Convert Keras Core to Keras 3.

Author: fchollet

README.md

@@ -1,9 +1,9 @@
-[![](https://github.com/keras-team/keras-core/workflows/Tests/badge.svg?branch=main)](https://github.com/keras-team/keras-core/actions?query=workflow%3ATests+branch%3Amain)
-[![](https://badge.fury.io/py/keras-core.svg)](https://badge.fury.io/py/keras-core)
+[![](https://github.com/keras-team/keras/workflows/Tests/badge.svg?branch=main)](https://github.com/keras-team/keras/actions?query=workflow%3ATests+branch%3Amain)
+[![](https://badge.fury.io/py/keras.svg)](https://badge.fury.io/py/keras)
 
-# Keras Core: A new multi-backend Keras
+# Keras 3: A new multi-backend Keras
 
-Keras Core is a new multi-backend implementation of the Keras API, with support for TensorFlow, JAX, and PyTorch.
+Keras 3 is a new multi-backend implementation of the Keras API, with support for TensorFlow, JAX, and PyTorch.
 
 **WARNING:** At this time, this package is experimental.
 It has rough edges and not everything might work as expected.
@@ -13,7 +13,7 @@ Once ready, this package will become Keras 3.0 and subsume `tf.keras`.
 
 ## Local installation
 
-Keras Core is compatible with Linux and MacOS systems. To install a local development version:
+Keras 3 is compatible with Linux and MacOS systems. To install a local development version:
 
 1. Install dependencies:
 
@@ -28,7 +28,7 @@ python pip_build.py --install
 ```
 
 You should also install your backend of choice: `tensorflow`, `jax`, or `torch`.
-Note that `tensorflow` is required for using certain Keras Core features: certain preprocessing layers as
+Note that `tensorflow` is required for using certain Keras 3 features: certain preprocessing layers as
 well as `tf.data` pipelines.
 
 ## Configuring your backend
@@ -46,16 +46,16 @@ In Colab, you can do:
 import os
 os.environ["KERAS_BACKEND"] = "jax"
 
-import keras_core as keras
+import keras as keras
 ```
 
-**Note:** The backend must be configured before importing `keras_core`, and the backend cannot be changed after 
+**Note:** The backend must be configured before importing `keras`, and the backend cannot be changed after 
 the package has been imported.
 
 ## Backwards compatibility
 
-Keras Core is intended to work as a drop-in replacement for `tf.keras` (when using the TensorFlow backend). Just take your
-existing `tf.keras` code, change the `keras` imports to `keras_core`, make sure that your calls to `model.save()` are using
+Keras 3 is intended to work as a drop-in replacement for `tf.keras` (when using the TensorFlow backend). Just take your
+existing `tf.keras` code, change the `keras` imports to `keras`, make sure that your calls to `model.save()` are using
 the up-to-date `.keras` format, and you're done.
 
 If your `tf.keras` model does not include custom components, you can start running it on top of JAX or PyTorch immediately.
@@ -66,7 +66,7 @@ to a backend-agnostic implementation in just a few minutes.
 In addition, Keras models can consume datasets in any format, regardless of the backend you're using:
 you can train your models with your existing `tf.data.Dataset` pipelines or PyTorch `DataLoaders`.
 
-## Why use Keras Core?
+## Why use Keras 3?
 
 - Run your high-level Keras workflows on top of any framework -- benefiting at will from the advantages of each framework,
 e.g. the scalability and performance of JAX or the production ecosystem options of TensorFlow.

benchmarks/layer_benchmark/README.md

@@ -1,7 +1,7 @@
 # Benchmark the layer performance
 
 This directory contains benchmarks to compare the performance of
-`keras_core.layers.XXX` and `tf.keras.layers.XXX`. We compare the performance of
+`keras.layers.XXX` and `tf.keras.layers.XXX`. We compare the performance of
 both the forward pass and train step (forward & backward pass). 
 
