Chatterbot: how can i import data from .txt file. and train that data.

Created on 2 Nov 2017 · 24Comments · Source: gunthercox/ChatterBot

from chatterbot.trainers import ListTrainer
from chatterbot import ChatBot

f = open('dataq2.txt','r').read()

chatbot = ChatBot(
    "Terminal",
    storage_adapter="chatterbot.storage.SQLStorageAdapter", #allows the chat bot to connect to SQL databases
    input_adapter="chatterbot.input.TerminalAdapter", #allows a user to type into their terminal to communicate with the chat bot.
    output_adapter="chatterbot.output.TerminalAdapter", # print chatbot responce
    database="../database.db"   # specify the path to the database that the chat bot will use
)

chatbot.set_trainer(ListTrainer)
chatbot.train(f)


print("Type your question here...")
while True:
    try:
        chatbot_input = chatbot.get_response(None)
    # Press ctrl-c or ctrl-d to exit
    except (KeyboardInterrupt, EOFError, SystemExit):
        break

this is my code. it is running but not giving proper answer (which is stored in dataq2.txt ) it do not train data which is in dataq2.txt file. or may be it trains that but can not store that.
dataq2.txt have QA data. what i want is if I write 1st line than it should reply line 2.

dataq2 is in the form of this

Q:hello
A:hi
Q:who
A:girl 
or
Q:
hello
A:
hi
Q:
who
A:
girl

here if i input hello than it should reply hi.
please help me

answered question

Source

PoojaPatel05

Most helpful comment

Here is your solution

from chatterbot.trainers import ListTrainer
from chatterbot import ChatBot

f = open('dataq2.txt','r')

train_data = []

for line in f:
    m = re.search('(Q:|A:)?(.+)', line)
    if m:
        train_data.append(m.groups()[1])

chatbot = ChatBot(
    "Terminal",
    storage_adapter="chatterbot.storage.SQLStorageAdapter", #allows the chat bot to connect to SQL databases
    input_adapter="chatterbot.input.TerminalAdapter", #allows a user to type into their terminal to communicate with the chat bot.
    output_adapter="chatterbot.output.TerminalAdapter", # print chatbot responce
    database="../database.db"   # specify the path to the database that the chat bot will use
)

chatbot.set_trainer(ListTrainer)
chatbot.train(train_data)


print("Type your question here...")
while True:
    try:
        chatbot_input = chatbot.get_response(None)
    # Press ctrl-c or ctrl-d to exit
    except (KeyboardInterrupt, EOFError, SystemExit):
        break

vkosuri on 3 Nov 2017

👍4 😄2 ❤1

All 24 comments

@PoojaPatel05 Yes you could train your bot with some small modifications.

from chatterbot.trainers import ListTrainer
from chatterbot import ChatBot

f = open('dataq2.txt','r').read()

train_data = []

for line in f:
    train_data.append(re.match('(Q:|A:)?(.+)', line).group()[1])

chatbot = ChatBot(
    "Terminal",
    storage_adapter="chatterbot.storage.SQLStorageAdapter", #allows the chat bot to connect to SQL databases
    input_adapter="chatterbot.input.TerminalAdapter", #allows a user to type into their terminal to communicate with the chat bot.
    output_adapter="chatterbot.output.TerminalAdapter", # print chatbot responce
    database="../database.db"   # specify the path to the database that the chat bot will use
)

chatbot.set_trainer(ListTrainer)
chatbot.train(train_data)


print("Type your question here...")
while True:
    try:
        chatbot_input = chatbot.get_response(None)
    # Press ctrl-c or ctrl-d to exit
    except (KeyboardInterrupt, EOFError, SystemExit):
        break

FYI... I haven't verified above code, let me know your inputs on this.

vkosuri on 2 Nov 2017

@PoojaPatel05 I made some changes on your script if you want you could use them.

vkosuri on 2 Nov 2017

Could you please let me know fallowing details?

which version of chatterbot are you using?
Are you using python or anaconda?

vkosuri on 2 Nov 2017

I am using anaconda. and chatterbot version is 0.7.6

PoojaPatel05 on 2 Nov 2017

Python regular expression module, More information available here https://docs.python.org/2/library/re.html,

To avoid your issue import re module at the top of the script

import re
import Chatterbot

# paste your code here

vkosuri on 2 Nov 2017

in this line:
train_data.append(re.match('(Q:|A:)?(.+)', line).group()[1])
its showing error
IndexError: string index out of range

PoojaPatel05 on 2 Nov 2017

train_data = []

for line in f:
train_data.append(re.match('(Q:|A:)?(.+)', line).group()[1])

CAN YOU EXPLAIN THIS LINES?

