Sync from tflite-micro at 76663ed.

Signed-off-by: CFU-Playground-Bot <cfu-playground-bot@google.com>

conf/tflite-micro.version

@@ -1 +1 @@
-b671143
+76663ed

third_party/tflite-micro/tensorflow/lite/kernels/internal/reference/hard_swish.h

@@ -12,8 +12,8 @@ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
 ==============================================================================*/
-#ifndef TENSORFLOW_LITE_KERNELS_INTERNAL_REFERENCE_ACTIVATIONS_H_
-#define TENSORFLOW_LITE_KERNELS_INTERNAL_REFERENCE_ACTIVATIONS_H_
+#ifndef TENSORFLOW_LITE_KERNELS_INTERNAL_REFERENCE_HARD_SWISH_H_
+#define TENSORFLOW_LITE_KERNELS_INTERNAL_REFERENCE_HARD_SWISH_H_
 
 #include <algorithm>
 
@@ -165,4 +165,4 @@ inline void HardSwish(const HardSwishParams& params,
 }  // namespace reference_ops
 }  // namespace tflite
 
-#endif  // TENSORFLOW_LITE_KERNELS_INTERNAL_REFERENCE_CONV_H_
+#endif  // TENSORFLOW_LITE_KERNELS_INTERNAL_REFERENCE_HARD_SWISH_H_

third_party/tflite-micro/tensorflow/lite/micro/examples/magic_wand/train/train_magic_wand_model.ipynb

@@ -74,9 +74,9 @@
       "outputs": [],
       "source": [
         "# Clone the repository from GitHub\n",
-        "!git clone --depth 1 -q https://github.com/tensorflow/tensorflow\n",
+        "!git clone --depth 1 -q https://github.com/tensorflow/tflite-micro\n",
         "# Copy the training scripts into our workspace\n",
-        "!cp -r tensorflow/tensorflow/lite/micro/examples/magic_wand/train train"
+        "!cp -r tflite-micro/tensorflow/lite/micro/examples/magic_wand/train train"
       ]
     },
     {

third_party/tflite-micro/tensorflow/lite/micro/examples/mnist_lstm/BUILD

@@ -0,0 +1,11 @@
+load("@tflm_pip_deps//:requirements.bzl", "requirement")
+
+py_binary(
+    name = "train",
+    srcs = ["train.py"],
+    srcs_version = "PY3",
+    deps = [
+        requirement("numpy"),
+        requirement("tensorflow-cpu"),
+    ],
+)

