I see that this competition has just started and people starts working seriously on it. How you guys spend so much time on it. Do you guys have day time job or are you students.\nI am just starting so it would help me estimate my time.
\n\nAnd you guys don't get bored of doing the same thing again and again??
", "messages": [ { "id": "119141", "postDate": "05/07/2016 15:56:06", "content": "I see that this competition has just started and people starts working seriously on it. How you guys spend so much time on it. Do you guys have day time job or are you students.\nI am just starting so it would help me estimate my time.
\n\nAnd you guys don't get bored of doing the same thing again and again??
", "rawMarkdown": "I see that this competition has just started and people starts working seriously on it. How you guys spend so much time on it. Do you guys have day time job or are you students.\r\nI am just starting so it would help me estimate my time.\r\n\r\nAnd you guys don't get bored of doing the same thing again and again??", "votes": null }, { "id": "119154", "postDate": "05/07/2016 17:21:42", "content": "No. See: https://en.wikipedia.org/wiki/Addiction
", "rawMarkdown": "No. See: https://en.wikipedia.org/wiki/Addiction", "votes": null }, { "id": "119163", "postDate": "05/07/2016 17:57:22", "content": "[quote=yilisg;119154]
\n\nNo. See: https://en.wikipedia.org/wiki/Addiction
\n\n[/quote]\nFrom wikipedia: Addiction is a medical condition that is characterized by compulsive engagement in rewarding stimuli, despite adverse consequences. Kaggle doesn't really have adverse consequences, so it isn't something like gambling or drug addiction. Addiction is also a serious issue and shouldn't be treated lightly. Each Kaggle competition is also unique, though obviously there are similarities.
", "rawMarkdown": "[quote=yilisg;119154]\r\n\r\nNo. See: https://en.wikipedia.org/wiki/Addiction\r\n\r\n[/quote]\r\nFrom wikipedia: Addiction is a medical condition that is characterized by compulsive engagement in rewarding stimuli, despite adverse consequences. Kaggle doesn't really have adverse consequences, so it isn't something like gambling or drug addiction. Addiction is also a serious issue and shouldn't be treated lightly. Each Kaggle competition is also unique, though obviously there are similarities.", "votes": null }, { "id": "119166", "postDate": "05/07/2016 18:18:34", "content": "My question is how much you get time to perform on your real job? Because real job is very demanding, and its pay too.
", "rawMarkdown": "My question is how much you get time to perform on your real job? Because real job is very demanding, and its pay too.", "votes": null }, { "id": "119168", "postDate": "05/07/2016 18:36:25", "content": "I'm always impressed at how fast some folks like NxGTR can tackle large datasets problems and make top submissions in little to no time. But, as he once wrote: https://www.kaggle.com/c/dato-native/forums/t/15826/0-95-already/88775#post88775
\n\nI also agree that some problems may seem similar to each other (like the Santander one and others where you have an anonymized ready-to-xgboost small dataset), but that's definitely not the case of the competitions currently in the pool.
", "rawMarkdown": "I'm always impressed at how fast some folks like NxGTR can tackle large datasets problems and make top submissions in little to no time. But, as he once wrote: https://www.kaggle.com/c/dato-native/forums/t/15826/0-95-already/88775#post88775\r\n\r\nI also agree that some problems may seem similar to each other (like the Santander one and others where you have an anonymized ready-to-xgboost small dataset), but that's definitely not the case of the competitions currently in the pool.", "votes": null }, { "id": "119169", "postDate": "05/07/2016 19:17:01", "content": "Why the down-voting? I think it's a great question!
\n\nFirst, as to addiction - yes, there can definitely be a kaggle addiction. Adverse consequences can arise if you are slacking off at your day job to have more time to work on kaggle. Or if you go into debt to buy a killer workstation. Or if you neglect relationships because you spend all your free time on kaggle. (Well, at least that is what my kaggle addiction therapist claims, but, just like all of you here, I can stop any time I want.)
\n\n@statisticsRus - Here's the deal. Many of us have done this for a few years now. So, it is much easier to hit the ground running. For example, see my contest launch checklist that helps me get through the new contest administrivia as quickly and consistently as possible. And when we work on a problem that is similar, we don't start from scratch. We take that time to develop a set of tools to make things easier. Take a look at the set of tools kaggler Guido Tapia has developed over the years, for example.
