similarité

.gcloudignore

@@ -1,19 +0,0 @@
-# This file specifies files that are *not* uploaded to Google Cloud
-# using gcloud. It follows the same syntax as .gitignore, with the addition of
-# "#!include" directives (which insert the entries of the given .gitignore-style
-# file at that point).
-#
-# For more information, run:
-#   $ gcloud topic gcloudignore
-#
-.gcloudignore
-# If you would like to upload your .git directory, .gitignore file or files
-# from your .gitignore file, remove the corresponding line
-# below:
-.git
-.gitignore
-
-# Python pycache:
-__pycache__/
-# Ignored by the build system
-/setup.cfg

App/utils/dataset/Normalisation - dictionnaire.tsv

@@ -0,0 +1,146 @@
+ABBREVIATIONS	CORRESPONDANCES	
+dissolv	dissolvant	
+diss	dissolvant	
+masc	mascara	
+mlvernis	ml vernis	
+ong	ongle	
+soi	soins	
+bjs	bourjois	10ML BL CERNES AL FAB 200 BJS BOURJOIS COTY
+pdr	poudre	
+plm	plm	10G PLM 04JAUN.TRANS.BO.GR.BIO
+poud	poudre	
+bg	bg	1.6G CRAYON YEUX 06NOIS BG BIO
+yx	yeux	
+eye	yeux	
+y	yeux	
+cra	crayon 	
+cr	crème	
+cray	crayon 	
+ess	essentiel	
+leg	legume	
+ver	vert	
+vrt	vert	
+bio	biologique	
+lsirop	l sirop	
+spec	special	
+cdp	compagnie de province	
+demaq	demaquillant	
+trse	trousse	
+eaa	eucerin anti age	
+eaf	eafit	
+epil	epilation	
+veg	vegetale	
+pfum	parfum	
+gaill	gaillac	
+juranc	jurancon	
+bor	bordeaux	
+bord	bordeaux	
+hle	huile	
+aoc	appelation d'origine contrôlée	
+aop	appelation d'origine protégée	
+rg	rouge	
+rges	rouge	
+rge	rouge	
+rse	rose	
+rs	rose	
+bl	blanc	
+bdx	Bordeaux	
+vdt	vin de table	
+vdp	vin de pays	
+blc	blanc	
+bib	bag in box	
+citr	citron	
+co	coco	
+gourm	gourmand	
+patis	patisserie	
+p'tits	petit	
+p'tit	petit	
+p tit	petit	
+pt	pepite	
+rev	revil	
+succ	sucettes	
+succet	sucettes	
+chocohouse	choco house	
+sach	sachet	
+choc	chocolat	
+tab	tablette	
+hte	haute	
+spagh	spaghetti	
+scht	sachet	
+nr	noir	
+caf	cafe	
+barr	barre	
+pces	pieces	
+pc	pieces	
+acidu	acidule	
+blnc	blanc	
+frui	fruit	
+gourman	gourmand	
+bte	boîte	
+bt	boîte	
+ptit	petit	
+corb	corbeil	
+ptits	petit	
+pti	petit	
+nois	noisette	
+poul	poulain	
+barq	barquette	
+barqu	barquette	
+fizz	fizzy	
+st	saint	
+mich	michel	
+cal	calendrier	
+calend	calendrier	
+calendr	calendrier	
+caram	caramel	
+cava	cavalier	
+har	haribo	
+choco	chocolat	
+lt	lait	
+choc'n	chocolat noir	
+choc n	chocolat noir	
+degust	degustation	
+degus	degustation	
+bis	biscuit	
+coffr	coffret	
+coff	coffret	
+cof	coffet	
+conf	confiserie	
+confis	confiserie	
+croco	crocodile	
+dble	double	
+dess	dessert	
+doyp	doypack	
+harib	harib	
+et	etui	
+exc	excellence	
+excel	excellence	
+frit	friture	
+fritu	friture	
+fritur	friture	
+gd	grand	
+gr	grand	
+grd	grand	
+grchoc	grand chocolat	
+lat	lait	
+ass	assorti	
+assoti	assorti	
+noug	nougatine	
+nougat	nougatine	
+sct	secret	
+cho	chocolat	
+bisc	biscuit	
+am	amande	
+liq	liqueur	
+tabl	tablette	
+asst	assorti	
+bil	bille	
+vali	valisette	
+cda	chevaliers d argouges	
+tub	tubo	
+gril	grille	
+amandesgrilles	amandes grilles	
+ball	ballotin	
+piecestubo	pieces tubo	
+bonb	bonbon	
+dej	dejeuner	

App/utils/divers_function.py

@@ -123,4 +123,26 @@ def cosine_similarity_between_expressions(expr1, expr2):
     vectors = vectorizer.fit_transform([expr1, expr2])
     similarity = cosine_similarity(vectors[0], vectors[1])
 
