Use my pre-trained model directly which was trained on American Express Credit Default Prediction corpus.
\nLightGBM is a gradient boosting model which basically uses ensemble learning methods. Gradient Boosting is a popular boosting algorithm. In gradient boosting, each predictor corrects its predecessor’s error. In contrast to Adaboost, the weights of the training instances are not tweaked, instead, each predictor is trained using the residual errors of the predecessor as labels.
\nIf you found the model helpful please upvote.
\nThanks
", "messages": [ { "id": "1814623", "postDate": "06/08/2022 05:42:42", "content": "\n\nUse my pre-trained model directly which was trained on American Express Credit Default Prediction corpus.
\nLightGBM is a gradient boosting model which basically uses ensemble learning methods. Gradient Boosting is a popular boosting algorithm. In gradient boosting, each predictor corrects its predecessor’s error. In contrast to Adaboost, the weights of the training instances are not tweaked, instead, each predictor is trained using the residual errors of the predecessor as labels.
\nIf you found the model helpful please upvote.
\nThanks
", "rawMarkdown": "[Trained LightGBM Model](https://www.kaggle.com/datasets/bhavikardeshna/american-express-credit-lightgbm-model)\n\n[My Notebook for using model](https://www.kaggle.com/bhavikardeshna/lightgbm-ensemble-learning)\n\nUse my pre-trained model directly which was trained on ***American Express Credit Default Prediction*** corpus.\n\nLightGBM is a gradient boosting model which basically uses ensemble learning methods. Gradient Boosting is a popular boosting algorithm. In gradient boosting, each predictor corrects its predecessor’s error. In contrast to Adaboost, the weights of the training instances are not tweaked, instead, each predictor is trained using the residual errors of the predecessor as labels.\n\nIf you found the model helpful please **upvote**.\n\nThanks", "votes": null }, { "id": "1814938", "postDate": "06/08/2022 13:48:11", "content": "Hi @bhavikardeshna Most people are on Kaggle because they want to learn something. It is hard to do something useful with a trained model without even knowing what features it expects as input. You'll get more applause if you show us the (well-explained) code with which you have engineered your features, trained the model and cross-validated it.
\nThis discussion gives some more ideas about what kinds of contributions people like to see.
", "rawMarkdown": "Hi @bhavikardeshna Most people are on Kaggle because they want to learn something. It is hard to do something useful with a trained model without even knowing what features it expects as input. You'll get more applause if you show us the (well-explained) code with which you have engineered your features, trained the model and cross-validated it. \n\n[This discussion](https://www.kaggle.com/competitions/tabular-playground-series-jan-2022/discussion/302569) gives some more ideas about what kinds of contributions people like to see.", "votes": null }, { "id": "1815159", "postDate": "06/08/2022 18:28:32", "content": "@ambrosm sure
\nI am working on it. I will share a guide regarding the trained LightGBM model soon😃
Thanks
", "rawMarkdown": "ambrosm sure\nI am working on it. I will share a guide regarding the trained LightGBM model soon😃\n\nThanks", "votes": null } ], "comments": [ { "id": 1814938, "author_name": "ambrosm", "author_url": "", "post_date": "06/08/2022 13:48:11", "content": "Hi @bhavikardeshna Most people are on Kaggle because they want to learn something. It is hard to do something useful with a trained model without even knowing what features it expects as input. You'll get more applause if you show us the (well-explained) code with which you have engineered your features, trained the model and cross-validated it.
\nThis discussion gives some more ideas about what kinds of contributions people like to see.
", "votes": null, "replies": [] }, { "id": 1815159, "author_name": "bhavikardeshna", "author_url": "", "post_date": "06/08/2022 18:28:32", "content": "@ambrosm sure
\nI am working on it. I will share a guide regarding the trained LightGBM model soon😃
Thanks
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1814623": "[Trained LightGBM Model](https://www.kaggle.com/datasets/bhavikardeshna/american-express-credit-lightgbm-model)\n\n[My Notebook for using model](https://www.kaggle.com/bhavikardeshna/lightgbm-ensemble-learning)\n\nUse my pre-trained model directly which was trained on ***American Express Credit Default Prediction*** corpus.\n\nLightGBM is a gradient boosting model which basically uses ensemble learning methods. Gradient Boosting is a popular boosting algorithm. In gradient boosting, each predictor corrects its predecessor’s error. In contrast to Adaboost, the weights of the training instances are not tweaked, instead, each predictor is trained using the residual errors of the predecessor as labels.\n\nIf you found the model helpful please **upvote**.\n\nThanks", "1814938": "Hi @bhavikardeshna Most people are on Kaggle because they want to learn something. It is hard to do something useful with a trained model without even knowing what features it expects as input. You'll get more applause if you show us the (well-explained) code with which you have engineered your features, trained the model and cross-validated it. \n\n[This discussion](https://www.kaggle.com/competitions/tabular-playground-series-jan-2022/discussion/302569) gives some more ideas about what kinds of contributions people like to see.", "1815159": "ambrosm sure\nI am working on it. I will share a guide regarding the trained LightGBM model soon😃\n\nThanks" }, "source": "meta" }