Cannot get my NLTK based Chatbot run on heroku but it works locally

Question

My chatbot is working fine on my local system, but I am getting an error when I deploy it on Heroku.

2021-05-10T08:22:26.384671+00:00 heroku[router]: at=info method=POST path="/get" host=ihaci- 
chat.herokuapp.com request_id=33f6310b-fc0f-4dfb-be87-aaedaca496df fwd="203.189.245.148" dyno=web.1 
connect=0ms service=7ms status=500 bytes=463 protocol=https
2021-05-10T08:22:26.382886+00:00 app[web.1]: [2021-05-10 08:22:26,382] ERROR in app: Exception on 
/get [POST]
2021-05-10T08:22:26.382924+00:00 app[web.1]: Traceback (most recent call last):
2021-05-10T08:22:26.382925+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/flask/app.py", line 2447, in wsgi_app
2021-05-10T08:22:26.382928+00:00 app[web.1]:     response = self.full_dispatch_request()
2021-05-10T08:22:26.382929+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/flask/app.py", line 1952, in full_dispatch_request
2021-05-10T08:22:26.382929+00:00 app[web.1]:     rv = self.handle_user_exception(e)
2021-05-10T08:22:26.382930+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/flask/app.py", line 1821, in handle_user_exception
2021-05-10T08:22:26.382930+00:00 app[web.1]:     reraise(exc_type, exc_value, tb)
2021-05-10T08:22:26.382930+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/flask/_compat.py", line 39, in reraise
2021-05-10T08:22:26.382931+00:00 app[web.1]:     raise value
2021-05-10T08:22:26.382931+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/flask/app.py", line 1950, in full_dispatch_request
2021-05-10T08:22:26.382932+00:00 app[web.1]:     rv = self.dispatch_request()
2021-05-10T08:22:26.382932+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/flask/app.py", line 1936, in dispatch_request
2021-05-10T08:22:26.382933+00:00 app[web.1]:     return self.view_functions[rule.endpoint] 
(**req.view_args)
2021-05-10T08:22:26.382933+00:00 app[web.1]:   File "/app/app.py", line 55, in chatbot_response
2021-05-10T08:22:26.382934+00:00 app[web.1]:     ints = predict_class(msg, model)
2021-05-10T08:22:26.382934+00:00 app[web.1]:   File "/app/app.py", line 86, in predict_class
2021-05-10T08:22:26.382934+00:00 app[web.1]:     p = bow(sentence, words, show_details=False)
2021-05-10T08:22:26.382935+00:00 app[web.1]:   File "/app/app.py", line 71, in bow
2021-05-10T08:22:26.382935+00:00 app[web.1]:     sentence_words = clean_up_sentence(sentence)
2021-05-10T08:22:26.382935+00:00 app[web.1]:   File "/app/app.py", line 63, in clean_up_sentence
2021-05-10T08:22:26.382936+00:00 app[web.1]:     sentence_words = nltk.word_tokenize(sentence)
2021-05-10T08:22:26.382936+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/nltk/tokenize/__init__.py", line 129, in word_tokenize
2021-05-10T08:22:26.382937+00:00 app[web.1]:     sentences = [text] if preserve_line else 
sent_tokenize(text, language)
2021-05-10T08:22:26.382937+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/nltk/tokenize/__init__.py", line 106, in sent_tokenize
2021-05-10T08:22:26.382938+00:00 app[web.1]:     tokenizer = 
load("tokenizers/punkt/{0}.pickle".format(language))
2021-05-10T08:22:26.382938+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/nltk/data.py", line 752, in load
2021-05-10T08:22:26.382939+00:00 app[web.1]:     opened_resource = _open(resource_url)
2021-05-10T08:22:26.382939+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/nltk/data.py", line 877, in _open
2021-05-10T08:22:26.382939+00:00 app[web.1]:     return find(path_, path + [""]).open()
2021-05-10T08:22:26.382940+00:00 app[web.1]:   File "/app/.heroku/python/lib/python3.9/site- 
packages/nltk/data.py", line 585, in find
2021-05-10T08:22:26.382940+00:00 app[web.1]:     raise LookupError(resource_not_found)
2021-05-10T08:22:26.382940+00:00 app[web.1]: LookupError: 
2021-05-10T08:22:26.382941+00:00 app[web.1]: 
**********************************************************************
2021-05-10T08:22:26.382941+00:00 app[web.1]:   Resource [93mpunkt[0m not found.
2021-05-10T08:22:26.382942+00:00 app[web.1]:   Please use the NLTK Downloader to obtain the resource:
2021-05-10T08:22:26.382942+00:00 app[web.1]: 
2021-05-10T08:22:26.382942+00:00 app[web.1]:   [31m>>> import nltk
2021-05-10T08:22:26.382943+00:00 app[web.1]:   >>> nltk.download('punkt')
2021-05-10T08:22:26.382943+00:00 app[web.1]:   [0m
2021-05-10T08:22:26.382944+00:00 app[web.1]:   For more information see: 
https://www.nltk.org/data.html
2021-05-10T08:22:26.382944+00:00 app[web.1]: 
2021-05-10T08:22:26.382945+00:00 app[web.1]:   Attempted to load 
[93mtokenizers/punkt/PY3/english.pickle[0m
2021-05-10T08:22:26.382945+00:00 app[web.1]: 
2021-05-10T08:22:26.382945+00:00 app[web.1]:   Searched in:
2021-05-10T08:22:26.382945+00:00 app[web.1]:     - '/app/nltk_data'
2021-05-10T08:22:26.382946+00:00 app[web.1]:     - '/app/.heroku/python/nltk_data'
2021-05-10T08:22:26.382946+00:00 app[web.1]:     - '/app/.heroku/python/share/nltk_data'
2021-05-10T08:22:26.382947+00:00 app[web.1]:     - '/app/.heroku/python/lib/nltk_data'
2021-05-10T08:22:26.382947+00:00 app[web.1]:     - '/usr/share/nltk_data'
2021-05-10T08:22:26.382947+00:00 app[web.1]:     - '/usr/local/share/nltk_data'
2021-05-10T08:22:26.382948+00:00 app[web.1]:     - '/usr/lib/nltk_data'
2021-05-10T08:22:26.382948+00:00 app[web.1]:     - '/usr/local/lib/nltk_data'
2021-05-10T08:22:26.382948+00:00 app[web.1]:     - ''
2021-05-10T08:22:26.382949+00:00 app[web.1]: 
**********************************************************************
2021-05-10T08:22:26.382949+00:00 app[web.1]: 
2021-05-10T08:22:26.384684+00:00 app[web.1]: 10.31.72.133 - - [10/May/2021:08:22:26 +0000] "POST /get 
HTTP/1.1" 500 290 "https://ihaci-chat.herokuapp.com/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) 
AppleWebKit/537.36 (KHTML, like Gecko) Chrome/90.0.4430.93 Safari/537.36 Edg/90.0.818.56"