-    return similarity[0][0]
+    return similarity[0][0]
+
+def ajout_simularite(data) :
+    data["ITEM_DESC_avant_clean"] = data["ITEM_DESC_x"].apply(data_cleaning)
+    data["ITEM_DESC_apres_clean"] = data["ITEM_DESC_y"].apply(data_cleaning)
+
+    stop = stopwords.words('french')
+    data['ITEM_DESC_avant_clean'] = data['ITEM_DESC_avant_clean'].apply(lambda x: " ".join(x for x in x.split() if x not in stop))
+    data['ITEM_DESC_apres_clean'] = data['ITEM_DESC_apres_clean'].apply(lambda x: " ".join(x for x in x.split() if x not in stop))
+
+    stop = stopwords.words('english')
+    data['ITEM_DESC_avant_clean'] = data['ITEM_DESC_avant_clean'].apply(lambda x: " ".join(x for x in x.split() if x not in stop))
+    data['ITEM_DESC_apres_clean'] = data['ITEM_DESC_apres_clean'].apply(lambda x: " ".join(x for x in x.split() if x not in stop))
+
+    data['ITEM_DESC_avant_clean'] = data['ITEM_DESC_avant_clean'].apply(remove_stop_words)
+    data['ITEM_DESC_apres_clean'] = data['ITEM_DESC_apres_clean'].apply(remove_stop_words)
+
+    data['ITEM_DESC_avant_clean'] = data['ITEM_DESC_avant_clean'].apply(standardization)
+    data['ITEM_DESC_apres_clean'] = data['ITEM_DESC_apres_clean'].apply(standardization)
+
+    data["Cosinus similarité"]  =  data.apply(lambda row: cosine_similarity_between_expressions(row['ITEM_DESC_apres_clean'], row['ITEM_DESC_avant_clean']), axis=1)
+
+    return data

app.py

@@ -11,7 +11,7 @@ from App.utils.filter_dataframe import *
 # Page configuration
 st.set_page_config(
     page_title="Gestion des ruptures",
-    page_icon="Carrefour_logo.png",
+    page_icon="images/Carrefour_logo.png",
     layout="wide"
 )
 hide_streamlit_style = """
@@ -22,7 +22,6 @@ hide_streamlit_style = """
 st.markdown(hide_streamlit_style, unsafe_allow_html=True)
 
 
-
 def app():
         st.title("Gestion des ruptures ")
 
@@ -72,7 +71,7 @@ def app():
                         st.subheader("Show data with ratios")
                         merged_final.loc[:, "Evaluation"]= True
                         merged_final  = st.data_editor(merged_final)
-                        #st.dataframe(merged_final)
+                        
 
                         csv = convert_df(merged_final)
                         st.download_button(label="Download data as CSV",
@@ -105,7 +104,7 @@ def app():
                         st.subheader("Data without decision-making")
                         df.loc[:, "Evaluation"] = True
                         df = st.data_editor(df)
-                        #st.dataframe(df)
+                        
                         st.download_button(label="Download data as CSV",
                                            data=csv,
                                            file_name='sample_df.csv',
@@ -115,7 +114,7 @@ def app():
                         st.subheader("Data with proposed changes")
                         finale_df.loc[:, "Evaluation"] = True
                         finale_df = st.data_editor(finale_df)
-                        #st.dataframe(finale_df)
+                        
                         csv_f = convert_df(finale_df)
                         st.download_button(label="Download data as CSV",
                                                data=csv_f,
@@ -131,7 +130,7 @@ def app():
                         st.subheader("Data without decision-making")
                         priority_data.loc[:, "Evaluation"] = True
                         priority_data = st.data_editor(priority_data)
-                        #st.dataframe(priority_data)
+                        
                         csv_f = convert_df(priority_data)
                         st.download_button(label="Download data as CSV",
                                                data=csv_f,
@@ -142,7 +141,7 @@ def app():
                         st.subheader("Equality case")
                         df_equa.loc[:, "Evaluation"]= True
                         df_equa = st.data_editor(df_equa)
-                        #st.dataframe(df_equa)
+                        
                         csv_f = convert_df(df_equa)
                         st.download_button(label="Download data as CSV",
                                                data=csv_f,
@@ -162,13 +161,13 @@ def app():
                                                mime='text/csv',)
                         
 
-                    #df_finale= data_1_1(priority_data, product_id, class_id)  
+                     
                     max_poids_index = df_nequa_.groupby('BARCODE')['Poids'].idxmax()
 
-                    # Récupérer les lignes correspondantes
+                    
                     df_max_poids = df_nequa_.loc[max_poids_index]
                     df_max_poids.drop(["COUNTRY_KEY"], axis = 1, inplace= True)
-                    finale_df_  =  Merger(data,df_max_poids,  product_id, class_id)   #, _ = finale_merge(data, df_finale, product_id, class_id)  
+                    finale_df_  =  Merger(data,df_max_poids,  product_id, class_id)    
                     with tab4 :
                         st.subheader("Cases of inequality")
                         finale_df_.loc[:, "Evaluation"]= True
@@ -188,7 +187,7 @@ def app():
                     # Récupérer les lignes correspondantes
                     df_max_poids1 = df_nequa_1.loc[max_poids_index1]
                     df_max_poids1.drop(["COUNTRY_KEY"], axis = 1, inplace= True)
-                    finale_df_1  =  Merger(data,df_max_poids1,  product_id, class_id)
+                    finale_df_1  =  ajout_simularite(Merger(data,df_max_poids1,  product_id, class_id))
                     finale_df_1.loc[:, "Evaluation"]= True
                     finale_df_1 = st.data_editor(finale_df_1)
                     csv_f = convert_df(finale_df_1)

query

@@ -1 +0,0 @@
-docker

start

@@ -1 +0,0 @@
-docker

stop

@@ -1 +0,0 @@
-docker