\n\nAre there kaggle contests that are boring? Yeah. But you can learn something new from all of them, or at least use them as an excuse to expand your repertoire of data science tools (e.g., switch from python to R, or try using only scikit-learn, etc.).
", "rawMarkdown": "Why the down-voting? I think it's a great question!\r\n\r\nFirst, as to addiction - yes, there can _definitely_ be a kaggle addiction. Adverse consequences can arise if you are slacking off at your day job to have more time to work on kaggle. Or if you go into debt to buy a killer workstation. Or if you neglect relationships because you spend all your free time on kaggle. (Well, at least that is what my kaggle addiction therapist _claims_, but, just like all of you here, I can stop any time I want.)\r\n\r\n@statisticsRus - Here's the deal. Many of us have done this for a few years now. So, it is much easier to hit the ground running. For example, see my [contest launch checklist][1] that helps me get through the new contest administrivia as quickly and consistently as possible. And when we work on a problem that is similar, we don't start from scratch. We take that time to develop a set of tools to make things easier. Take a look at the [set of tools][2] kaggler Guido Tapia has developed over the years, for example.\r\n\r\nAre there kaggle contests that are boring? Yeah. But you can learn something new from all of them, or at least use them as an excuse to expand your repertoire of data science tools (e.g., switch from python to R, or try using only scikit-learn, etc.). \r\n\r\n [1]: https://www.kaggle.com/forums/f/15/kaggle-forum/t/19959/my-standard-work-for-every-new-competition\r\n [2]: https://github.com/gatapia/py_ml_utils", "votes": null }, { "id": "119170", "postDate": "05/07/2016 19:20:27", "content": "@statisticsRus - p.s. My day job is psychotically busy. I'm not saying it is easy to find time for kaggle. But I'll posit that most of the long timers are still here because they love to learn, and kaggle is by far one of the best places around for that.
", "rawMarkdown": "statisticsRus - p.s. My day job is psychotically busy. I'm not saying it is easy to find time for kaggle. But I'll posit that most of the long timers are still here because they love to learn, and kaggle is by far one of the best places around for that.", "votes": null }, { "id": "119175", "postDate": "05/07/2016 20:05:31", "content": "Why I am getting negative votes, when I am trying to learn time management from experienced people.\nAll of us have a family and a job. \nPretty sad and demoralizing community here I find.
", "rawMarkdown": "Why I am getting negative votes, when I am trying to learn time management from experienced people.\r\nAll of us have a family and a job. \r\nPretty sad and demoralizing community here I find.", "votes": null }, { "id": "119176", "postDate": "05/07/2016 20:14:21", "content": "@ inversion:\nThanks for the answer.\nI just went through your contest launch checklist. So you use 3 linux machine. Is it possible to participate using a mac book pro??
", "rawMarkdown": "inversion:\r\nThanks for the answer.\r\nI just went through your contest launch checklist. So you use 3 linux machine. Is it possible to participate using a mac book pro??", "votes": null }, { "id": "119188", "postDate": "05/07/2016 21:30:43", "content": "[quote=statisticsRus;119176]\n@ inversion:\nIs it possible to participate using a mac book pro??\n[/quote]
\n\nAbsolutely.
\n\nStart with the getting started and playground data sets as you figure things out. Some competitions can be very hardware intensive, but not all of them. Check out the Rossmann competition where Gert, the winner (by a large margin), used only a laptop.
", "rawMarkdown": "[quote=statisticsRus;119176]\r\n@ inversion:\r\nIs it possible to participate using a mac book pro??\r\n[/quote]\r\n\r\nAbsolutely.\r\n\r\nStart with the getting started and playground data sets as you figure things out. Some competitions can be very hardware intensive, but not all of them. Check out the Rossmann competition where Gert, the winner (by a large margin), [used only a laptop][1].\r\n\r\n\r\n [1]: https://www.kaggle.com/c/rossmann-store-sales/forums/t/17896/share-your-solution/101749#post101749", "votes": null }, { "id": "119191", "postDate": "05/07/2016 22:03:59", "content": "I used my 4 GB RAM MacPro on all competitions, and will use it on this one. Although i'm not a super master, my best results were on competitions with lots of data. Of course, you need to be precise while coding, your change-validate-test cycle is MUCH slower.
