TypeError: a bytes-like object is required, not 'str' about how_to_deploy_a_keras_model_to_production HOT 6 CLOSED

Python 3 doesn't handle base64 encoding (which is a pure ASCII encoding) the same way as Python 2.

You can use this function instead:

import base64

def convertImage(imgData1):
    imgstr = re.search(b'data:image/png;base64,(.*)', imgData1).group(1)
    with open('output.png', 'wb') as output:
        output.write(base64.b64decode(imgstr))

If you pass the image to Python through the "Predict" button, something like this gets returned:
b'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAARgAAAEYCAYAAACHjumMAAAdlklEQVR4Xu2de2jX1RvHH295XWppKAbOFDQvqJFibEwNpgZBin+UWOiwkChzpqKx2VxqKrm8VGSgTUElQ0ihUFfkbSPUnIrWjLxGorHBnNeVU388Xz

With the regex, you remove the part before the actual image data.

You also need to change the paths in the load.py file if you place your models in a "model" folder.

from how_to_deploy_a_keras_model_to_production.

I'm using python 3 and then got this error

from how_to_deploy_a_keras_model_to_production.

wow! I know the first part of base64 string . but I don't know how to remove it.
Amazing, now I can fix it ✌️ thank you

from how_to_deploy_a_keras_model_to_production.

OK, Thanks @frankherfert . Problem solved.

from how_to_deploy_a_keras_model_to_production.

Traceback (most recent call last):
File "netconf_tool.py", line 18, in
tn.read_until("login: ")
File "C:\lib\telnetlib.py", line 302, in read_until
i = self.cookedq.find(match)
TypeError: a bytes-like object is required, not 'str'

Code snippet:
!/usr/bin/env python

import getpass
import sys
import telnetlib
import re
import pexpect
import base64

OST = str(sys.argv[1])
#user = raw_input("Enter your remote account: ")
#password = getpass.getpass()
user = "gss"
password = "pureethernet"

tn = telnetlib.Telnet(HOST)
tn.read_until("login: ")
tn.write(user + "\n")
if password:
tn.read_until("Password: ")
tn.write(password + "\n")
f = open("C:\djangogirls\blog\templates\startup-config.txt", "rb")
output.write(base64.b64decode(f))

from how_to_deploy_a_keras_model_to_production.

frankherfert: Would you please help me when I have the same error with the following code:
self.kmeans = load(open('data/kmeans_2.dat', 'r'))

from how_to_deploy_a_keras_model_to_production.