Common problems in Data Scientist’s life: (1) which machine learning algorithm to choose and what hyper parameters values to use? (2) how to pre-process the data before analysis? (3) wrangling with OS dependent algorithms, which have different input format (4) long time waiting for results if running on single machine (5) which attributes are the most important? (6) what was the algorithm’s pipeline to obtain final result (provenance of the result)? (7) build from scratch your own infrastructure to deploy your model (8) awful sharing the results with collaborators by tons of emails
mljar.com was created to make Data Scientists live easier :) What does mljar do: (1) automatically train and tune machine learning models in parallel (2) search for data transformation that boost the performance (3) select the best match between machine learning algorithms and your data (4) assure provenance of the results (5) provide better understanding of your data with feature importance analysis (6) enable fast model deployment in the cloud (7) enable easy sharing of your results with your team and collaborators
With mljar.com: user need to upload data files, select which attributes to use and click ‘Compute’.