Kaggle Reviews

The hardware accelerators it offers, along with the generous weekly quotas, are a real lifesaver. I often work with the P100 GPU it offers and it has saved so much of my time.

Separating and merging code cells is quite easy with the intuitive keyboard shortcuts. The dual GPU (T4 *2) setup is especially useful for parallel learning.

It’s not easy to set up or freeze an environment with specific dependencies. Instead, I have to install the required packages again in every session (or create the dependencies installation file and run it). Also, when I run a cell using the keyboard shortcut (SHIFT + ENTER), it doesn’t visually move me to the next cell, so I end up having to scroll to find it.

Furthermore, it would be nice to be able to see the time it took for each cell to run (similar to google colab).

For me Kaggle is by far the best platform for machine learning. As it uses notbooks to run the code, execution of code becomes much easier as we can do cell execution. Also downloading any libraries is quite handy on Kaggle. Getting the data from different sources is also easy.

Only thing I think where kaggle is lagging is the starting of a cluster as sometimes it take a bit longer to start. So kaggle team can work on making it more efficient.

Most of the time I practice my machine learning algorithms on Kaggle as it is easy and it has all the computation power needed to run the code. Also there are some courses provided by Kaggle that have helped me in Data Science journey.

Kaggle is a best platform where the data scientist or data analyst find the best datasets for the create machine learning models it provides various dataset which are image, text , audio all type.

Kaggle provide many many options where you are not find the dataset which dataset is best for your models which you are a train for the your model. And some time you have to clean the data in the dataset.

For finding the dataset

.Kaggle is a great platform to showcase one's data science skills

It has a vast pool of datasets

It also gives opportunities to view other data scientist's notebooks and connect with them

Kaggle offers free GPU only for a limited time

Wide range of complete datasets. Very practical and fast to find what you are looking for.

I didn't find any downsides to using Kaggle.

Allows you to quickly find a wide variety of datasets.

It serves as a coding platform, has unlimited datasets that are highly diverse and almost cover all topics existing, it hosts multiple competitions that not only helps to solve pressing problems through AI through various user contributions but also provides hands on experience to AI and ML engineers worldwide with different experience levels. It also has a huge community of professionals, engineers and enthusiasts that always seem to have the answers to my questions with great support levels.

Not much honestly but sometimes for competitions some rules are very clear or not mentioned at all for example in my last competition with Kaggle there was no mention of when to submit the code itself and I was surprised that It needed to be submitted before a certain deadline that wasn't mentioned.

Kaggle is contributing in solving the world's most stressing problems through AI not only that but also helps incorporate AI in almost all industries with providing hands on experience to millions of users and promoting community support and collaboration. When I started with AI I would only go to Kaggle for datasets and to ask for support. I also love the ranks it gives users and the small accomplishments that push you to go further. I gained huge experience through the enrollment in the competitions it offers and I still do.

I am a Data Enthusiast and was really struggling to have a start and I found Kaggle. The best part of Kaggle is that it provides a range of Datasets, Online Editor, and computing resources and they are free and enough for new users.

Kaggle is best and doesn't have such a thing to dislike but still talking about it. It provides GPU access for only 37hrs per week. But many of us are ok with it.

I will recommend Kaggle for Data Enthusiast who are really looking for a carrier in Data.

I am a Proposed DataScientist and am using Kaggle and its data to test and build different machine learning models ranging from health detection models to object detection models. These benefits really help me and other learners to grow in their carrier.

The thing that I love about Kaggle is that it has a wide variety of never-ending datasets and a global community of data scientists to learn from.

As such for now absolutely there is nothing that I don't l like about Kaggle, everything is best. But the UI can still be Improved more.

I am able to learn and Impart my knowledge, Plus the certifications and medals add a plus to the resume as well. It has helped me a lot in solving a lot of problems that I was facing while starting up in the field of Data Science and Machine Learning.

I learn a lot of information, especially in competitions. There are very good models written in the field of machine learning. On site; There are various and qualified data sets that are open to everyone. A great platform for data scientists. You can run the codes written in R or Python in the internet browser. Its performance is pretty good.

Some users publish the results of their models in a misleading way. The moderation system may come.

I found datasets to train my models. I have studied many models and learned a lot of information. It was helpful in solving the problems in the models I am currently using.

The datasets provided with all the tools and resources

Nothing

Its too good to be true

They can incorporate more divisions for different streams - data engineering, data science, data analytics