On Fri, Nov 3, 2017 at 10:17 AM, Pooja Patel pooja.tagove@gmail.com wrote:

hey, Please help me.
May be it have a silly solution. I dont know. but I am new in this. so
please help me

On Thu, Nov 2, 2017 at 6:41 PM, Mallikarjunarao Kosuri <
[email protected]> wrote:

Python regular expression module, More information available here
https://docs.python.org/2/library/re.html,

To avoid your import re module at the top

import reimport Chatterbot

paste your code here

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/gunthercox/ChatterBot/issues/1048#issuecomment-341416225,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AfFEgsbULng_ktoWZW5EWTjIqw2JOAmsks5syb-agaJpZM4QPlgl
.

--
Pooja Patel
Python Developer
w: www.tagove.com
https://u3476773.ct.sendgrid.net/wf/click?upn=AvVTWHpii7nfl55rPBWDlFzoSSG-2BuGg5dAeZQNDakXg-3D_ls62w2VqncErQRLLYqULEUYLnD2M4VQenauY8ZodonLGqFgeX4dvd6ZyrtANiPzAcOslZAPxp-2F98z04-2BFw1-2BtZloOEzXXgh7hLucUKp4pnbAGbkOIpKJ3gPrbhlBLqcVU3MIE3lGgOYZNFqbW6JnRK-2FNNfefFZGcowRXklQ-2B7Say56IAV18mPZWzgvAHlCkqkWaLCLU0Dy4pveYMkZ1NBAWQEvEompfUlqJvuV3vdUY-3D

--
Pooja Patel
Python Developer
w: www.tagove.com
https://u3476773.ct.sendgrid.net/wf/click?upn=AvVTWHpii7nfl55rPBWDlFzoSSG-2BuGg5dAeZQNDakXg-3D_ls62w2VqncErQRLLYqULEUYLnD2M4VQenauY8ZodonLGqFgeX4dvd6ZyrtANiPzAcOslZAPxp-2F98z04-2BFw1-2BtZloOEzXXgh7hLucUKp4pnbAGbkOIpKJ3gPrbhlBLqcVU3MIE3lGgOYZNFqbW6JnRK-2FNNfefFZGcowRXklQ-2B7Say56IAV18mPZWzgvAHlCkqkWaLCLU0Dy4pveYMkZ1NBAWQEvEompfUlqJvuV3vdUY-3D

PoojaPatel05 on 3 Nov 2017

Here is your solution

from chatterbot.trainers import ListTrainer
from chatterbot import ChatBot

f = open('dataq2.txt','r')

train_data = []

for line in f:
    m = re.search('(Q:|A:)?(.+)', line)
    if m:
        train_data.append(m.groups()[1])

chatbot = ChatBot(
    "Terminal",
    storage_adapter="chatterbot.storage.SQLStorageAdapter", #allows the chat bot to connect to SQL databases
    input_adapter="chatterbot.input.TerminalAdapter", #allows a user to type into their terminal to communicate with the chat bot.
    output_adapter="chatterbot.output.TerminalAdapter", # print chatbot responce
    database="../database.db"   # specify the path to the database that the chat bot will use
)

chatbot.set_trainer(ListTrainer)
chatbot.train(train_data)


print("Type your question here...")
while True:
    try:
        chatbot_input = chatbot.get_response(None)
    # Press ctrl-c or ctrl-d to exit
    except (KeyboardInterrupt, EOFError, SystemExit):
        break

vkosuri on 3 Nov 2017

👍4 😄2 ❤1

Thank you so much.
It works. Thanks alot

On Fri, Nov 3, 2017 at 12:34 PM, Mallikarjunarao Kosuri <
[email protected]> wrote:

Here is your solution

from chatterbot.trainers import ListTrainerfrom chatterbot import ChatBot

f = open('test_data.txt','r')

train_data = []
for line in f:
# print line
m = re.search('(Q:|A:)?(.+)', line)
if m:
train_data.append(m.groups()[1])

chatbot = ChatBot(
"Terminal",
storage_adapter="chatterbot.storage.SQLStorageAdapter", #allows the chat bot to connect to SQL databases
input_adapter="chatterbot.input.TerminalAdapter", #allows a user to type into their terminal to communicate with the chat bot.
output_adapter="chatterbot.output.TerminalAdapter", # print chatbot responce
database="../database.db" # specify the path to the database that the chat bot will use
)

chatbot.set_trainer(ListTrainer)
chatbot.train(train_data)

print("Type your question here...")while True:
try:
chatbot_input = chatbot.get_response(None)
# Press ctrl-c or ctrl-d to exit
except (KeyboardInterrupt, EOFError, SystemExit):
break

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/gunthercox/ChatterBot/issues/1048#issuecomment-341633602,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AfFEgrEzf2XVkC9TumsOhiVzT6BTLJCdks5syrrjgaJpZM4QPlgl
.

PoojaPatel05 on 3 Nov 2017

🎉1

It is giving answer which is below the line of question. But what if answer
is in paragraph?

