{ "id": 356603, "title": "Competition metric- log loss-- references and refresher", "url": "/competitions/tabular-playground-series-oct-2022/discussion/356603", "author_name": "Ravi Ramakrishnan", "post_date": "2022-10-01T07:22:55.534000", "votes": 7, "comment_count": 0, "views": 0, "content": "

Hello all,

\n

The competition metric is the commonly used log-loss, a metric to identify how distant (or close) the predicted probability of an event is to the truth value of an event. I believe this is one of the most commonly used metrics in most of the use-cases. The below references could be handy for a refresher in this regard-

\n
    \n
  1. https://towardsdatascience.com/intuition-behind-log-loss-score-4e0c9979680a
    2. \n
  2. https://www.kaggle.com/dansbecker/what-is-log-loss
    3. \n
  3. https://medium.com/@fzammito/whats-considered-a-good-log-loss-in-machine-learning-a529d400632d
    4. \n
  4. https://dasha.ai/en-us/blog/log-loss-function
    5. \n
  5. http://scikit-learn.org/stable/modules/generated/sklearn.metrics.log_loss.html
    6. \n
  6. https://iryndin.dev/post/log-loss/
    7. \n
\n

Hope these help! All the best!!

", "messages": [ { "id": 1965168, "postDate": "2022-10-01T07:22:55.533Z", "content": "

Hello all,

\n

The competition metric is the commonly used log-loss, a metric to identify how distant (or close) the predicted probability of an event is to the truth value of an event. I believe this is one of the most commonly used metrics in most of the use-cases. The below references could be handy for a refresher in this regard-

\n
    \n
  1. https://towardsdatascience.com/intuition-behind-log-loss-score-4e0c9979680a
    2. \n
  2. https://www.kaggle.com/dansbecker/what-is-log-loss
    3. \n
  3. https://medium.com/@fzammito/whats-considered-a-good-log-loss-in-machine-learning-a529d400632d
    4. \n
  4. https://dasha.ai/en-us/blog/log-loss-function
    5. \n
  5. http://scikit-learn.org/stable/modules/generated/sklearn.metrics.log_loss.html
    6. \n
  6. https://iryndin.dev/post/log-loss/
    7. \n
\n

Hope these help! All the best!!

", "rawMarkdown": "Hello all,\n\nThe competition metric is the commonly used log-loss, a metric to identify how distant (or close) the predicted probability of an event is to the truth value of an event. I believe this is one of the most commonly used metrics in most of the use-cases. The below references could be handy for a refresher in this regard- \n\n1. https://towardsdatascience.com/intuition-behind-log-loss-score-4e0c9979680a\n2. https://www.kaggle.com/dansbecker/what-is-log-loss\n3. https://medium.com/@fzammito/whats-considered-a-good-log-loss-in-machine-learning-a529d400632d\n4. https://dasha.ai/en-us/blog/log-loss-function\n5. http://scikit-learn.org/stable/modules/generated/sklearn.metrics.log_loss.html\n6. https://iryndin.dev/post/log-loss/\n\nHope these help! All the best!!", "votes": 7 } ], "comments": [], "raw_markdown_by_id": { "1965168": "Hello all,\n\nThe competition metric is the commonly used log-loss, a metric to identify how distant (or close) the predicted probability of an event is to the truth value of an event. I believe this is one of the most commonly used metrics in most of the use-cases. The below references could be handy for a refresher in this regard- \n\n1. https://towardsdatascience.com/intuition-behind-log-loss-score-4e0c9979680a\n2. https://www.kaggle.com/dansbecker/what-is-log-loss\n3. https://medium.com/@fzammito/whats-considered-a-good-log-loss-in-machine-learning-a529d400632d\n4. https://dasha.ai/en-us/blog/log-loss-function\n5. http://scikit-learn.org/stable/modules/generated/sklearn.metrics.log_loss.html\n6. https://iryndin.dev/post/log-loss/\n\nHope these help! All the best!!" } }