Hi there,
\nI'm fairly new to Deep Learning Competitions on Kaggle. I want to try out this competition, but I don't know where I should look to get started. Does anyone have any tips they would like to share? It can be both about competitions in general as well as about this competition specifically.
\nI greatly appreciate your help!
", "messages": [ { "id": "2553936", "postDate": "12/08/2023 16:23:20", "content": "Hi there,
\nI'm fairly new to Deep Learning Competitions on Kaggle. I want to try out this competition, but I don't know where I should look to get started. Does anyone have any tips they would like to share? It can be both about competitions in general as well as about this competition specifically.
\nI greatly appreciate your help!
", "rawMarkdown": "Hi there,\n\nI'm fairly new to Deep Learning Competitions on Kaggle. I want to try out this competition, but I don't know where I should look to get started. Does anyone have any tips they would like to share? It can be both about competitions in general as well as about this competition specifically. \n\nI greatly appreciate your help!", "votes": null }, { "id": "2554010", "postDate": "12/08/2023 17:41:53", "content": "I would say the best way to get started with any competition on Kaggle is to look for similar past competitions (There are lots of past segmentation competitions on Kaggle) and read through the discussions , code and especially winning solutions. That should be a good starting point.
", "rawMarkdown": "I would say the best way to get started with any competition on Kaggle is to look for similar past competitions (There are lots of past segmentation competitions on Kaggle) and read through the discussions , code and especially winning solutions. That should be a good starting point.", "votes": null }, { "id": "2554029", "postDate": "12/08/2023 18:00:10", "content": "I may suggest you to first try the starter competitions like MNIST and Disaster tweets NLP challenge.
\nThen you could peruse past completed competitions and study the approaches across the competition, high scoring kernels and winning solutions. This will provide a foundation for you to compete well. You could also curate and develop your own pipeline to perform EDA, train models and infer and predict that could be used for most use cases. This will be handy to start off across your future competitions. You should have one such pipeline for tabular data, one for NLP tasks and one for CV tasks. This will help you a lot with baseline models and will quickly place you on the LB.
\nIn the meanwhile, try and develop interest areas and connect with Kaggle users in these areas and form a team, this will help you team up when you participate in these competitions. Shared ideas and hardware is of utmost help in these assignments.
Good luck and happy learning @limjeffrey
", "rawMarkdown": "I may suggest you to first try the starter competitions like MNIST and Disaster tweets NLP challenge. \nThen you could peruse past completed competitions and study the approaches across the competition, high scoring kernels and winning solutions. This will provide a foundation for you to compete well. You could also curate and develop your own pipeline to perform EDA, train models and infer and predict that could be used for most use cases. This will be handy to start off across your future competitions. You should have one such pipeline for tabular data, one for NLP tasks and one for CV tasks. This will help you a lot with baseline models and will quickly place you on the LB. \nIn the meanwhile, try and develop interest areas and connect with Kaggle users in these areas and form a team, this will help you team up when you participate in these competitions. Shared ideas and hardware is of utmost help in these assignments. \n\nGood luck and happy learning @limjeffrey", "votes": null }, { "id": "2562048", "postDate": "12/15/2023 04:08:13", "content": "Hi! Here are some competitions to start with
\nI would say the best way to get started with any competition on Kaggle is to look for similar past competitions (There are lots of past segmentation competitions on Kaggle) and read through the discussions , code and especially winning solutions. That should be a good starting point.
", "votes": null, "replies": [] }, { "id": 2554029, "author_name": "ravi20076", "author_url": "", "post_date": "12/08/2023 18:00:10", "content": "I may suggest you to first try the starter competitions like MNIST and Disaster tweets NLP challenge.
\nThen you could peruse past completed competitions and study the approaches across the competition, high scoring kernels and winning solutions. This will provide a foundation for you to compete well. You could also curate and develop your own pipeline to perform EDA, train models and infer and predict that could be used for most use cases. This will be handy to start off across your future competitions. You should have one such pipeline for tabular data, one for NLP tasks and one for CV tasks. This will help you a lot with baseline models and will quickly place you on the LB.
\nIn the meanwhile, try and develop interest areas and connect with Kaggle users in these areas and form a team, this will help you team up when you participate in these competitions. Shared ideas and hardware is of utmost help in these assignments.
Good luck and happy learning @limjeffrey
", "votes": null, "replies": [] }, { "id": 2562048, "author_name": "mlsputnik", "author_url": "", "post_date": "12/15/2023 04:08:13", "content": "Hi! Here are some competitions to start with
\n