feat: generated files

app.py

@@ -3,6 +3,8 @@ import pandas as pd
 import plotly.express as px
 import matplotlib.pyplot as plt
 import numpy as np
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
 
 # Page configuration
 st.set_page_config(page_title="Customer Insights App", page_icon=":bar_chart:")
@@ -16,6 +18,11 @@ euros_proveedor = pd.read_csv("euros_proveedor.csv", sep=',')
 df['CLIENTE'] = df['CLIENTE'].astype(str)
 nombres_proveedores['codigo'] = nombres_proveedores['codigo'].astype(str)
 euros_proveedor['CLIENTE'] = euros_proveedor['CLIENTE'].astype(str)
+fieles_df = pd.read_csv("clientes_relevantes.csv")
+# Cargo csv del histórico de cestas
+cestas = pd.read_csv("cestas.csv")
+# Cargo csv de productos y descripcion
+productos = pd.read_csv("productos.csv")
 
 # Convert all columns except 'CLIENTE' to float in euros_proveedor
 for col in euros_proveedor.columns:
@@ -87,7 +94,7 @@ st.markdown("""
 """)
 
 # Navigation menu
-page = st.selectbox("Select the tool you want to use", ["", "Customer Analysis", "Basket Recommendations"])
+page = st.selectbox("Select the tool you want to use", ["", "Customer Analysis", "Articles Recommendations"])
 
 # Home Page
 if page == "":
@@ -188,11 +195,14 @@ elif page == "Customer Analysis":
             st.warning(f"No data found for customer {customer_code}. Please check the code.")
 
 # Customer Recommendations Page
-elif page == "Basket Recommendations":
-    st.title("Basket Recommendations")
-    st.markdown("""Get tailored product recommendations for your customers based on their purchasing basket.""")
+elif page == "Articles Recommendations":
+    st.title("Articles Recommendations")
 
-    # Use the same customer search as in Customer Analysis
+    st.markdown("""
+        Get tailored recommendations for your customers based on their basket.
+    """)
+
+    # Campo input para cliente
     partial_code = st.text_input("Enter part of Customer Code for Recommendations (or leave empty to see all)")
     if partial_code:
         filtered_customers = df[df['CLIENTE'].str.contains(partial_code)]
@@ -201,30 +211,122 @@ elif page == "Basket Recommendations":
     customer_list = filtered_customers['CLIENTE'].unique()
     customer_code = st.selectbox("Select Customer Code for Recommendations", customer_list)
 
-    # Basket input section
-    if customer_code:
-        st.write(f"### Purchase Basket for Customer {customer_code}")
+    # DEfinicion de la funcion recomienda
+    def recomienda(new_basket):
+        # Calcular la matriz TF-IDF
+        tfidf = TfidfVectorizer()
+        tfidf_matrix = tfidf.fit_transform(cestas['Cestas'])
 
-        # Create input fields for article codes and units
-        basket_items = []
-        num_lines = st.number_input("How many lines do you want to add?", min_value=1, value=1, step=1)
-        
-        for i in range(num_lines):
-            col1, col2 = st.columns(2)
-            with col1:
-                article_code = st.text_input(f"Enter Article Code for Line {i + 1}")
-            with col2:
-                units = st.number_input(f"Enter Units for Line {i + 1}", min_value=1, value=1, step=1)
+        # Convertir la nueva cesta en formato TF-IDF
+        new_basket_str = ' '.join(new_basket)
+        new_basket_tfidf = tfidf.transform([new_basket_str])
 
-            # Add each article code and units to the list
-            if article_code and units:
-                basket_items.append((article_code, units))
+        # Comparar la nueva cesta con las anteriores
+        similarities = cosine_similarity(new_basket_tfidf, tfidf_matrix)
 
-        # Generate the concatenated string
-        input_cesta = ''.join([f"{code * units}" for code, units in basket_items])
 
