| import streamlit as st #line:1:import streamlit as st | |
| from bokeh .models .widgets import Button #line:2:from bokeh.models.widgets import Button | |
| from bokeh .models import CustomJS #line:3:from bokeh.models import CustomJS | |
| from streamlit_bokeh_events import streamlit_bokeh_events #line:4:from streamlit_bokeh_events import streamlit_bokeh_events | |
| import pandas as pd #line:6:import pandas as pd | |
| import numpy as np #line:7:import numpy as np | |
| from nltk .stem import SnowballStemmer #line:8:from nltk.stem import SnowballStemmer | |
| from inflector import Inflector ,English ,Spanish #line:9:from inflector import Inflector, English, Spanish | |
| inflector =Inflector (Spanish )#line:10:inflector = Inflector(Spanish) | |
| import unidecode #line:11:import unidecode | |
| import unicodedata #line:12:import unicodedata | |
| n =1 #line:13:n=1 | |
| p =0 #line:14:p=0 | |
| query="none" | |
| df =pd .read_excel ('preguntas_qh_tags2.xlsx',engine ="openpyxl")#line:15:df = pd.read_excel('preguntas_qh_tags2.xlsx',engine="openpyxl") | |
| def remove_accents (O0O00OO0O000OOO0O ):#line:17:def remove_accents(input_str): | |
| O000O0OO000000OOO =unicodedata .normalize ('NFKD',O0O00OO0O000OOO0O )#line:18:nfkd_form = unicodedata.normalize('NFKD', input_str) | |
| return u"".join ([O0O000O00O00O00OO for O0O000O00O00O00OO in O000O0OO000000OOO if not unicodedata .combining (O0O000O00O00O00OO )])#line:19:return u"".join([c for c in nfkd_form if not unicodedata.combining(c)]) | |
| stt_button =Button (label ="Speak")#line:25:stt_button = Button(label="Speak")#, width=100) | |
| stt_button .js_on_event ("button_click",CustomJS (code =""" | |
| var recognition = new webkitSpeechRecognition(); | |
| recognition.lang = "es-ES"; | |
| recognition.continuous = true; | |
| recognition.interimResults = true; | |
| recognition.onresult = function (e) { | |
| var value = ""; | |
| for (var i = e.resultIndex; i < e.results.length; ++i) { | |
| if (e.results[i].isFinal) { | |
| value += e.results[i][0].transcript; | |
| } | |
| } | |
| if ( value != "") { | |
| document.dispatchEvent(new CustomEvent("GET_TEXT", {detail: value})); | |
| } | |
| } | |
| recognition.start(); | |
| """))#line:47:""")) | |
| result =streamlit_bokeh_events (stt_button ,events ="GET_TEXT",key ="listen",refresh_on_update =False ,override_height =75 ,debounce_time =0 )#line:56:debounce_time=0) | |
| placeholder =st .empty ()#line:59:placeholder = st.empty() | |
| placeholder .text ("Escuchando...")#line:62:placeholder.text("Escuchando...") | |
| if result :#line:66:if result: | |
| if "GET_TEXT"in result :#line:67:if "GET_TEXT" in result: | |
| placeholder .text ("Entendido!")#line:69:placeholder.text("Entendido!") | |
| st .text ("Entendido!")#line:70:st.text("Entendido!") | |
| placeholder .empty ()#line:72:placeholder.empty() | |
| #st .write (result .get ("GET_TEXT"))#line:74:st.write(result.get("GET_TEXT")) | |
| query =result .get ("GET_TEXT")#line:80:query=result.get("GET_TEXT") | |
| if query !="none":#line:86:if query!="salir": | |
| #st .text ("query: "+query )#line:81:st.text("query: "+query) | |
| query =query .lower ()#line:82:query = query.lower() | |
| if query !="salir":#line:87:if query!="none": | |
| #st .text ("query: "+query )#line:89:st.text("query: "+query) | |
| result =query .split ()#line:90:result=query.split() | |
| #st .text ("query: "+query )#line:91:st.text("query: "+query) | |
| df2 =[]#line:92:df2=[] | |
| for index ,row in df .iterrows ():#line:93:for index,row in df.iterrows(): | |
| list_words =df .loc [index ,'TAGS2']#line:96:list_words=df.loc[index,'TAGS2']#busca columna tag2 | |