\n\nThe good news is that after feature extraction, your dataset is not so big after all. This is the case here, MS Malware classification is another example. Good luck!
", "rawMarkdown": "I used my 4 GB RAM MacPro on all competitions, and will use it on this one. Although i'm not a super master, my best results were on competitions with lots of data. Of course, you need to be precise while coding, your change-validate-test cycle is MUCH slower. \r\n\r\nThe good news is that after feature extraction, your dataset is not so big after all. This is the case here, [MS Malware classification][1] is another example. Good luck!\r\n\r\n\r\n [1]: https://www.kaggle.com/c/malware-classification/", "votes": null }, { "id": "119212", "postDate": "05/08/2016 04:49:20", "content": "Kaggle provides invaluable data in various domains for improving day job engagement. In my industry, these have help my work and to provide relevant implementations for my clients. So spending hours in understanding new data domains will be invaluable to data scientists who need to tackle new domains in every project.
", "rawMarkdown": "Kaggle provides invaluable data in various domains for improving day job engagement. In my industry, these have help my work and to provide relevant implementations for my clients. So spending hours in understanding new data domains will be invaluable to data scientists who need to tackle new domains in every project.", "votes": null }, { "id": "119250", "postDate": "05/08/2016 13:47:45", "content": "Great question -- this is something that I am curious about not only Kaggle. People always seem to have time to spend on Coursera, and I never manage to finish anything hehe. In my particular case, I work in a research institute and my supervisor likes that we spend time on Kaggle to help students.
", "rawMarkdown": "Great question -- this is something that I am curious about not only Kaggle. People always seem to have time to spend on Coursera, and I never manage to finish anything hehe. In my particular case, I work in a research institute and my supervisor likes that we spend time on Kaggle to help students.", "votes": null }, { "id": "119361", "postDate": "05/09/2016 14:40:01", "content": "All I can say is Super Question! :) - as I am looking at this forum every day and deciding whether to take the plunge or not (given what is stacked up in my day job and how bad it went on the weekends on the last one :)). It is seriously difficult to find time if you have a demanding day job plus family/children. But all the answers from inversion makes sense and helps + it REALLY helps if you team up. So if you are not getting time - do not get demotivated - just read the forums - that by itself is worth a LOT!!.
", "rawMarkdown": "All I can say is Super Question! :) - as I am looking at this forum every day and deciding whether to take the plunge or not (given what is stacked up in my day job and how bad it went on the weekends on the last one :)). It is seriously difficult to find time if you have a demanding day job plus family/children. But all the answers from inversion makes sense and helps + it REALLY helps if you team up. So if you are not getting time - do not get demotivated - just read the forums - that by itself is worth a LOT!!.", "votes": null }, { "id": "119365", "postDate": "05/09/2016 15:05:05", "content": "Another thing I'll add to the discussion - if you look at all the contests and public data sets to explore, there is more kaggle than anyone can handle. So, everyone here is having to decide what the right balance is for them.
\n\nJust figure out what is right for you and don't worry if you can't participate in 3 contests at once. (BTW, for me, 2 contests is the sweet spot. When I'm active in 3, it tends to spread my time too thin and I find I am no longer getting out of kaggle what I want to. Guys like DataGeek seem to be able to be active in every kaggle contest, plus contests on other platforms. Not me.)
