Help protect the Great Barrier Reef with TensorFlow on Kaggle Join Challenge


Abstract inspector for all Yggdrasil models.

dataspec Gets the dataspec.
header Gets the generic model header.
metadata Gets the model's metadata.
task Task solved by the model.



View source

Model self evaluation.

The model self evaluation is a cheap alternative to the use of a separate validation dataset or cross-validation. The exact implementation depends on the model e.g. Out-of-bag evaluation, internal train-validation.

During training, some models (e.g. Gradient Boosted Tree) used this evaluation for early stopping (if early stopping is enabled).

While this evaluation is computed during training, it can be used as a low quality model evaluation.

The evaluation, or None is not evaluation is available.


View source

Export the training logs (and possibly other metadata) to TensorBoard.

Usage examples in Colab:

%load_ext tensorboard
%tensorboard --logdir "/tmp/tensorboard_logs"

Note that you can compare multiple models runs using sub-directories. For examples:


%load_ext tensorboard
%tensorboard --logdir "/tmp/tb_logs"

path Output directory for the logs.


View source

Input features of the model.


View source

Label predicted by the model.


View source

Possible classes of the label.

If the task is not a classification, or if the labels are dense integers, returns None.

The list of label values, or None.


View source

Unique key describing the type of the model.

Note that different learners can output similar model types, and a given learner can output different model types.


View source

Objective solved by the model i.e. Task + extra information.


View source

Evaluation metrics and statistics about the model during training.

The training logs show the quality of the model (e.g. accuracy evaluated on the out-of-bag or validation dataset) according to the number of trees in the model. Logs are useful to characterize the balance between model size and model quality.


View source

Variable importances (VI) i.e impact of each feature to the model.

VIs generally indicates how much a variable contributes to the model predictions or quality. Different VIs have different semantics and are generally not comparable.

The VIs returned by variable_importances() depends on the learning algorithm and its hyper-parameters. For example, the hyperparameter compute_oob_variable_importances=True of the Random Forest learner enables the computation of permutation out-of-bag variable importances.

See for the definition of the variable importances.

Values are sorted by decreasing value/importance except if stated otherwise.

Usage example:

# Train a Random Forest. Enable the computation of OOB (out-of-bag) variable
# importances.
model = tfdf.keras.RandomForestModel(compute_oob_variable_importances=True)

# Print all the variable importances

# List the available variable importances

# Show a specific variable importance
# Each line is: (feature name, (index of the feature), importance score)
>> [("bill_length_mm" (1; #1), 0.0713061951754389),
>> ("island" (4; #4), 0.007298519736842035),
>> ("flipper_length_mm" (1; #3), 0.004505893640351366),

Variable importances.