scikit-learn Reviews & Product Details

Scikit learn is the library for machine learning which is well documented that a naive machine learning developer can also use it.The algorithms that are implemented in the library are the common machine learning algorithms and they can scale for almost every size of data.You can easily use the machine learning algorithms in a normal python program and take the advantage of the data analytics through ML using scikit learn. Review collected by and hosted on G2.com.

Scikit learn has not left any false clue besides it that is you cannot even find a single evidence for not liking it. Review collected by and hosted on G2.com.

The best machine learning library that I have found on the web. It is the library which is used by the experts for machine learning exercises. Using scikit-learn you can easily get your classifier or else regression model developed in a single line and then just train your data through that classifier by passing training data to it and also you can save the trained model and use it in future.You can also customize the famous ML algorithms and tune them according to your usage. Review collected by and hosted on G2.com.

Nothing to dislike about the best Machine Learning Library. Review collected by and hosted on G2.com.

- It is open source.

- It has a huge community support.

- One can easily find tutorials to learn it.

- Detailed documentation with details. Review collected by and hosted on G2.com.

It has been my helping hand when it comes to Machine Learning. I have no problem or dislikes for this very great and helpful library. Review collected by and hosted on G2.com.

Most of the complex problems are solved easily with the help of it's potential of selecting algorithms. It also covers most of the machine learning tasks. It has a great interface and is a well-updated module. The scalability and robustness makes it very easy to use. Review collected by and hosted on G2.com.

It is not very likely used where there is a high requirement of statistical information. Review collected by and hosted on G2.com.

It covers most of the machine learning tasks. It scales to most data problems. The selection of solid algorithms. A well-updated module. It's API documentation. The support for customer. It is robust and easy to use. Review collected by and hosted on G2.com.

It doesn't support GPU acceleration. It has less of a focus on statistics than R does. Review collected by and hosted on G2.com.

The documentation is clean and clear one can easily understand. If you face any problems you can easily find the solution over the internet as there are a lot of people using it around the world. I almost use is everywhere I use Machine Learning. Review collected by and hosted on G2.com.

No dislikes for such a well documented and helpful library. Review collected by and hosted on G2.com.

Scikit-learn is a well-documented Python library that gives easy access to many prominent machine learning algorithms. The library is designed in such a way as to have a consistent API regardless of which algorithm you choose to use, so it is easy to pick up and try a new algorithm you have never used before. Review collected by and hosted on G2.com.

As with any library of this type (compilation of many different algorithms), it doesn't always contain the content you're looking for. Scikit-learn only contains the most popular algorithms, so if you're looking for an implementation of a more specialized algorithm, it's very possible you won't find it in the library. Review collected by and hosted on G2.com.

You can do classification, clustering, regression, pre processing and so many. If you are working in machine learning based research, I would highly recommend this package. Review collected by and hosted on G2.com.

Nothing is dislike. Every thing comes without cost and its really efficient. You just need to know basic python coding Review collected by and hosted on G2.com.

It has the best libraries that can run on data. It is mainly helpful when you are doing supervised or unsupervised machine learning on your data Review collected by and hosted on G2.com.

python is slow. Therefore using the libraries makes dataanalysis slow. Review collected by and hosted on G2.com.

sklearn provides consistent interface and the documentation is thorough. It is also highly extensible. Review collected by and hosted on G2.com.

I would prefer that cross_val_score provides a mechanism for out of sample evaluation. Assuming your sample is rebalanced, you may want the nth fold used for evaluation to be an unbalanced, out of sample dataset so as to the true performance of your model in the wild. cross_val_score does not provide this functionality. The pipeline class should also provide a mechanism to chain very many transformations and allow a grid search of best parameters across all the transformations. This is particularly useful in NLP pipeline where you stemming, removing stop words, ngram-ing, etc. could be a separate transformation and you want to know which transformation and parameters (e.g. the n in ngram) produced the best result. Review collected by and hosted on G2.com.