Is the datarobot tool similar to the lime package in R, in that it will run your data through multiple models and give them scores for which one is best based off of your data?
Similar, yes, but the depth of models is much greater. Plus the guardrails to make sure the testing and validation is done correctly are a huge win. Plus, you have a deployable model at the end with drift tracking and MLOps.
Great question Michael. I've run a number of different data sets through DataRobot and I've been able to get predictions above 0.8 without any 'taget leakage'. I work a lot with healthcare data that tends to have social and behavorial components. DataRobot outputs what they call "Prediction Explanations". These explanations are the reasons why the chosen algorithm gave the probility it gave to a record. I find these explanations incredibly useful when reporting back the results because they give true insight to the data. I hope my answer helps.
![ไลฟ์สุ่ม iPhone 295 บาท... โกงมั้ย?[ โกงมั้ยครับ ep.97 ] | DOM](http://i.ytimg.com/vi/gdstHEE4UUE/mqdefault.jpg)
Is the datarobot tool similar to the lime package in R, in that it will run your data through multiple models and give them scores for which one is best based off of your data?
Similar, yes, but the depth of models is much greater. Plus the guardrails to make sure the testing and validation is done correctly are a huge win. Plus, you have a deployable model at the end with drift tracking and MLOps.
Will this also work properly with machine data for sensors, where accuracy should be more than 95
Does the accuracy top out around 0.8 using automated solutions like DataRobot?
Great question Michael. I've run a number of different data sets through DataRobot and I've been able to get predictions above 0.8 without any 'taget leakage'. I work a lot with healthcare data that tends to have social and behavorial components. DataRobot outputs what they call "Prediction Explanations". These explanations are the reasons why the chosen algorithm gave the probility it gave to a record. I find these explanations incredibly useful when reporting back the results because they give true insight to the data. I hope my answer helps.