Monitor your logs from anywhere
Project description
Monitor the metrics of a model remotely, while training the model.
How to install the package?
1. Install the monitor-logs package using.
pip install monitor-logs
2. Import the package.
from monitor_logs import api
How to use the package?
1. Call the API from the training script using.
api.call(epoch_num, actual, preds, "username", "session_name")
epoch_num - Epoch num from the training
actual - 1D array of ground truth labels
preds - 1D array of predicted labels
username - username of the user
session_name - Name of the session (User Defined).
- Since you need the session name to read the logs, make sure to give unique and readable name.
- Please make sure to follow naming convention standard.
- Special characters such as underscore, hash, dollar, asterisk can be used.
- session_name shouldn't start with special characters.
2. Login to our website or Create a new account here: Home Page.
3. Once the model starts training, go to this website: http://139.162.6.184:5000/. Select your session name and metric to view the model logs.
Find the GitHub Page here.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
monitor_logs-0.1.3.tar.gz
(2.4 kB
view hashes)