app.py:

# libraries
import random
import numpy as np
import pickle
import json
from flask import Flask, render_template, request
from flask_ngrok import run_with_ngrok
import nltk
from keras.models import load_model
from nltk.stem import WordNetLemmatizer
lemmatizer = WordNetLemmatizer()


# chat initialization
model = load_model("chatbot_model.h5")
intents = json.loads(open('intents.json',encoding="utf8").read())
words = pickle.load(open("words.pkl", "rb"))
classes = pickle.load(open("classes.pkl", "rb"))

app = Flask(__name__)
run_with_ngrok(app)

@app.route("/")
def home():
    return render_template("index.html")


@app.route("/get", methods=["POST"])
def chatbot_response():
    msg = request.form["msg"]
    #checks is a user has given a name, in order to give a personalized feedback
    if msg.startswith('my name is'):
        name = msg[11:]
        ints = predict_class(msg, model)
        res1 = getResponse(ints, intents)
        res =res1.replace("{n}",name)
    elif msg.startswith('hi my name is'):
        name = msg[14:]
        ints = predict_class(msg, model)
        res1 = getResponse(ints, intents)
        res =res1.replace("{n}",name)
    #if no name is passed execute normally
    else:
        ints = predict_class(msg, model)
        res = getResponse(ints, intents)
    return res



 # chat functionalities
def clean_up_sentence(sentence):
    sentence_words = nltk.word_tokenize(sentence)
    sentence_words = [lemmatizer.lemmatize(word.lower()) for word in sentence_words]
    return sentence_words


# return bag of words array: 0 or 1 for each word in the bag that exists in the sentence
def bow(sentence, words, show_details=True):
 # tokenize the pattern
    sentence_words = clean_up_sentence(sentence)
# bag of words - matrix of N words, vocabulary matrix
    bag = [0] * len(words)
    for s in sentence_words:
        for i, w in enumerate(words):
            if w == s:
            # assign 1 if current word is in the vocabulary position
                bag[i] = 1
                if show_details:
                    print("found in bag: %s" % w)
    return np.array(bag)


def predict_class(sentence, model):
    # filter out predictions below a threshold
    p = bow(sentence, words, show_details=False)
    res = model.predict(np.array([p]))[0]
    ERROR_THRESHOLD = 0.25
    results = [[i, r] for i, r in enumerate(res) if r > ERROR_THRESHOLD]
    # sort by strength of probability
    results.sort(key=lambda x: x[1], reverse=True)
    return_list = []
    for r in results:
        return_list.append({"intent": classes[r[0]], "probability": str(r[1])})
    return return_list


def getResponse(ints, intents_json):
    tag = ints[0]["intent"]
    list_of_intents = intents_json["intents"]
    for i in list_of_intents:
        if i["tag"] == tag:
            result = random.choice(i["responses"])
            break
    return result


if __name__ == "__main__":
    app.run()

Can anyone help me to understand what this error exactly is? What should I do to solve this specific error?

Answer 1

Your Error is clear in its nature from Heroku logs, your chat bot is working fine in local system because your chat bot is using NLTK, and NLTK needs a resource which is already downloaded in your local system. The resource is however not found in the Heroku server itself.

Write the following script in a temp file and execute it once per deployment to make your chat bot run smoothly on Heroku. Also you can have a check of existence of this resource in your chat bot itself to download it automatically if it not found!

import nltk
nltk.download('punkt')