third_party/tflite-micro/tensorflow/lite/micro/examples/mnist_lstm/train.py

@@ -0,0 +1,148 @@
+# Copyright 2022 The TensorFlow Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =============================================================================
+"""
+LSTM model training for MNIST recognition
+
+This script is based on:
+https://www.tensorflow.org/lite/models/convert/rnn
+https://colab.research.google.com/github/tensorflow/tensorflow/blob/master/tensorflow/lite/examples/experimental_new_converter/Keras_LSTM_fusion_Codelab.ipynb
+
+Run:
+`bazel build tensorflow/lite/micro/examples/mnist_lstm:train`
+`bazel-bin/tensorflow/lite/micro/examples/mnist_lstm/train`
+"""
+import os
+
+from absl import app
+from absl import flags
+from absl import logging
+import numpy as np
+import tensorflow as tf
+
+FLAGS = flags.FLAGS
+
+flags.DEFINE_integer('epochs', 1, 'number of epochs to train the model.')
+flags.DEFINE_string('save_dir', '/tmp/lstm_trained_model',
+                    'the directory to save the trained model.')
+flags.DEFINE_boolean('save_tf_model', False,
+                     'store the original unconverted tf model.')
+
+
+def create_model(units=20):
+  """Create a keras LSTM model for MNIST recognition
+
+    Args:
+        units (int, optional): dimensionality of the output space for the model. Defaults to 20.
+
+    Returns:
+        tf.keras.Model: A Keras LSTM model
+    """
+
+  model = tf.keras.models.Sequential([
+      tf.keras.layers.Input(shape=(28, 28), name='input'),
+      tf.keras.layers.LSTM(units, return_sequences=True),
+      tf.keras.layers.Flatten(),
+      tf.keras.layers.Dense(10, activation=tf.nn.softmax, name='output')
+  ])
+  model.compile(optimizer='adam',
+                loss='sparse_categorical_crossentropy',
+                metrics=['accuracy'])
+  model.summary()
+  return model
+
+
+def get_train_data():
+  """Get MNIST train and test data
+
+    Returns:
+        list of tuples: (data, label) pairs for train and test
+    """
+  (x_train, y_train), _ = tf.keras.datasets.mnist.load_data()
+  x_train = x_train / 255.  # normalize pixel values to 0-1
+  x_train = x_train.astype(np.float32)
+  return (x_train, y_train)
+
+
+def train_lstm_model(epochs):
+  """Train keras LSTM model on MNIST dataset
+
+    Args:
+        epochs (int) : number of epochs to train the model
+
+    Returns:
+        tf.keras.Model: A trained keras LSTM model
+    """
+  model = create_model()
+  x_train, y_train = get_train_data()
+  callback = tf.keras.callbacks.EarlyStopping(
+      monitor='val_loss',
+      patience=1)  #early stop if validation loss does not drop anymore
+  model.fit(x_train,
+            y_train,
+            epochs=epochs,
+            validation_split=0.2,
+            batch_size=32,
+            callbacks=[callback])
+  return model
+
+
+def save_tflite_model(model, save_dir):
+  """Convert the saved TF model to tflite model, then save it as .tflite flatbuffer format
+
+    Args:
+        model (tf.keras.Model): the trained LSTM Model
+        save_dir (str): directory to save the model
+    """
+  converter = tf.lite.TFLiteConverter.from_keras_model(model)
+  save_name = 'lstm.tflite'
+  tflite_model = converter.convert()
+
+  if not os.path.exists(save_dir):
+    os.makedirs(save_dir)
+  with open(save_dir + '/' + save_name, 'wb') as f:
+    f.write(tflite_model)
+  logging.info("Tflite model saved to %s", save_dir)
+
+
+def train_save_model(save_dir, epochs=3, save_raw_model=False):
+  """train and save LSTM model using keras
+
+    Args:
+        save_dir (string): save directory for the trained model
+        epochs (int, optional): number of epochs to train the model. Defaults to 3
+        save_raw_model (bool): store the original unconverted tf model. Defaults to False
+    """
+  trained_model = train_lstm_model(epochs)
+
+  # converter requires fixed shape input to work, alternative: b/225231544
+  fixed_input = tf.keras.layers.Input(shape=[28, 28],
+                                      batch_size=1,
+                                      dtype=trained_model.inputs[0].dtype,
+                                      name='fixed_input')
+  fixed_output = trained_model(fixed_input)
+  run_model = tf.keras.models.Model(fixed_input, fixed_output)
+
+  if save_raw_model:
+    run_model.save(save_dir, save_format="tf")
+    logging.info("TF model saved to %s", save_dir)
+  save_tflite_model(run_model, save_dir)
+
+
+def main(_):
+  train_save_model(FLAGS.save_dir, FLAGS.epochs, FLAGS.save_tf_model)
+
+
+if __name__ == '__main__':
+  app.run(main)

third_party/tflite-micro/tensorflow/lite/micro/micro_interpreter.h

@@ -120,7 +120,7 @@ class MicroInterpreter {
   // created. i.e. after Init and Prepare is called for the very first time.
   TfLiteStatus Reset();
 
-  // TODO: remove this in favor of Reset()
+  // TODO(b/244457206): remove this in favor of Reset()
   // Reset all variable tensors to the default value.
   TfLiteStatus ResetVariableTensors();
 

third_party/tflite-micro/tensorflow/lite/micro/tools/make/targets/cortex_m_generic_makefile.inc

@@ -95,6 +95,12 @@ else ifeq ($(TARGET_ARCH), cortex-m85)
   # GCC does not yet support cortex-m85 option hence go with cortex-m55 for now.
   GCC_TARGET_ARCH := cortex-m55
 
+else ifeq ($(TARGET_ARCH), project_generation)
+  # No flags needed here as project_generation does not build anything.
+  # Make explicit there is to be no gcc download as project_generation does
+  # not build anything.
+  SKIP_ARM_GCC_DOWNLOAD = yes
+
 else
   $(error "TARGET_ARCH=$(TARGET_ARCH) is not supported")
 endif