-        # Show the result
-        st.write(f"Generated input_cesta: {input_cesta}")
+        # Obtener los índices de las cestas más similares
+        similar_indices = similarities.argsort()[0][-3:]  # Las 3 más similares
 
-        # Placeholder for passing the input_cesta to the model
-        st.write("Model is not yet implemented, but input_cesta is ready.")
+        # Crear un diccionario para contar las recomendaciones
+        recommendations_count = {}
+        total_similarity = 0
+
+        # Recomendar productos de cestas similares
+        for idx in similar_indices:
+            sim_score = similarities[0][idx]
+            total_similarity += sim_score
+            products = cestas.iloc[idx]['Cestas'].split()
+        
+        for product in products:
+            if product.strip() not in new_basket:  # Evitar recomendar lo que ya está en la cesta
+                if product.strip() in recommendations_count:
+                    recommendations_count[product.strip()] += sim_score
+                else:
+                    recommendations_count[product.strip()] = sim_score
+
+        # Calcular la probabilidad relativa de cada producto recomendado
+        recommendations_with_prob = []
+        if total_similarity > 0:  # Verificar que total_similarity no sea cero
+            recommendations_with_prob = [(product, score / total_similarity) for product, score in recommendations_count.items()]
+        else:
+            print("No se encontraron similitudes suficientes para calcular probabilidades.")
+
+        recommendations_with_prob.sort(key=lambda x: x[1], reverse=True)  # Ordenar por puntuación
+
+        # Crear un nuevo DataFrame para almacenar las recomendaciones con descripciones y probabilidades
+        recommendations_df = pd.DataFrame(columns=['ARTICULO', 'DESCRIPCION', 'PROBABILIDAD'])
+
+        # Agregar las recomendaciones al DataFrame usando pd.concat
+        for product, prob in recommendations_with_prob:
+        # Buscar la descripción en el DataFrame de productos
+            description = productos.loc[productos['ARTICULO'] == product, 'DESCRIPCION']
+            if not description.empty:
+            # Crear un nuevo DataFrame temporal para la recomendación
+                    temp_df = pd.DataFrame({
+                        'ARTICULO': [product],
+                        'DESCRIPCION': [description.values[0]],  # Obtener el primer valor encontrado
+                        'PROBABILIDAD': [prob]
+                     })
+            # Concatenar el DataFrame temporal al DataFrame de recomendaciones
+            recommendations_df = pd.concat([recommendations_df, temp_df], ignore_index=True)
+
+            return recommendations_df   
+
+    # Comprobar si el cliente está en el CSV de fieles
+    
+    is_fiel = customer_code in fieles_df['CLIENTE'].astype(str).values
+
+    if customer_code:
+        if is_fiel:
+            st.write(f"### Customer {customer_code} is a loyal customer.")
+            option = st.selectbox("Select Recommendation Type", ["Select an option", "By Purchase History", "By Current Basket"])
+
+            if option == "By Purchase History":
+                st.warning("Option not available... aún")
+            elif option == "By Current Basket": 
+                    
+                st.write("Enter the items in the basket:")
+                
+                # Input para los artículos y unidades
+                items = st.text_input("Enter items (comma-separated):").split(',')
+                quantities = st.text_input("Enter quantities (comma-separated):").split(',')
+
+                # Crear una lista de artículos basada en la entrada
+                new_basket = [item.strip() for item in items]
+
+                # Asegurarse de que las longitudes de artículos y cantidades coincidan
+                if len(new_basket) == len(quantities):
+                    # Procesar la lista para recomendar
+                    recommendations_df = recomienda(new_basket)
+
+                    if not recommendations_df.empty:
+                        st.write("### Recommendations based on the current basket:")
+                        st.dataframe(recommendations_df)
+                    else:
+                        st.warning("No recommendations found for the provided basket.")
+                else:
+                    st.warning("The number of items must match the number of quantities.")
+        else:
+            st.write(f"### Customer {customer_code} is not a loyal customer.")
+            st.write("Recommendation based on the basket. Please enter the items:")
+            
+            # Input para los artículos y unidades
+            items = st.text_input("Enter items (comma-separated):").split(',')
+            quantities = st.text_input("Enter quantities (comma-separated):").split(',')
+
+            # Crear una lista de artículos basada en la entrada
+            new_basket = [item.strip() for item in items]
+
+            # Asegurarse de que las longitudes de artículos y cantidades coincidan
+            if len(new_basket) == len(quantities):
+                # Procesar la lista para recomendar
+                recommendations_df = recomienda(new_basket)
+
+                if not recommendations_df.empty:
+                    st.write("### Recommendations based on the current basket:")
+                    st.dataframe(recommendations_df)
+                else:
+                    st.warning("No recommendations found for the provided basket.")
+            else:
+                st.warning("The number of items must match the number of quantities.")