", "rawMarkdown": "Another thing I'll add to the discussion - if you look at all the contests and public data sets to explore, there is more kaggle than anyone can handle. So, everyone here is having to decide what the right balance is for them. \r\n\r\nJust figure out what is right for you and don't worry if you can't participate in 3 contests at once. (BTW, for me, 2 contests is the sweet spot. When I'm active in 3, it tends to spread my time too thin and I find I am no longer getting out of kaggle what I want to. Guys like DataGeek seem to be able to be active in every kaggle contest, plus contests on other platforms. Not me.)", "votes": null }, { "id": "120863", "postDate": "05/21/2016 02:36:43", "content": "I face the same dilemma. A full time job + full time family+ other commitments= little time for kaggle. By the time I download the data and get some basic scripts up an running, most people are way ahead and i am already demoralized. I have stopped competing. I participate to learn with a clear goal in mind. Once it is achieved, I stop. I do not have access to linux environment. So setting up a windows environment sucks up a shit load of time. AWS is clumsy. Most people here have a vast arsenal of tools and software setup and ready to go and up to date. I cannot keep up with that pace. I think the key is to keep participating consistently. You will create your own workflow. Strong scripting skills help. Most people bring in experience and code snippets from other projects. Unless you build up that critical mass of experience and code/tools base, you are going to struggle.\nMy suggestion: Find a solid linux machine(s) or use AWS. Set up the most essential libraries. And just keep doing kaggle.
", "rawMarkdown": "I face the same dilemma. A full time job + full time family+ other commitments= little time for kaggle. By the time I download the data and get some basic scripts up an running, most people are way ahead and i am already demoralized. I have stopped competing. I participate to learn with a clear goal in mind. Once it is achieved, I stop. I do not have access to linux environment. So setting up a windows environment sucks up a shit load of time. AWS is clumsy. Most people here have a vast arsenal of tools and software setup and ready to go and up to date. I cannot keep up with that pace. I think the key is to keep participating consistently. You will create your own workflow. Strong scripting skills help. Most people bring in experience and code snippets from other projects. Unless you build up that critical mass of experience and code/tools base, you are going to struggle.\r\nMy suggestion: Find a solid linux machine(s) or use AWS. Set up the most essential libraries. And just keep doing kaggle.", "votes": null }, { "id": "124129", "postDate": "06/15/2016 18:32:11", "content": "My full time job is very relaxing. Nothing intense
", "rawMarkdown": "My full time job is very relaxing. Nothing intense", "votes": null } ], "comments": [ { "id": 119154, "author_name": "yilihome", "author_url": "", "post_date": "05/07/2016 17:21:42", "content": "No. See: https://en.wikipedia.org/wiki/Addiction
", "votes": null, "replies": [] }, { "id": 119163, "author_name": "jturkewitz", "author_url": "", "post_date": "05/07/2016 17:57:22", "content": "[quote=yilisg;119154]
\n\nNo. See: https://en.wikipedia.org/wiki/Addiction
\n\n[/quote]\nFrom wikipedia: Addiction is a medical condition that is characterized by compulsive engagement in rewarding stimuli, despite adverse consequences. Kaggle doesn't really have adverse consequences, so it isn't something like gambling or drug addiction. Addiction is also a serious issue and shouldn't be treated lightly. Each Kaggle competition is also unique, though obviously there are similarities.
", "votes": null, "replies": [] }, { "id": 119166, "author_name": "statisticsrus", "author_url": "", "post_date": "05/07/2016 18:18:34", "content": "My question is how much you get time to perform on your real job? Because real job is very demanding, and its pay too.
", "votes": null, "replies": [] }, { "id": 119168, "author_name": "fernandoprocy", "author_url": "", "post_date": "05/07/2016 18:36:25", "content": "I'm always impressed at how fast some folks like NxGTR can tackle large datasets problems and make top submissions in little to no time. But, as he once wrote: https://www.kaggle.com/c/dato-native/forums/t/15826/0-95-already/88775#post88775
\n\nI also agree that some problems may seem similar to each other (like the Santander one and others where you have an anonymized ready-to-xgboost small dataset), but that's definitely not the case of the competitions currently in the pool.
", "votes": null, "replies": [] }, { "id": 119169, "author_name": "inversion", "author_url": "", "post_date": "05/07/2016 19:17:01", "content": "Why the down-voting? I think it's a great question!
\n\nFirst, as to addiction - yes, there can definitely be a kaggle addiction. Adverse consequences can arise if you are slacking off at your day job to have more time to work on kaggle. Or if you go into debt to buy a killer workstation. Or if you neglect relationships because you spend all your free time on kaggle. (Well, at least that is what my kaggle addiction therapist claims, but, just like all of you here, I can stop any time I want.)
