Internal change

PiperOrigin-RevId: 375132461

CHANGELOG.md

@@ -1,10 +1,11 @@
 # Changelog
 
-## 0.1.4 - ????
+## 0.1.4 - 2021-05-21
 
 ### Features
 
 -   Stop the training when interrupting a colab cell / typing ctrl-c.
+-   `model.fit` support training callbacks and a validation dataset.
 
 ### Bug fix
 

configure/setup.py

@@ -20,7 +20,7 @@
 from setuptools.command.install import install
 from setuptools.dist import Distribution
 
-_VERSION = "0.1.3"
+_VERSION = "0.1.4"
 
 with open("README.md", "r", encoding="utf-8") as fh:
   long_description = fh.read()

documentation/known_issues.md

@@ -51,29 +51,6 @@ numerical/categorical features would be 4 GB in memory.
 * Train a manual ensemble on slices of the dataset, i.e. train N models on N
 slices of data, and average the predictions.
 
-## No support for training callbacks.
-
-Training callbacks will not get the expected metrics passed to on_epoch_end
-since TF-DF algorithms are trained for only one epoch, and the validation
-data is evaluated before the model is trained. Evaluation callbacks are
-supported.
-
-**Workaround**
-
-By design TF-DF algorithms train for only one epoch, so callbacks that override
-on_epoch_end can be instantiated and called manually with the metrics from
-model.evaluate(). Specifically:
-
-```diff {.bad}
-- cb = tf.keras.callbacks.Callback()
-- model.fit(train_ds, validation_data=val_ds, callbacks=[cb])
-```
-
-```diff {.good}
-+ model.fit(train_ds)
-+ cb.on_epoch_end(epoch=1, logs=model.evaluate(val_ds, ...))
-```
-
 ## No support for GPU / TPU.
 
 TF-DF does not support GPU or TPU training. Compiling with AVX instructions,

tensorflow_decision_forests/__init__.py

@@ -45,7 +45,7 @@
 
 """
 
-__version__ = "0.1.3"
+__version__ = "0.1.4"
 __author__ = "Mathieu Guillame-Bert"
 
 from tensorflow_decision_forests import keras

tensorflow_decision_forests/tensorflow/ops/training/kernel.cc

@@ -644,7 +644,7 @@ std::atomic<bool> stop_training;
 void (*previous_signal_handler)(int);
 
 // Number of learners training.
-std::atomic<int> active_learners = 0;
+std::atomic<int> active_learners(0);
 
 void StopTrainingSignalHandler(int signal) { stop_training = true; }
 }  // namespace