-
为什么拿一折的train_idx进行特征选择后的256维特征,最后拿去进行5折交叉实验呢?不应该是每折实验都进行一次特征选择吗?
-
The core of model building is evaluation - does the new model improve on the old one? Does it generalize outside of the training data? Answering those questions requires an appropriate cross validatio…
-
I'm sorry for bothering you with another question. I have the case where I do
not need to use cross-validation (I can generate as much data as I want). This
means that the number of runs is equal to…
-
When you call compare_models with the fold parameter, it's not used when you eventually call create_model
```
best_model = compare_models(fold=n_folds,include=['lr','nb','knn','rf'])
~\AppData…
-
I'm a bit confused by the accuracy function in timm.utils.metrics, it takes the parameter topk? Does this specify number of fold cross validation?
-
**Describe the bug**
**ValueError: cross_val_predict only works for partitions**
When having 3 models, xgboost, catboost and decision tree trained using 'timeseries' cross-validation, then stack…
-
Hi,
I'm attempting to implement a non-standard metric to include in cross-validation. I have created a class for it. Sorry for length of exposition, I'm a little unsteady on my feet so want to give…
-
## 🐛 Bug
The wandb logger adds a 'global_step' to the metric dict which appears in all other loggers (e.g. Tensorboard). Only the wandb logger is adding 'global_step' to metric and I think it is no…
-
1. I emailed Moez about a month ago asking if you planned to add a neural network to your list of algorithms. I'll be talking to some AI-oriented physicians about PyCaret and they may ask
2. When I…
-
## Context
- In #601 and #602 we saw that our NER model still produces signficant errors in terms of `f1-score`.
- For instance, both `CELL_TYPE` and `CELL_COMPARTMENT` have `f1-score < 0.60`.
-…