I joined this competition 25 days ago before the end date. I didn't know anything about BERT or transformers. I knew the basics of sentiment analysis but BERT was a whole new concept to me but I liked the problem statement so I decided to work on it. I went through a lot of articles to understand BERT. I also went through a lot of public kernels to get started with the competition (special thanks to @tanulsingh077 introduction notebook which helped me to get started with the competition).
\n\nAlthough I didn't score a medal, I got to learn a lot of new things by going through many kernels and discussions. I will now try to implement the solutions given by the winners to enhance my knowledge.
\n\nTo any data science enthusiast, who is a newbie and gets overwhelmed by the idea of participating in a competition without having pre-requisite knowledge, my advice is to just get started with the project. Research about the problem statement. Go through public kernels and discussion forums. You may or may not win a medal but you will get to learn a lot! In the end, it's all about learning and trust me you will get to learn a lot.
", "messages": [ { "id": "907111", "postDate": "06/29/2020 17:31:25", "content": "I joined this competition 25 days ago before the end date. I didn't know anything about BERT or transformers. I knew the basics of sentiment analysis but BERT was a whole new concept to me but I liked the problem statement so I decided to work on it. I went through a lot of articles to understand BERT. I also went through a lot of public kernels to get started with the competition (special thanks to @tanulsingh077 introduction notebook which helped me to get started with the competition).
\n\nAlthough I didn't score a medal, I got to learn a lot of new things by going through many kernels and discussions. I will now try to implement the solutions given by the winners to enhance my knowledge.
\n\nTo any data science enthusiast, who is a newbie and gets overwhelmed by the idea of participating in a competition without having pre-requisite knowledge, my advice is to just get started with the project. Research about the problem statement. Go through public kernels and discussion forums. You may or may not win a medal but you will get to learn a lot! In the end, it's all about learning and trust me you will get to learn a lot.
", "rawMarkdown": "I joined this competition 25 days ago before the end date. I didn't know anything about BERT or transformers. I knew the basics of sentiment analysis but BERT was a whole new concept to me but I liked the problem statement so I decided to work on it. I went through a lot of articles to understand BERT. I also went through a lot of public kernels to get started with the competition (special thanks to @tanulsingh077 introduction notebook which helped me to get started with the competition).\n\nAlthough I didn't score a medal, I got to learn a lot of new things by going through many kernels and discussions. I will now try to implement the solutions given by the winners to enhance my knowledge. \n\nTo any data science enthusiast, who is a newbie and gets overwhelmed by the idea of participating in a competition without having pre-requisite knowledge, my advice is to just get started with the project. Research about the problem statement. Go through public kernels and discussion forums. You may or may not win a medal but you will get to learn a lot! In the end, it's all about learning and trust me you will get to learn a lot.", "votes": null }, { "id": "907137", "postDate": "06/29/2020 17:47:14", "content": "I'm a beginner here. I have been doing some courses on Coursera to get started. What is your suggestion?
", "rawMarkdown": "I'm a beginner here. I have been doing some courses on Coursera to get started. What is your suggestion?", "votes": null }, { "id": "907197", "postDate": "06/29/2020 18:47:27", "content": "Pick up any dataset of your liking on Kaggle and start working towards it. Study the data. Research about different ways used to solve the problem statement. Start learning the concepts and implement them on the dataset. It may take some time but you will get to learn a lot of new things. Happy kaggling 😄!
", "rawMarkdown": "Pick up any dataset of your liking on Kaggle and start working towards it. Study the data. Research about different ways used to solve the problem statement. Start learning the concepts and implement them on the dataset. It may take some time but you will get to learn a lot of new things. Happy kaggling 😄!", "votes": null }, { "id": "907202", "postDate": "06/29/2020 18:54:14", "content": "Yeah the learning curve is more important than medal .
", "rawMarkdown": "Yeah the learning curve is more important than medal .", "votes": null }, { "id": "907575", "postDate": "06/30/2020 03:29:57", "content": "Great piece of motivational advice!
", "rawMarkdown": "Great piece of motivational advice!", "votes": null } ], "comments": [ { "id": 907137, "author_name": "anmspro", "author_url": "", "post_date": "06/29/2020 17:47:14", "content": "I'm a beginner here. I have been doing some courses on Coursera to get started. What is your suggestion?
", "votes": null, "replies": [ { "id": 907197, "author_name": "nkaps98", "author_url": "", "post_date": "06/29/2020 18:47:27", "content": "Pick up any dataset of your liking on Kaggle and start working towards it. Study the data. Research about different ways used to solve the problem statement. Start learning the concepts and implement them on the dataset. It may take some time but you will get to learn a lot of new things. Happy kaggling 😄!
", "votes": null, "replies": [] } ] }, { "id": 907202, "author_name": "yash612", "author_url": "", "post_date": "06/29/2020 18:54:14", "content": "Yeah the learning curve is more important than medal .
", "votes": null, "replies": [] }, { "id": 907575, "author_name": "gauravdahiya", "author_url": "", "post_date": "06/30/2020 03:29:57", "content": "Great piece of motivational advice!
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "907111": "I joined this competition 25 days ago before the end date. I didn't know anything about BERT or transformers. I knew the basics of sentiment analysis but BERT was a whole new concept to me but I liked the problem statement so I decided to work on it. I went through a lot of articles to understand BERT. I also went through a lot of public kernels to get started with the competition (special thanks to @tanulsingh077 introduction notebook which helped me to get started with the competition).\n\nAlthough I didn't score a medal, I got to learn a lot of new things by going through many kernels and discussions. I will now try to implement the solutions given by the winners to enhance my knowledge. \n\nTo any data science enthusiast, who is a newbie and gets overwhelmed by the idea of participating in a competition without having pre-requisite knowledge, my advice is to just get started with the project. Research about the problem statement. Go through public kernels and discussion forums. You may or may not win a medal but you will get to learn a lot! In the end, it's all about learning and trust me you will get to learn a lot.", "907137": "I'm a beginner here. I have been doing some courses on Coursera to get started. What is your suggestion?", "907197": "Pick up any dataset of your liking on Kaggle and start working towards it. Study the data. Research about different ways used to solve the problem statement. Start learning the concepts and implement them on the dataset. It may take some time but you will get to learn a lot of new things. Happy kaggling 😄!", "907202": "Yeah the learning curve is more important than medal .", "907575": "Great piece of motivational advice!" }, "source": "meta" }