\n\n@statisticsRus - Here's the deal. Many of us have done this for a few years now. So, it is much easier to hit the ground running. For example, see my contest launch checklist that helps me get through the new contest administrivia as quickly and consistently as possible. And when we work on a problem that is similar, we don't start from scratch. We take that time to develop a set of tools to make things easier. Take a look at the set of tools kaggler Guido Tapia has developed over the years, for example.
\n\nAre there kaggle contests that are boring? Yeah. But you can learn something new from all of them, or at least use them as an excuse to expand your repertoire of data science tools (e.g., switch from python to R, or try using only scikit-learn, etc.).
", "votes": null, "replies": [] }, { "id": 119170, "author_name": "inversion", "author_url": "", "post_date": "05/07/2016 19:20:27", "content": "@statisticsRus - p.s. My day job is psychotically busy. I'm not saying it is easy to find time for kaggle. But I'll posit that most of the long timers are still here because they love to learn, and kaggle is by far one of the best places around for that.
", "votes": null, "replies": [] }, { "id": 119175, "author_name": "statisticsrus", "author_url": "", "post_date": "05/07/2016 20:05:31", "content": "Why I am getting negative votes, when I am trying to learn time management from experienced people.\nAll of us have a family and a job. \nPretty sad and demoralizing community here I find.
", "votes": null, "replies": [] }, { "id": 119176, "author_name": "statisticsrus", "author_url": "", "post_date": "05/07/2016 20:14:21", "content": "@ inversion:\nThanks for the answer.\nI just went through your contest launch checklist. So you use 3 linux machine. Is it possible to participate using a mac book pro??
", "votes": null, "replies": [] }, { "id": 119188, "author_name": "inversion", "author_url": "", "post_date": "05/07/2016 21:30:43", "content": "[quote=statisticsRus;119176]\n@ inversion:\nIs it possible to participate using a mac book pro??\n[/quote]
\n\nAbsolutely.
\n\nStart with the getting started and playground data sets as you figure things out. Some competitions can be very hardware intensive, but not all of them. Check out the Rossmann competition where Gert, the winner (by a large margin), used only a laptop.
", "votes": null, "replies": [] }, { "id": 119191, "author_name": "snowdog", "author_url": "", "post_date": "05/07/2016 22:03:59", "content": "I used my 4 GB RAM MacPro on all competitions, and will use it on this one. Although i'm not a super master, my best results were on competitions with lots of data. Of course, you need to be precise while coding, your change-validate-test cycle is MUCH slower.
\n\nThe good news is that after feature extraction, your dataset is not so big after all. This is the case here, MS Malware classification is another example. Good luck!
", "votes": null, "replies": [] }, { "id": 119212, "author_name": "drpatrickchan", "author_url": "", "post_date": "05/08/2016 04:49:20", "content": "Kaggle provides invaluable data in various domains for improving day job engagement. In my industry, these have help my work and to provide relevant implementations for my clients. So spending hours in understanding new data domains will be invaluable to data scientists who need to tackle new domains in every project.
", "votes": null, "replies": [] }, { "id": 119250, "author_name": "rpmcruz", "author_url": "", "post_date": "05/08/2016 13:47:45", "content": "Great question -- this is something that I am curious about not only Kaggle. People always seem to have time to spend on Coursera, and I never manage to finish anything hehe. In my particular case, I work in a research institute and my supervisor likes that we spend time on Kaggle to help students.
", "votes": null, "replies": [] }, { "id": 119361, "author_name": "rightfit", "author_url": "", "post_date": "05/09/2016 14:40:01", "content": "All I can say is Super Question! :) - as I am looking at this forum every day and deciding whether to take the plunge or not (given what is stacked up in my day job and how bad it went on the weekends on the last one :)). It is seriously difficult to find time if you have a demanding day job plus family/children. But all the answers from inversion makes sense and helps + it REALLY helps if you team up. So if you are not getting time - do not get demotivated - just read the forums - that by itself is worth a LOT!!.