clientes_relevantes.csv

@@ -0,0 +1,684 @@
+,Cliente
+0,1019
+5,1161
+6,1169
+7,1177
+8,1193
+11,1218
+12,1232
+16,1348
+19,1381
+20,1403
+22,1421
+23,1447
+25,1486
+28,1523
+30,1575
+32,1597
+35,1630
+39,1665
+41,1692
+43,1853
+47,2098
+49,2132
+52,2190
+53,2192
+54,2204
+55,2247
+56,2252
+57,2267
+59,2279
+60,2308
+62,2484
+63,3001
+64,3003
+65,3016
+66,4350
+67,5018
+68,5022
+69,5037
+70,5295
+72,6000
+74,6062
+75,6077
+76,6079
+77,6089
+78,6110
+87,6503
+88,6506
+90,6515
+94,6559
+96,6588
+97,6593
+100,7034
+101,7035
+103,7073
+105,7136
+106,7141
+107,7179
+108,7219
+109,7243
+110,7335
+111,7340
+112,7451
+118,7593
+121,7622
+122,7629
+123,7667
+124,7701
+125,7702
+127,7763
+129,7791
+131,7806
+132,8060
+133,8063
+136,8222
+137,8227
+139,8955
+141,10018
+145,12197
+147,12566
+149,12595
+150,12612
+151,13244
+152,14023
+153,14030
+155,14064
+156,14079
+158,14104
+159,14165
+161,14177
+162,14182
+166,78020
+170,78028
+173,78050
+174,78065
+181,78102
+184,78116
+185,78122
+187,78129
+189,78155
+191,78177
+193,78195
+195,78208
+196,78218
+198,78222
+200,78229
+202,78248
+203,78253
+205,78272
+206,78276
+207,78283
+210,78300
+213,78332
+214,78335
+222,78387
+224,78398
+225,78401
+231,78450
+232,78453
+233,78463
+234,78469
+237,78486
+239,78491
+244,78539
+252,78584
+253,78585
+256,78591
+257,78601
+260,78611
+267,78664
+270,78673
+271,78679
+276,78704
+281,78722
+283,78725
+291,78739
+294,78743
+295,78747
+296,78748
+299,78756
+300,78762
+301,78763
+304,78773
+309,78796
+310,78799
+311,78802
+312,78803
+315,78817
+317,78827
+318,78830
+319,78832
+326,78858
+327,78868
+328,78872
+329,78879
+334,78901
+338,78916
+343,78942
+347,78951
+348,78953
+351,78969
+355,78987
+357,78991
+358,79001
+359,79017
+360,79019
+361,79023
+362,79024
+363,79031
+365,79037
+369,79071
+371,79081
+373,79086
+374,79090
+376,79093
+377,79095
+379,79101
+382,79113
+383,79114
+385,79117
+386,79127
+387,79129
+389,79133
+392,79145
+393,79152
+394,79160
+395,79164
+400,79177
+401,79187
+402,79188
+403,79192
+404,79193
+405,79198
+408,79205
+409,79210
+410,79211
+411,79217
+412,79221
+415,79237
+417,79266
+420,79280
+421,79284
+422,79297
+423,79316
+424,79324
+425,79325
+426,79328
+427,79331
+428,79340
+430,79354
+431,79355
+433,79361
+434,79362
+436,79378
+444,79426
+445,79435
+447,79451
+448,79462
+449,79465
+450,79466
+451,79467
+452,79477
+453,79480
+454,79483
+455,79486
+456,79488
+458,79495
+459,79497
+463,79518
+465,79529
+466,79531
+471,79549
+472,79550
+473,79551
+476,79561
+477,79565
+478,79568
+479,79576
+480,79580