On Fri, Nov 3, 2017 at 4:00 PM, Mallikarjunarao Kosuri <
[email protected]> wrote:

Closed #1048 https://github.com/gunthercox/ChatterBot/issues/1048.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/gunthercox/ChatterBot/issues/1048#event-1324604391,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AfFEgo_I2l_l3mHJ9VVSEewBfowiODZ7ks5syusygaJpZM4QPlgl
.

PoojaPatel05 on 4 Nov 2017

@poojapatel05 can i have your trainer txt file

crynomore on 3 Feb 2018

i am using this python code

from chatterbot.trainers import ListTrainer

from chatterbot import ChatBot

bot = ChatBot('Test')

conv = open('chats.txt','r').readlines()

bot.set_trainer(ListTrainer)

bot.train(conv)

while True:
    request = input('You: ')
    response = bot.get_response(request)

    print('Bot: ',response)

i want a trainer file
can anyone help me

crynomore on 3 Feb 2018

I think are you looking this type example https://github.com/gunthercox/ChatterBot/issues/1048#issuecomment-341633602

vkosuri on 4 Feb 2018

👍1

@PoojaPatel05 can i have your trainer txt file

@beingnvn pooja want to say that, if we have question and answer with single line, then it working fine.
but what if we have answer of multiple lines(paragraphed answer), that time it gives error like:
No handlers could be found for logger "chatterbot.storage.storage_adapter"

sanjaytagove on 24 Sep 2018

Just a guess, try using three inverted commas
Ex: """Hi"""

crynomore on 25 Sep 2018

Just a guess, try using three inverted commas
Ex: """Hi"""

@beingnvn, is there any character limits in dialog statement for question and answer?
if we pass shorter character length then its okay, but it gives error while we putting long character stream.

sanjaytagove on 25 Sep 2018

Yes, https://github.com/gunthercox/ChatterBot/blob/4b5ae262bab0bc0c83555d39400049f20aaca9cd/chatterbot/constants.py

vkosuri on 26 Sep 2018

@vkosuri got your point.
but I have a question, by default it set to 400 characters, but it can train above 400 characters and up to 1011 characters. how this is possible?

sanjaytagove on 27 Sep 2018

import constants.py file and override the required property in your machine.

vkosuri on 28 Sep 2018

after putting:

from chatterbot import constants
constants.STATEMENT_TEXT_MAX_LENGTH = 2000;

it still not train large text statement.

sanjaytagove on 28 Sep 2018

Can someone update that code to the lastest version of ChatterBot?. I tried it, but it doesn't seem to work properly.
Here's the code so far (i changed 1 to 0 in the append because it kept giving me an index error)

from chatterbot import ChatBot
from chatterbot.trainers import ListTrainer
from chatterbot.trainers import ChatterBotCorpusTrainer
import re



f=open('FIA.txt','r+')

train_data=[]

for line in f:
    m=re.search('(Q:|A:)?(.+)',line)
    if m:
        train_data.append(m.group()[0])



chatb=ChatBot("EDI",
    storage_adapter="chatterbot.storage.SQLStorageAdapter",
    input_adapter="chatterbot.input.TerminalAdapter",
    output_adapter="chatterbot.output.TerminalAdapter",
    database="../database.db"
    )

trainerCorpus=ChatterBotCorpusTrainer(chatb)
trainerCorpus.train(
    "chatterbot.corpus.spanish"
)

trainer=ListTrainer(chatb)

trainer.train(
    train_data

)

print("Buenas")
while True:
    try:
        chatbot_input=chatb.get_response(input('Q:'))
        print('A:',chatbot_input)
    except(KeyboardInterrupt,EOFError,SystemExit):
        break

Alvarden15 on 27 May 2019

Nevermind, i found a solution. Although i'm using a .txt file; instead it's with a .yaml file.

Alvarden15 on 29 May 2019

Here is your solution

from chatterbot.trainers import ListTrainer
from chatterbot import ChatBot

f = open('dataq2.txt','r')

train_data = []

for line in f:
    m = re.search('(Q:|A:)?(.+)', line)
    if m:
        train_data.append(m.groups()[1])

chatbot = ChatBot(
    "Terminal",
    storage_adapter="chatterbot.storage.SQLStorageAdapter", #allows the chat bot to connect to SQL databases
    input_adapter="chatterbot.input.TerminalAdapter", #allows a user to type into their terminal to communicate with the chat bot.
    output_adapter="chatterbot.output.TerminalAdapter", # print chatbot responce
    database="../database.db"   # specify the path to the database that the chat bot will use
)

chatbot.set_trainer(ListTrainer)
chatbot.train(train_data)


print("Type your question here...")
while True:
    try:
        chatbot_input = chatbot.get_response(None)
    # Press ctrl-c or ctrl-d to exit
    except (KeyboardInterrupt, EOFError, SystemExit):
        break

I got this error:
UnicodeDecodeError: 'charmap' codec can't decode byte 0x9d in position 7159: character maps to