", "votes": null, "replies": [] }, { "id": 119365, "author_name": "inversion", "author_url": "", "post_date": "05/09/2016 15:05:05", "content": "Another thing I'll add to the discussion - if you look at all the contests and public data sets to explore, there is more kaggle than anyone can handle. So, everyone here is having to decide what the right balance is for them.
\n\nJust figure out what is right for you and don't worry if you can't participate in 3 contests at once. (BTW, for me, 2 contests is the sweet spot. When I'm active in 3, it tends to spread my time too thin and I find I am no longer getting out of kaggle what I want to. Guys like DataGeek seem to be able to be active in every kaggle contest, plus contests on other platforms. Not me.)
", "votes": null, "replies": [] }, { "id": 120863, "author_name": "wabbit", "author_url": "", "post_date": "05/21/2016 02:36:43", "content": "I face the same dilemma. A full time job + full time family+ other commitments= little time for kaggle. By the time I download the data and get some basic scripts up an running, most people are way ahead and i am already demoralized. I have stopped competing. I participate to learn with a clear goal in mind. Once it is achieved, I stop. I do not have access to linux environment. So setting up a windows environment sucks up a shit load of time. AWS is clumsy. Most people here have a vast arsenal of tools and software setup and ready to go and up to date. I cannot keep up with that pace. I think the key is to keep participating consistently. You will create your own workflow. Strong scripting skills help. Most people bring in experience and code snippets from other projects. Unless you build up that critical mass of experience and code/tools base, you are going to struggle.\nMy suggestion: Find a solid linux machine(s) or use AWS. Set up the most essential libraries. And just keep doing kaggle.
", "votes": null, "replies": [] }, { "id": 124129, "author_name": "yuxiaofelicia", "author_url": "", "post_date": "06/15/2016 18:32:11", "content": "My full time job is very relaxing. Nothing intense
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "119141": "I see that this competition has just started and people starts working seriously on it. How you guys spend so much time on it. Do you guys have day time job or are you students.\r\nI am just starting so it would help me estimate my time.\r\n\r\nAnd you guys don't get bored of doing the same thing again and again??", "119154": "No. See: https://en.wikipedia.org/wiki/Addiction", "119163": "[quote=yilisg;119154]\r\n\r\nNo. See: https://en.wikipedia.org/wiki/Addiction\r\n\r\n[/quote]\r\nFrom wikipedia: Addiction is a medical condition that is characterized by compulsive engagement in rewarding stimuli, despite adverse consequences. Kaggle doesn't really have adverse consequences, so it isn't something like gambling or drug addiction. Addiction is also a serious issue and shouldn't be treated lightly. Each Kaggle competition is also unique, though obviously there are similarities.", "119166": "My question is how much you get time to perform on your real job? Because real job is very demanding, and its pay too.", "119168": "I'm always impressed at how fast some folks like NxGTR can tackle large datasets problems and make top submissions in little to no time. But, as he once wrote: https://www.kaggle.com/c/dato-native/forums/t/15826/0-95-already/88775#post88775\r\n\r\nI also agree that some problems may seem similar to each other (like the Santander one and others where you have an anonymized ready-to-xgboost small dataset), but that's definitely not the case of the competitions currently in the pool.", "119169": "Why the down-voting? I think it's a great question!\r\n\r\nFirst, as to addiction - yes, there can _definitely_ be a kaggle addiction. Adverse consequences can arise if you are slacking off at your day job to have more time to work on kaggle. Or if you go into debt to buy a killer workstation. Or if you neglect relationships because you spend all your free time on kaggle. (Well, at least that is what my kaggle addiction therapist _claims_, but, just like all of you here, I can stop any time I want.)\r\n\r\n@statisticsRus - Here's the deal. Many of us have done this for a few years now. So, it is much easier to hit the ground running. For example, see my [contest launch checklist][1] that helps me get through the new contest administrivia as quickly and consistently as possible. And when we work on a problem that is similar, we don't start from scratch. We take that time to develop a set of tools to make things easier. Take a look at the [set of tools][2] kaggler Guido Tapia has developed over the years, for example.