| list_words =list_words .split (",")#line:97:list_words = list_words.split(",") | |
| df2 .append (list_words )#line:98:df2.append(list_words) | |
| ls5 =[]#line:103:ls5=[] | |
| list_words =[]#line:104:list_words=[] | |
| for indexw ,word in enumerate (result ):#line:105:for indexw, word in enumerate(result): | |
| result [indexw ]=inflector .singularize (str (word ))#line:106:result[indexw]=inflector.singularize(str(word)) | |
| #st .text ("result:"+str (result ))#line:107:st.text("result:" + str(result)) | |
| count_words =np .zeros (len (df .index ),dtype =int )#line:108:count_words = np.zeros(len(df.index), dtype=int) | |
| for index1 ,row in enumerate (df2 ):#line:109:for index1,row in enumerate(df2): | |
| ls4 =[]#line:112:ls4=[] | |
| for word in row :#line:123:for word in row : | |
| if word !=[]:#line:126:if word!=[]: | |
| for num in range (100 ):#line:128:for num in range (100): | |
| num =num /10 #line:130:num=num/10 | |
| num =str (num )#line:131:num=str(num) | |
| if word ==num :#line:133:if word==num: | |
| num =num .split (".")#line:135:num = num.split(".") | |
| ls4 .append (num [0 ])#line:137:ls4.append(num[0]) | |
| ls4 .append ("con")#line:139:ls4.append("con") | |
| ls4 .append (num [1 ])#line:140:ls4.append(num[1]) | |
| p =p +1 #line:141:p=p+1 | |
| if p ==0 :#line:143:if p==0: | |
| ls4 .append (word )#line:144:ls4.append(word) | |
| p =0 #line:145:p=0 | |
| ls5 .append (ls4 )#line:146:ls5.append(ls4) | |
| #if str (remove_accents (ls5 [index1 ][0 ]))==remove_accents (result [0 ]):#line:150:if str(remove_accents(ls5[index1][0]))==remove_accents(result[0]):# and ls[1]==first_word: | |
| if n==1: | |
| for resulted in result :#line:152:for resulted in result[1:]: | |
| for index2 ,word in enumerate (ls5 [index1 ]):#line:158:for index2,word in enumerate(ls5[index1][1:]) : | |
| if str (resulted )=="exactitec"or str (resulted )=="exactite"or str (resulted )=="exactec":#line:159:if str(resulted)=="h": | |
| resulted ="exactech" | |
| if str (resulted )=="h":#line:159:if str(resulted)=="h": | |
| resulted ="dh"#line:160:resulted="dh" | |
| if str (resulted )=="colón":#line:161:if str(resulted)=="colón": | |
| resulted ="coron"#line:162:resulted="coron" | |
| if str (resulted )=="trave"or str (resulted )=="travé":#line:163:if str(resulted)=="trave" or str(resulted)=="travé": | |
| resulted ="nope"#line:164:resulted="nope" | |
| if str (resulted )=="lasersolvo"or str (resulted )=="lásersolvo"or str (resulted )=="lásersolbo"or str (resulted )=="lasersolbo":#line:165:if str(resulted)=="lasersolvo"or str(resulted)=="lásersolvo"or str(resulted)=="lásersolbo"or str(resulted)=="lasersolbo": | |
| resulted ="solvo"#line:166:resulted="solvo" | |
| if str (resulted )=="solbo":#line:167:if str(resulted)=="solbo": | |
| resulted ="solvo"#line:168:resulted="solvo" | |
| if str (resulted )=="maya":#line:169:if str(resulted)=="maya": | |
| resulted ="malla"#line:170:resulted="malla" | |
| if str (resulted )=="pilos"or str (resulted )=="pilo":#line:171:if str(resulted)=="pilos"or str(resulted)=="pilo": | |
| resulted ="philo"#line:172:resulted="philo" | |
| if str (resulted )=="filos"or str (resulted )=="filo":#line:173:if str(resulted)=="filos"or str(resulted)=="filo": | |
| resulted ="philo"#line:174:resulted="philo" | |
| if str (resulted )=="sinces"or str (resulted )=="sinc":#line:175:if str(resulted)=="sinces" or str(resulted)=="sinc": | |
| resulted ="synthe"#line:176:resulted="synthe" | |
| if str (resulted )=="sintes"or str (resulted )=="sint":#line:177:if str(resulted)=="sintes" or str(resulted)=="sint": | |
