Ranuga Disansa comments

Results 57 comments of


                                            Ranuga Disansa

trafficstars

Add Pytorch Lightning

Fix of issue #1826

Typo or misnaming in automl/timeseries/ts_model.py

Hi, I have created a PR #1312 feel free to check it :) Best regards

Where can I get the corresponding model for each cross validation

``` kf = KFold(n_splits=5, shuffle=True, random_state=42) model = RandomForestClassifier() for fold, (train_index, test_index) in enumerate(kf.split(X)): model.fit(X_train, y_train) ``` With the use of a code snippet similar to the above you...

Parameter tuning in final_estimator with automl ensemble

The parameters of the CatBoostClassifier are not automatically tuned. Set `passthrough` to False to automatically tune the parameters. Thank you

Update Optuna

Before transitioning to newer versions, it typically involves a substantial amount of effort. I'm planning to submit a pull request to update Optuna to version 2.10.0, but the rationale behind...

custom_metric() function error

According to #1217 you referred to, that issue is fixed with a new version `v2.1.1` of FLAML.. Please check if that fixes this issue Thank you

Need estimators like support vector machine, Decision Tree, Naïve Bayes Classifier.

> Custom estimator. Use custom estimator for: tuning an estimator that is not built-in; customizing search space for a built-in estimator. You have the ability to create your own custom...

Shifting of Time Series data

1. Shifted Model: Exogenous feature are shifted, an advantage is that it provides inputs that directly align with forecast on the other hand it reduces the dataset size. 2. Unshifted...

Represent "best error" in runtime logger instance using scientific notation?

I have created a PR #1305 which solves the issue mentioned.. Best regards, Ranuga

How to change value of max_iter in Logistic Regression in AutoML ?

Hi, This is a warning that was created when specifically in small datasets... In turn, this warning states that the algorithm has reached the solution. So thats simply the explanation...