_count=T,names=L,numeric_col_indexes=A.numeric_col_indexes,data=[[]for A in range(T)]);E=0
 				if F and E==0:next(M);E+=1
 				if G:
 					try:
@@ -63,11 +63,11 @@ class ReaderCsv(ReaderInterface):
 						N=next(M);E+=1
 						if H and E>H:break
 						if J is not _A and float(N[A.index_col_index])>J:break
-						if A.ignore_lines_patterns and A._ignore_line(R[E-1].rstrip('\n')):continue
+						if A.ignore_lines_patterns and A._ignore_line(S[E-1].rstrip('\n')):continue
 						if I is not _A and float(N[A.index_col_index])<I:continue
-						for(T,Q)in enumerate(N):
-							if T in D.numeric_col_indexes:Q=float(Q)
-							D.data[T].append(Q)
+						for(U,Q)in enumerate(N):
+							if U in D.numeric_col_indexes:Q=float(Q)
+							D.data[U].append(Q)
 						D.origin_line_nb.append(E)
 				except StopIteration:pass
 				D.rows_count=len(D.origin_line_nb);A.cols_dataset=D;A.raw_lines_number=D.rows_count+(1 if F else 0);A.metrics=_A
@@ -78,8 +78,8 @@ class ReaderCsv(ReaderInterface):
 						if B.curve_parser.parameters:O=B.curve_parser.parameters.x
 						elif B.index_col:O=B.index_col
 						if not O:raise ValueError('Curve parser COL_X requires a parameter x, or index_col to be defined.')
-						A.curves,X=D.get_curves_col_x(O)
-						if A.curves:A.cols_curve=ColsCurves(type=ReaderCurveParserNameConfig.COL_X,info=X,curves=A.curves)
+						A.curves,Y=D.get_curves_col_x(O)
+						if A.curves:A.cols_curve=ColsCurves(type=ReaderCurveParserNameConfig.COL_X,info=Y,curves=A.curves)
 					elif B.curve_parser.name==ReaderCurveParserNameConfig.COLS_COUPLE:raise NotImplementedError('cols_couple not implemented')
 					else:raise Exception('Curve parser not supported.')
 	def compare(A,compare_floats,param_reader,param_is_ref=_B):

{scilens-0.4.7.dist-info → scilens-0.4.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: scilens
-Version: 0.4.7
+Version: 0.4.9
 Summary: A CesGensLaB framework for data collecting and deep analysis
 Home-page: https://scilens.dev
 License: Proprietary

{scilens-0.4.7.dist-info → scilens-0.4.9.dist-info}/RECORD

@@ -32,7 +32,7 @@ scilens/config/models/reader_format_csv.py,sha256=KWEH0c12n6hdaWAKdHXYMT5SvH4UqB
 scilens/config/models/reader_format_netcdf.py,sha256=Skr5lZACqVRrrlvs7R1RVFBpMTDthLJN-Fbs24jtb7o,1271
 scilens/config/models/reader_format_txt.py,sha256=35rvhg311DN9miE47QM9NN_gcZHglYG2lBZPLJy4iyg,1462
 scilens/config/models/reader_format_txt_fixed_cols.py,sha256=XmmFS0LdGR81XCjJIyW4cngivFo_8drckqe_k8fhqJk,1385
-scilens/config/models/reader_metrics.py,sha256=qJBnq6gix-vkNeCz7LuVu04n8q6Rx1KUmkqy-J6BCDY,900
+scilens/config/models/reader_metrics.py,sha256=uLwyd2vE4GRHbArYRv1BRH_acTL1hbFB7HgZRDJiAJg,965
 scilens/config/models/readers.py,sha256=NpEjXbLv11k9N8UsnuS55KEYlRijF9C3eOrI0SB9dqM,1723
 scilens/config/models/report.py,sha256=_4W96v5izWCC_CrMzh3PM0mZIUAa3L26ZQqjV0Mv0f4,1008
 scilens/config/models/report_html.py,sha256=IsKlmE7QbSTiPrOaxPc48Ne7xPE5yqTcLjxqN_i-unY,3803
@@ -49,10 +49,10 @@ scilens/processors/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJW
 scilens/processors/models/results.py,sha256=KoWxh13Zgi7PuPql8hkf4VjCis42ZxAuzIgJxBWVaX8,119
 scilens/processors/processor_interface.py,sha256=jzMp1529JXnMGTJijVy6b_1zmARAMNv70f2lgys7vn4,452
 scilens/readers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-scilens/readers/cols_dataset.py,sha256=OpCTS9FmJxCHLoCrrD56sHY3f9dxwTsw2Xyxf_vNv_8,3270
+scilens/readers/cols_dataset.py,sha256=6Cq9aIG4UMOdam_zx64lf2f0e2f0lI2VZDNZklccBaI,3964
 scilens/readers/exceptions.py,sha256=JzmxcjnR5sH-IOWVeCC5A1bSwxv-jCAtIJvDjzx1CTI,32
 scilens/readers/mat_dataset.py,sha256=Z9TYDWaH2aqdniLNDjlpR6VVNHMSARjh52clhdMyOn4,1496
-scilens/readers/reader_csv.py,sha256=5TQ9TAVSNUje7Mbqy9T1SfJ1f4Ic_bFBTD9S5OuMrGU,5539
+scilens/readers/reader_csv.py,sha256=FKxx26Vzf_h25cFX37DwFXy_-toUGw3481iMbyRa-vg,5550
 scilens/readers/reader_interface.py,sha256=r1pu9LyweTGXU8YfI3FPZy1Em4stzmJb-6j90j1tPQQ,938
 scilens/readers/reader_manager.py,sha256=DFinxIk3IIIcB6JxybGcv-mXt3jhXgCwUtzR0TqhB2Q,2684
 scilens/readers/reader_txt.py,sha256=U3hGIorj-Nv-jq6zYtvbDv2LQBHTgW52PHbV8A5FMA8,4526
@@ -113,7 +113,7 @@ scilens/utils/template.py,sha256=9dlXX3nmfzDRUwzPJOkoxk15UXivZ2SW-McdCwokFa4,443
 scilens/utils/time_tracker.py,sha256=DdVBoMpVLXrX0qZZXyLm4g38EwDVLlRcBqcpNex1mYY,545
 scilens/utils/vectors.py,sha256=4N2BZSC5n3HgZqPujDGF5NdjVmSL1rOHb_qw4OoABQY,103
 scilens/utils/web.py,sha256=MAFWpIFOKz7QhqDoFh-Qwstvc76KpcxstSgHFT8FOL4,901
-scilens-0.4.7.dist-info/METADATA,sha256=Q6exfgps2ADhfvgJqlhrQdKbffPAfgrdvGBM47bd9PY,1367
-scilens-0.4.7.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-scilens-0.4.7.dist-info/entry_points.txt,sha256=DaKGgxUEUv34GJAoXtta6ecL37ercejep9sCSSRQK2s,48
-scilens-0.4.7.dist-info/RECORD,,
+scilens-0.4.9.dist-info/METADATA,sha256=pz6dHk21Xl_u0bwZeCj6clWQivZddEpxCJ2KsLVW2Xk,1367
+scilens-0.4.9.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+scilens-0.4.9.dist-info/entry_points.txt,sha256=DaKGgxUEUv34GJAoXtta6ecL37ercejep9sCSSRQK2s,48
+scilens-0.4.9.dist-info/RECORD,,