 To run the benchmark, use the command below and change the flags according to

benchmarks/layer_benchmark/base_benchmark.py

@@ -4,7 +4,7 @@
 import tensorflow as tf
 from absl import flags
 
-import keras_core
+import keras
 
 FLAGS = flags.FLAGS
 
@@ -66,7 +66,7 @@ def on_predict_batch_end(self, batch, logs=None):
             self.state["throughput"] = throughput
 
 
-class KerasCoreBenchmarkMetricsCallback(keras_core.callbacks.Callback):
+class KerasCoreBenchmarkMetricsCallback(keras.callbacks.Callback):
     def __init__(self, start_batch=1, stop_batch=None):
         self._callback = BenchmarkMetricsCallback(start_batch, stop_batch)
 
@@ -108,36 +108,36 @@ def __init__(
         input_shape,
         flat_call_inputs=True,
         jit_compile=True,
-        keras_core_layer=None,
+        keras_layer=None,
         tf_keras_layer=None,
     ):
         self.layer_name = layer_name
-        _keras_core_layer_class = getattr(keras_core.layers, layer_name)
+        _keras_layer_class = getattr(keras.layers, layer_name)
         _tf_keras_layer_class = getattr(tf.keras.layers, layer_name)
 
-        if keras_core_layer is None:
-            # Sometimes you want to initialize the keras_core layer and tf_keras
+        if keras_layer is None:
+            # Sometimes you want to initialize the keras layer and tf_keras
             # layer in a different way. For example, `Bidirectional` layer,
-            # which takes in `keras_core.layers.Layer` and
+            # which takes in `keras.layers.Layer` and
             # `tf.keras.layer.Layer` separately.
-            self._keras_core_layer = _keras_core_layer_class(**init_args)
+            self._keras_layer = _keras_layer_class(**init_args)
         else:
-            self._keras_core_layer = keras_core_layer
+            self._keras_layer = keras_layer
 
         if tf_keras_layer is None:
             self._tf_keras_layer = _tf_keras_layer_class(**init_args)
         else:
             self._tf_keras_layer = tf_keras_layer
 
         self.input_shape = input_shape
-        self._keras_core_model = self._build_keras_core_model(
+        self._keras_model = self._build_keras_model(
             input_shape, flat_call_inputs
         )
         self._tf_keras_model = self._build_tf_keras_model(
             input_shape, flat_call_inputs
         )
 
-        self._keras_core_model.compile(
+        self._keras_model.compile(
             loss="mse", optimizer="sgd", jit_compile=jit_compile
         )
         self._tf_keras_model.compile(
@@ -148,19 +148,19 @@ def __init__(
         self.jit_compile = jit_compile
         self.input_shape = input_shape
 
-    def _build_keras_core_model(self, input_shape, flat_call_inputs=True):
+    def _build_keras_model(self, input_shape, flat_call_inputs=True):
         inputs = []
         if not isinstance(input_shape[0], (tuple, list)):
             input_shape = [input_shape]
 
         for shape in input_shape:
-            inputs.append(keras_core.Input(shape=shape))
+            inputs.append(keras.Input(shape=shape))
 
         if flat_call_inputs:
-            outputs = self._keras_core_layer(*inputs)
+            outputs = self._keras_layer(*inputs)
         else:
-            outputs = self._keras_core_layer(inputs)
-        return keras_core.Model(inputs=inputs, outputs=outputs)
+            outputs = self._keras_layer(inputs)
+        return keras.Model(inputs=inputs, outputs=outputs)
 
     def _build_tf_keras_model(self, input_shape, flat_call_inputs=True):
         inputs = []
@@ -195,7 +195,7 @@ def benchmark_predict(self, num_samples, batch_size, data=None):
             stop_batch=num_iterations
         )
 
-        self._keras_core_model.predict(
+        self._keras_model.predict(
             data,
             batch_size=batch_size,
             callbacks=[callback],
@@ -207,15 +207,15 @@ def benchmark_predict(self, num_samples, batch_size, data=None):
             callbacks=[tf_keras_callback],
         )
 