\r\n\r\nAre there kaggle contests that are boring? Yeah. But you can learn something new from all of them, or at least use them as an excuse to expand your repertoire of data science tools (e.g., switch from python to R, or try using only scikit-learn, etc.). \r\n\r\n [1]: https://www.kaggle.com/forums/f/15/kaggle-forum/t/19959/my-standard-work-for-every-new-competition\r\n [2]: https://github.com/gatapia/py_ml_utils", "119170": "statisticsRus - p.s. My day job is psychotically busy. I'm not saying it is easy to find time for kaggle. But I'll posit that most of the long timers are still here because they love to learn, and kaggle is by far one of the best places around for that.", "119175": "Why I am getting negative votes, when I am trying to learn time management from experienced people.\r\nAll of us have a family and a job. \r\nPretty sad and demoralizing community here I find.", "119176": "inversion:\r\nThanks for the answer.\r\nI just went through your contest launch checklist. So you use 3 linux machine. Is it possible to participate using a mac book pro??", "119188": "[quote=statisticsRus;119176]\r\n@ inversion:\r\nIs it possible to participate using a mac book pro??\r\n[/quote]\r\n\r\nAbsolutely.\r\n\r\nStart with the getting started and playground data sets as you figure things out. Some competitions can be very hardware intensive, but not all of them. Check out the Rossmann competition where Gert, the winner (by a large margin), [used only a laptop][1].\r\n\r\n\r\n [1]: https://www.kaggle.com/c/rossmann-store-sales/forums/t/17896/share-your-solution/101749#post101749", "119191": "I used my 4 GB RAM MacPro on all competitions, and will use it on this one. Although i'm not a super master, my best results were on competitions with lots of data. Of course, you need to be precise while coding, your change-validate-test cycle is MUCH slower. \r\n\r\nThe good news is that after feature extraction, your dataset is not so big after all. This is the case here, [MS Malware classification][1] is another example. Good luck!\r\n\r\n\r\n [1]: https://www.kaggle.com/c/malware-classification/", "119212": "Kaggle provides invaluable data in various domains for improving day job engagement. In my industry, these have help my work and to provide relevant implementations for my clients. So spending hours in understanding new data domains will be invaluable to data scientists who need to tackle new domains in every project.", "119250": "Great question -- this is something that I am curious about not only Kaggle. People always seem to have time to spend on Coursera, and I never manage to finish anything hehe. In my particular case, I work in a research institute and my supervisor likes that we spend time on Kaggle to help students.", "119361": "All I can say is Super Question! :) - as I am looking at this forum every day and deciding whether to take the plunge or not (given what is stacked up in my day job and how bad it went on the weekends on the last one :)). It is seriously difficult to find time if you have a demanding day job plus family/children. But all the answers from inversion makes sense and helps + it REALLY helps if you team up. So if you are not getting time - do not get demotivated - just read the forums - that by itself is worth a LOT!!.", "119365": "Another thing I'll add to the discussion - if you look at all the contests and public data sets to explore, there is more kaggle than anyone can handle. So, everyone here is having to decide what the right balance is for them. \r\n\r\nJust figure out what is right for you and don't worry if you can't participate in 3 contests at once. (BTW, for me, 2 contests is the sweet spot. When I'm active in 3, it tends to spread my time too thin and I find I am no longer getting out of kaggle what I want to. Guys like DataGeek seem to be able to be active in every kaggle contest, plus contests on other platforms. Not me.)", "120863": "I face the same dilemma. A full time job + full time family+ other commitments= little time for kaggle. By the time I download the data and get some basic scripts up an running, most people are way ahead and i am already demoralized. I have stopped competing. I participate to learn with a clear goal in mind. Once it is achieved, I stop. I do not have access to linux environment. So setting up a windows environment sucks up a shit load of time. AWS is clumsy. Most people here have a vast arsenal of tools and software setup and ready to go and up to date. I cannot keep up with that pace. I think the key is to keep participating consistently. You will create your own workflow. Strong scripting skills help. Most people bring in experience and code snippets from other projects. Unless you build up that critical mass of experience and code/tools base, you are going to struggle.\r\nMy suggestion: Find a solid linux machine(s) or use AWS. Set up the most essential libraries. And just keep doing kaggle.", "124129": "My full time job is very relaxing. Nothing intense" }, "source": "meta" }