| resulted ="synthe"#line:178:resulted="synthe" | |
| if str (resulted )=="axos"or str (resulted )=="axo":#line:179:if str(resulted)=="axos" or str(resulted)=="axo": | |
| resulted ="axso"#line:180:resulted="axso" | |
| if str (resulted )=="uno":#line:181:if str(resulted)=="uno": | |
| resulted ="1"#line:182:resulted="1" | |
| if str (resulted )=="dos"or str (resulted )=="do":#line:184:if str(resulted)=="dos" or str(resulted)=="do": | |
| resulted ="2"#line:185:resulted="2" | |
| if str (resulted )=="tres"or str (resulted )=="tr":#line:187:if str(resulted)=="tres" or str(resulted)=="tr": | |
| resulted ="3"#line:188:resulted="3" | |
| if str (resulted )=="cuatro"or str (resulted )=="cuatr":#line:190:if str(resulted)=="cuatro"or str(resulted)=="cuatr": | |
| resulted ="4"#line:191:resulted="4" | |
| if str (resulted )=="cinco"or str (resulted )=="cinc":#line:193:if str(resulted)=="cinco"or str(resulted)=="cinc": | |
| resulted ="5"#line:194:resulted="5" | |
| if str (resulted )=="seis"or str (resulted )=="sei":#line:196:if str(resulted)=="seis"or str(resulted)=="sei": | |
| resulted ="6"#line:197:resulted="6" | |
| if str (resulted )=="siete":#line:199:if str(resulted)=="siete": | |
| resulted ="7"#line:200:resulted="7" | |
| if str (resulted )=="ocho"or str (resulted )=="och":#line:202:if str(resulted)=="ocho"or str(resulted)=="och": | |
| resulted ="8"#line:203:resulted="8" | |
| if str (resulted )=="nueve"or str (resulted )=="nuev":#line:205:if str(resulted)=="nueve"or str(resulted)=="nuev": | |
| resulted ="9"#line:206:resulted="9" | |
| if str (resulted )=="cero":#line:208:if str(resulted)=="cero": | |
| resulted ="0"#line:209:resulted="0" | |
| if str(resulted)=="veintiuno": | |
| resulted="21" | |
| if str(resulted)=="veinte": | |
| resulted="20" | |
| if str(resulted)=="veintidos"or str(resulted)=="veintido" : | |
| resulted="22" | |
| if str(resulted)=="veintitres" or str(resulted)=="veintitre" : | |
| resulted="23" | |
| if str(resulted)=="veinticuatro": | |
| resulted="24" | |
| if str(resulted)=="veinticinco": | |
| resulted="25" | |
| if str(resulted)=="veintiseis"or str(resulted)=="veintisei" : | |
| resulted="26" | |
| if str(resulted)=="veintisiete": | |
| resulted="27" | |
| if str(resulted)=="veintiocho": | |
| resulted="28" | |
| if str(resulted)=="veintinueve": | |
| resulted="29" | |
| if str(resulted)=="treinta": | |
| resulted="30" | |
| if str (remove_accents (word )).lower ()==str (remove_accents (resulted )).lower ():#line:213:if str(remove_accents(word)).lower() == str(remove_accents(resulted)).lower(): | |
| count_words [index1 ]=count_words [index1 ]+1 #line:216:count_words[index1]=count_words[index1]+1 | |
| ls5 [index1 ].pop (index2 )#line:217:ls5[index1].pop(index2) | |
| indexmax =np .argwhere (count_words ==np .amax (count_words ))#line:219:indexmax = np.argwhere(count_words == np.amax(count_words)) | |
| all_zeros =not np .any (indexmax )#line:220:all_zeros = not np.any(indexmax) | |
| #st .text (count_words )#line:222:st.text(count_words) | |
| if not np .all (count_words ==0 ):#line:224:if not np.all(count_words==0): | |
| for indexin in indexmax :#line:226:for indexin in indexmax: | |
| #st .text (indexin )#line:228:st.text(indexin) | |
| #st .text (count_words [indexin ])#line:229:st.text(count_words[indexin]) | |
| st .text ("Si has preguntado...\n")#line:230:st.text("Si has preguntado...\n") | |
| st .text (df .iloc [indexin ,2])#line:231:st.text(df.iloc[indexin,2]) | |
| st .text ("La respuesta es...\n")#line:232:st.text("La respuesta es...\n") | |
| st .text (df .iloc [indexin ,3])#line:233:st.text(df.iloc[indexin,3]) | |
| print ("GRACIAS!") |