-        keras_core_throughput = (
+        keras_throughput = (
             callback._callback.state["throughput"] * batch_size
         )
         tf_keras_throughput = (
             tf_keras_callback._callback.state["throughput"] * batch_size
         )
         print(
-            f"Keras Core throughput of forward pass of {self.layer_name}: "
-            f"{keras_core_throughput:.2f} samples/sec."
+            f"Keras 3 throughput of forward pass of {self.layer_name}: "
+            f"{keras_throughput:.2f} samples/sec."
         )
         print(
             f"TF Keras throughput of forward pass of {self.layer_name}: "
@@ -240,15 +240,15 @@ def benchmark_train(self, num_samples, batch_size, data=None, label=None):
             if self.flat_call_inputs:
                 # Scale by a small factor to avoid zero gradients.
                 label = (
-                    keras_core.backend.convert_to_numpy(
-                        self._keras_core_layer(*data)
+                    keras.backend.convert_to_numpy(
+                        self._keras_layer(*data)
                     )
                     * 1.001
                 )
             else:
                 label = (
-                    keras_core.backend.convert_to_numpy(
-                        self._keras_core_layer(data)
+                    keras.backend.convert_to_numpy(
+                        self._keras_layer(data)
                     )
                     * 1.001
                 )
@@ -259,7 +259,7 @@ def benchmark_train(self, num_samples, batch_size, data=None, label=None):
             stop_batch=num_iterations
         )
 
-        self._keras_core_model.fit(
+        self._keras_model.fit(
             data,
             label,
             batch_size=batch_size,
@@ -272,15 +272,15 @@ def benchmark_train(self, num_samples, batch_size, data=None, label=None):
             callbacks=[tf_keras_callback],
         )
 
-        keras_core_throughput = (
+        keras_throughput = (
             callback._callback.state["throughput"] * batch_size
         )
         tf_keras_throughput = (
             tf_keras_callback._callback.state["throughput"] * batch_size
         )
         print(
-            f"Keras Core throughput of forward & backward pass of "
-            f"{self.layer_name}: {keras_core_throughput:.2f} samples/sec."
+            f"Keras 3 throughput of forward & backward pass of "
+            f"{self.layer_name}: {keras_throughput:.2f} samples/sec."
         )
         print(
             f"TF Keras  throughput of forward & backward pass of "

benchmarks/layer_benchmark/rnn_benchmark.py

@@ -16,7 +16,7 @@
 from absl import app
 from absl import flags
 
-import keras_core
+import keras
 from benchmarks.layer_benchmark.base_benchmark import LayerBenchmark
 
 FLAGS = flags.FLAGS
@@ -194,16 +194,16 @@ def benchmark_bidirectional(
 ):
     layer_name = "Bidirectional"
     init_args = {}
-    keras_core_layer = keras_core.layers.Bidirectional(
-        keras_core.layers.LSTM(32)
+    keras_layer = keras.layers.Bidirectional(
+        keras.layers.LSTM(32)
     )
     tf_keras_layer = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(32))
     benchmark = LayerBenchmark(
         layer_name,
         init_args,
         input_shape=[256, 256],
         jit_compile=jit_compile,
-        keras_core_layer=keras_core_layer,
+        keras_layer=keras_layer,
         tf_keras_layer=tf_keras_layer,
     )
 
@@ -225,8 +225,8 @@ def benchmark_time_distributed(
 ):
     layer_name = "TimeDistributed"
     init_args = {}
-    keras_core_layer = keras_core.layers.TimeDistributed(
-        keras_core.layers.Conv2D(16, (3, 3))
+    keras_layer = keras.layers.TimeDistributed(
+        keras.layers.Conv2D(16, (3, 3))
     )
     tf_keras_layer = tf.keras.layers.TimeDistributed(
         tf.keras.layers.Conv2D(16, (3, 3))
@@ -236,7 +236,7 @@ def benchmark_time_distributed(
         init_args,
         input_shape=[10, 32, 32, 3],
         jit_compile=jit_compile,
-        keras_core_layer=keras_core_layer,
+        keras_layer=keras_layer,
         tf_keras_layer=tf_keras_layer,
     )
 

