CodeMonitor

Code Monitor is a simple way to send feedback about your code to your smartphone as a Telegram message!

Set up

Telegram side

1. Open Telegram and search for CodeMonitor bot.

2. Send any message to it.

3. Once you send any message the Bot will return a message like 'Your code is: "123456789"' this is your chat code, it will be used to identify your chat on python side.

Python side

1. install the Code Monitor dependency through the following command

pip install CodeMonitor

2. now import it in your code

from CodeMonitor.telegram import Messenger, FitMonitor

3. Once CodeMonitor is imported you'll be able to use two classes: Messenger and FitMonitor, first let's talk about Messenger class.

Messenger

This class allows you to send any string you want to the Telegram chat, only requiring the chat code as mentioned on the Telegram side section.

messenger = Messenger("123456789")

When the class is instantiated it send the message "All ready!", to send your own messages do like the following code snippet.

messenger.send_message("your awesome message!")

FitMonitor

This class is a callback that send feedback about the training of a Keras Neural Network model, it is very usefull if you're training a model for a very long time and need to leave your computer but still want to know how good your training is doing. At the end of each epoch this callback is going to send information about the accuracy and loss (and validation accuracy and loss when possible) of the model.

model.fit(X_train, Y_train,          
          epochs = 10,
          validation_data = (X_test, Y_test), 
          callbacks=[FitMonitor("123456789")])

In the latest version (1.0.2) it is possible to specify which of the logs of your network you want to embed in the messages that will me be sent, for instance, if you just want to send only the loss of your model you can use the parameter "log_keys" as in the following code snippet, otherwise if you want to send all of the logs you can just omit this parameter or set it equal "all".

model.fit(X_train, Y_train,          
          epochs = 10,
          validation_data = (X_test, Y_test), 
          callbacks=[FitMonitor("123456789", log_keys=["loss"])])