+483,79593
+484,79594
+485,79609
+486,79610
+487,79614
+489,79631
+495,79653
+496,79655
+497,79660
+498,79667
+499,79673
+500,79674
+501,79686
+502,79687
+504,79694
+506,79703
+509,79727
+513,79742
+514,79754
+515,79757
+516,79762
+517,79764
+519,79769
+522,79775
+524,79786
+526,79789
+528,79803
+529,79807
+530,79812
+531,79819
+532,79827
+535,79843
+536,79844
+537,79854
+538,79859
+539,79863
+543,79907
+544,79908
+545,79909
+546,79915
+547,79920
+548,79921
+549,79931
+550,79932
+551,79933
+556,79953
+557,79965
+558,79981
+559,79991
+561,80018
+562,80022
+563,80024
+564,80026
+565,80027
+566,80030
+567,80033
+568,80036
+569,80038
+570,80039
+571,80042
+572,80044
+573,80045
+574,80048
+575,80058
+580,80087
+581,80095
+583,80106
+584,80108
+585,80123
+586,80129
+588,80139
+590,80147
+592,80156
+595,80164
+599,80180
+600,80183
+604,80197
+605,80202
+607,80210
+610,80225
+611,80231
+613,80243
+614,80249
+619,80259
+622,80287
+624,80298
+625,80301
+626,80303
+627,80307
+628,80308
+629,80310
+630,80312
+631,80315
+632,80317
+633,80318
+640,80361
+641,80363
+642,80375
+643,80377
+645,80380
+646,80389
+647,80390
+648,80392
+649,80393
+650,80396
+651,80401
+652,80404
+653,80405
+656,80418
+657,80419
+658,80429
+660,80440
+663,80445
+664,80447
+666,80453
+667,80455
+669,80460
+671,80468
+672,80471
+675,80479
+677,80489
+678,80495
+681,80511
+682,80512
+684,80520
+685,80524
+686,80532
+687,80538
+688,80540
+690,80544
+691,80546
+692,80555
+694,80561
+696,80570
+700,80582
+701,80584
+702,80586
+705,80604
+708,80612
+709,80615
+715,80637
+718,80661
+721,80680
+723,80689
+724,80699
+727,80725
+730,80734
+732,80739
+733,80741
+734,80747
+735,80751
+736,80753
+741,80801
+744,80828
+746,80852
+751,80867
+752,80871
+753,80872
+754,80873
+755,80875
+756,80877
+757,80881
+758,80886
+759,80894
+760,80897
+761,80911
+762,80912
+764,80923
+765,80924
+766,80930
+772,80952
+773,80955
+774,80963
+778,80979
+779,80980
+780,80988
+781,80992
+782,81003
+783,81014
+788,81045
+791,81077
+796,81107
+797,81110
+798,81121
+802,81135
+804,81140
+805,81144
+806,81149
+807,81150
+808,81151
+810,81160
+811,81162
+814,81173
+815,81178
+816,81188
+818,81205
+819,81207
+820,81208
+827,81239
+830,81253
+834,81275
+835,81280
+840,81295
+847,81312
+850,81315
+851,81323
+852,81325
+855,81339
+857,81349
+859,81363
+863,81380
+864,81389
+865,81390
+870,81415
+871,81418
+874,81430
+876,81439
+877,81441
+882,81461
+883,81465
+884,81473
+885,81474
+887,81476
+889,81487
+901,81533
+914,81569
+915,81570
+921,81599
+922,81602
+923,81604
+941,81765
+956,81981
