Lightgbm could not convert string to float. See if that works for you.

Lightgbm could not convert string to float. See if that works for you.

Lightgbm could not convert string to float. I'd like to use Shap (Shapley) to interpret features. 3. This model parsing is only needed for the interaction values so I moved it to be only done when computing interaction values (not regular SHAP values). I got an error from Shap: ValueError: could not convert string to float: 'Yes'. Tested and it seems to work for me. However, Shap gave me errors on categorical features. See if that works for you. Am I missing any settings? Jun 3, 2022 · There are many things that could be happening here and it's very hard to narrow down the possible cause from the current description. May 14, 2022 · My model uses LGBMClassifier. Also you're using an old LightGBM version, does the problem persist if you update it to the latest one (3. For example, I have a feature "Smoker" and its values include "Yes" and "No". 2)? Dec 19, 2020 · From what I can see in your error, it looks like the error may stem from passing a data set to your model_predictor that has strings in the gender section and not numbers. Aug 22, 2018 · I tried one hot encoding and it worked by it is not dealing with these large categories. Jul 21, 2018 · It looks like the model loading does not handle the categorical features right now. In addition, I want to draw a decision tree so doctors can understand it; in case of hot encoding, the classes are coded and it is hard to read Sep 10, 2017 · I try to use the scikit-learn api for categorical handling in lightgbm with the latest master branch: Oct 5, 2021 · Hello, I'm having this error when I try to generate shap_values for a LightGBM model that have categorical input features: Here you have a simple code to reproduce the warning:. gmknc pelvgmx gkrwzb kntqv fngggt myvjt kuoh lnbm dms gqyivl