trained model takes longer to get a response

[Karamya]

I trained my corpus with storage_adapter= SQLStorageAdapter and database= database.sqlite3. The file size of the sqlite3 database is about 45 MB, which I got after about one week of training using a corpus size of ~78 MB. I am trying to get a response from the trained model and it is almost one and half hours and still I didn't get any response. Here is the code for training my data based on a corpus in the folder /data

from chatterbot.trainers import ChatterBotCorpusTrainer, Trainer
from chatterbot import ChatBot

bot = ChatBot(
    "Remedy",
    storage_adapter='chatterbot.storage.SQLStorageAdapter',
    #input_adapter = 'chatterbot.input.TerminalAdapter',
    #output_adapter = 'chatterbot.output.TerminalAdapter',
    logic_adapters=[
        {
            "import_path": "chatterbot.logic.BestMatch",
            "statement_comparison_function": "chatterbot.comparisons.levenshtein_distance",
            "response_selection_method": "chatterbot.response_selection.get_first_response"
        },
{
            'import_path': 'chatterbot.logic.LowConfidenceAdapter',
            'threshold': 0.65,
            'default_response': 'I am sorry, but I do not understand.'
        }

    ],
    database = './database.sqlite3'
)

bot.set_trainer(ChatterBotCorpusTrainer)
bot.train('chatterbot.corpus.german', './data/') 

After training my corpus, I used the following code to get a response,

from chatterbot import ChatBot
from chatterbot.trainers import ChatterBotCorpusTrainer, Trainer
bot = ChatBot("Remedy",
    storage_adapter='chatterbot.storage.SQLStorageAdapter',
    #input_adapter = 'chatterbot.input.TerminalAdapter',
    #output_adapter = 'chatterbot.output.TerminalAdapter',
    logic_adapters=[
        {
            "import_path": "chatterbot.logic.BestMatch",
            "statement_comparison_function": "chatterbot.comparisons.levenshtein_distance",
            "response_selection_method": "chatterbot.response_selection.get_first_response"
        },
{
            'import_path': 'chatterbot.logic.LowConfidenceAdapter',
            'threshold': 0.65,
            'default_response': 'I am sorry, but I do not understand.'
        }

    ],
    database = './database.sqlite3', read_only = True)
print(bot.get_response('Ich arbeite an einem Projekt'))

Can someone figure out the problem, why it takes really long to get a response (already waited close to 2 hours).

[vkosuri]

Known issue is going address by this feature #925

[ghost]

Does changing the database away from sqlite3 help the efficiency here? I trained the bot on 10,000 movie scripts and it's been hours since a response with a simple "hello"

[gunthercox]

@kyle-sorensen The database type won't effect the efficiency on any noticeable level for large datasets.

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.