+957,81982
+958,81991
+961,82019
+962,82024
+965,82043
+969,82092
+970,82096
+971,82104
+978,82152
+979,82157
+983,82180
+985,82191
+987,82219
+994,82247
+997,82302
+1001,82345
+1003,82360
+1012,82444
+1016,82466
+1021,82483
+1022,82487
+1023,82500
+1024,82507
+1025,82516
+1026,82521
+1028,82558
+1029,82565
+1032,82588
+1033,82589
+1043,82648
+1045,82694
+1046,82703
+1049,82715
+1051,82731
+1052,82737
+1054,82743
+1055,82763
+1056,82764
+1057,82768
+1061,82846
+1062,82848
+1063,82850
+1067,82891
+1073,82928
+1075,82944
+1079,83002
+1080,83023
+1081,83049
+1085,83080
+1095,83157
+1096,83168
+1099,83185
+1101,83192
+1102,83195
+1106,83238
+1107,83244
+1110,83254
+1114,83276
+1115,83278
+1120,83300
+1123,83319
+1124,83335
+1126,83340
+1127,83352
+1132,83380
+1139,83434
+1142,83451
+1143,83452
+1145,83473
+1149,83500
+1150,83502
+1151,83513
+1154,83529
+1156,83551
+1159,83566
+1166,83605
+1168,83613
+1169,83615
+1172,83624
+1174,83640
+1177,83665
+1181,83677
+1183,83694
+1184,83696
+1186,83703
+1199,83786
+1208,83906
+1211,83919
+1216,83968
+1217,83971
+1219,83984
+1223,84000
+1224,84001
+1226,84008
+1227,84014
+1232,84117
+1235,84140
+1239,84174
+1241,84177
+1244,84188
+1246,84203
+1251,84284
+1252,84297
+1256,84349
+1258,84364
+1260,84385
+1261,84386
+1266,84412
+1267,84435
+1271,84489
+1277,84537
+1284,84598
+1285,84599
+1287,84611
+1288,84619
+1289,84620
+1290,84623
+1291,84640
+1294,84648
+1295,84654
+1297,84679
+1303,84717
+1307,84781
+1313,84840
+1315,84845
+1316,84851
+1320,84883
+1327,84933
+1333,84969
+1335,84988
+1336,84992
+1341,85015
+1346,85028
+1350,85040
+1358,85085
+1360,85088
+1362,85099
+1365,85103
+1367,85106
+1369,85138
+1370,85139
+1372,85144
+1374,85150
+1383,85177
+1385,85179
+1386,85190
+1397,85276
+1400,85287
+1404,85305
+1405,85309
+1409,85333
+1411,85338
+1415,85348
+1418,85367
+1419,85368
+1426,85398
+1432,85430
+1445,85490
+1449,85508
+1457,85552
+1461,85563
+1463,85565
+1467,85582
+1490,85652
+1496,85668
+1498,85675
+1501,85680
+1502,85687
+1507,85711
+1510,85722
+1512,85729
+1516,85733
+1520,85737
+1531,85751
+1535,85755
+1575,85800
+1581,85809
+1595,85823
+1606,85834
+1617,85846
+1626,85857
+1671,85913
+1681,85925
+1689,85934
+1711,85962
+1723,85974
+1732,85986
+1733,85987
+1754,86012
+1771,86031
+1777,86040
+1791,86059
+1797,86067
+1804,86077
+1832,86110
+1851,86134
+1852,86135
+1869,86153
+1897,86184
+1903,86191
+1917,86205
+1954,90251
+1957,90742
+1958,90792
+1961,90876
+1965,91863
+1966,91876
+1967,91883
+1968,91884
+1971,91905
+1973,97148
+1974,98018
+1977,98317