benchmarks/model_benchmark/benchmark_utils.py

@@ -1,9 +1,9 @@
 import time
 
-import keras_core
+import keras
 
 
-class BenchmarkMetricsCallback(keras_core.callbacks.Callback):
+class BenchmarkMetricsCallback(keras.callbacks.Callback):
     def __init__(self, start_batch=1, stop_batch=None):
         self.start_batch = start_batch
         self.stop_batch = stop_batch

benchmarks/model_benchmark/bert_benchmark.py

@@ -21,7 +21,7 @@
 from absl import logging
 from model_benchmark.benchmark_utils import BenchmarkMetricsCallback
 
-import keras_core as keras
+import keras as keras
 
 flags.DEFINE_string("model_size", "small", "The size of model to benchmark.")
 flags.DEFINE_string(

benchmarks/model_benchmark/image_classification_benchmark.py

@@ -27,7 +27,7 @@
 from absl import logging
 from model_benchmark.benchmark_utils import BenchmarkMetricsCallback
 
-import keras_core as keras
+import keras as keras
 
 flags.DEFINE_string("model", "EfficientNetV2B0", "The model to benchmark.")
 flags.DEFINE_integer("epochs", 1, "The number of epochs.")

benchmarks/torch_ctl_benchmark/conv_model_benchmark.py

@@ -9,9 +9,9 @@
 import torch.nn as nn
 import torch.optim as optim
 
-import keras_core
+import keras
 from benchmarks.torch_ctl_benchmark.benchmark_utils import train_loop
-from keras_core import layers
+from keras import layers
 
 num_classes = 2
 input_shape = (3, 256, 256)
@@ -55,8 +55,8 @@ def forward(self, x):
         return x
 
 
-def run_keras_core_custom_training_loop():
-    keras_model = keras_core.Sequential(
+def run_keras_custom_training_loop():
+    keras_model = keras.Sequential(
         [
             layers.Input(shape=input_shape),
             layers.Conv2D(32, kernel_size=(3, 3), activation="relu"),
@@ -74,7 +74,7 @@ def run_keras_core_custom_training_loop():
         num_epochs=num_epochs,
         optimizer=optimizer,
         loss_fn=loss_fn,
-        framework="keras_core",
+        framework="keras",
     )
 
 
@@ -93,5 +93,5 @@ def run_torch_custom_training_loop():
 
 
 if __name__ == "__main__":
-    run_keras_core_custom_training_loop()
+    run_keras_custom_training_loop()
     run_torch_custom_training_loop()

benchmarks/torch_ctl_benchmark/dense_model_benchmark.py

@@ -9,9 +9,9 @@
 import torch.nn as nn
 import torch.optim as optim
 
-import keras_core
+import keras
 from benchmarks.torch_ctl_benchmark.benchmark_utils import train_loop
-from keras_core import layers
+from keras import layers
 
 num_classes = 2
 input_shape = (8192,)
@@ -55,8 +55,8 @@ def forward(self, x):
         return x
 
 
-def run_keras_core_custom_training_loop():
-    keras_model = keras_core.Sequential(
+def run_keras_custom_training_loop():
+    keras_model = keras.Sequential(
         [
             layers.Input(shape=input_shape),
             layers.Dense(64, activation="relu"),
@@ -73,7 +73,7 @@ def run_keras_core_custom_training_loop():
         num_epochs=num_epochs,
         optimizer=optimizer,
         loss_fn=loss_fn,
-        framework="keras_core",
+        framework="keras",
     )
 
 
@@ -92,5 +92,5 @@ def run_torch_custom_training_loop():
 
 
 if __name__ == "__main__":
-    run_keras_core_custom_training_loop()
+    run_keras_custom_training_loop()
     run_torch_custom_training_loop()