Upload 58 files

.gitattributes

@@ -40,3 +40,26 @@ output.mp3 filter=lfs diff=lfs merge=lfs -text
 requirements.txt filter=lfs diff=lfs merge=lfs -text
 sample.mp3 filter=lfs diff=lfs merge=lfs -text
 text2speech.mp3 filter=lfs diff=lfs merge=lfs -text
+models/char-clas/cz/Bi-RNN/model_2.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/char-clas/cz/Bi-RNN/model_2.meta filter=lfs diff=lfs merge=lfs -text
+models/char-clas/cz/CharClassifier.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/char-clas/en/Bi-RNN/model_1.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/char-clas/en/Bi-RNN/model_1.meta filter=lfs diff=lfs merge=lfs -text
+models/char-clas/en/CharClassifier.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/A-RNN/model.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/A-RNN/model.meta filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/Bi-RNN.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/CNN-CG.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/large/CNN-CG.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/RNN-A/Bi-RNN.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/RNN-A/Bi-RNN.meta filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/RNN/Bi-RNN-new.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/gap-clas/RNN/Bi-RNN-new.meta filter=lfs diff=lfs merge=lfs -text
+models/word-clas/CTC/Classifier1.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/word-clas/CTC/Classifier1.meta filter=lfs diff=lfs merge=lfs -text
+models/word-clas/en/CTC/Classifier2.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/word-clas/en/CTC/Classifier2.meta filter=lfs diff=lfs merge=lfs -text
+models/word-clas/en/SeqRNN/Classifier.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/word-clas/en/SeqRNN/Classifier.meta filter=lfs diff=lfs merge=lfs -text
+models/word-clas/en/WordClassifier2.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+models/word-clas/en/WordClassifier2.meta filter=lfs diff=lfs merge=lfs -text

models/README.md

@@ -0,0 +1,5 @@
+# Machine Learning Models
+You have to donwload pre-trained ML models before using some of the notebooks.
+
+Download and extract the models in this folder:
+<https://drive.google.com/file/d/1YbmsiJK3Wclfm6K8PrJuz-QROEKX1qis/view?usp=sharing>

models/char-clas/cz/Bi-RNN/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "model_2"
+all_model_checkpoint_paths: "model_2"

models/char-clas/cz/Bi-RNN/model_2.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c54a4b4c8d925bee557917ffa2d1f43befdc21035e8fe0b482caf2aa9051bf20
+size 145353836

models/char-clas/cz/Bi-RNN/model_2.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cae04a4752de5f50f2928d8eadb1fc0a68a6bd4b26dd936a94d5f4852b02532e
+size 1864590

models/char-clas/cz/CharClassifier.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:921c2981f453c8bd09a278142d10fe0b0ec54b34f00b6b99527bd7990e04ee0f
+size 6745052

models/char-clas/cz/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "CharClassifier"
+all_model_checkpoint_paths: "CharClassifier"

models/char-clas/en/Bi-RNN/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "model_1"
+all_model_checkpoint_paths: "model_1"

models/char-clas/en/Bi-RNN/model_1.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd3a4cb72afd75b88dafacf6de6e53086724b3bbdcc19a6f68f073a35952b1ed
+size 145169156

models/char-clas/en/Bi-RNN/model_1.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d7924276bbfa6a042d76638152276aa05d4a15061a9a28aab4e89bf231c8e89
+size 1861142

models/char-clas/en/CharClassifier.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79e8d85a7810fe30bfea4c9a474787ada5988a8b85202b5b1542b93e6de4159c
+size 4440692

models/char-clas/en/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "CharClassifier"
+all_model_checkpoint_paths: "CharClassifier"

models/freeze_graph.py

@@ -0,0 +1,263 @@
+# EDITED on 10. 9. 2017 for meta graph freezing
+#
+# Copyright 2015 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.
+# ==============================================================================
+"""Converts checkpoint variables into Const ops in a standalone GraphDef file.
+
+This script is designed to take a GraphDef proto, a SaverDef proto, and a set of
+variable values stored in a checkpoint file, and output a GraphDef with all of
+the variable ops converted into const ops containing the values of the
+variables.
+
+It's useful to do this when we need to load a single file in C++, especially in
+environments like mobile or embedded where we may not have access to the
+RestoreTensor ops and file loading calls that they rely on.
+
+An example of command-line usage is:
+bazel build tensorflow/python/tools:freeze_graph && \
+bazel-bin/tensorflow/python/tools/freeze_graph \
+--input_graph=some_graph_def.pb \
+--input_checkpoint=model.ckpt-8361242 \
+--output_graph=/tmp/frozen_graph.pb --output_node_names=softmax
+
+You can also look at freeze_graph_test.py for an example of how to use it.
+
+"""
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import argparse
+import sys
+
+from google.protobuf import text_format
+
+from tensorflow.contrib.saved_model.python.saved_model import reader
+from tensorflow.core.framework import graph_pb2
+from tensorflow.core.protobuf import saver_pb2
+from tensorflow.core.protobuf.meta_graph_pb2 import MetaGraphDef
+from tensorflow.python import pywrap_tensorflow
+from tensorflow.python.client import session
+from tensorflow.python.framework import graph_util
+from tensorflow.python.framework import importer
+from tensorflow.python.platform import app
+from tensorflow.python.platform import gfile
+from tensorflow.python.saved_model import loader
+from tensorflow.python.saved_model import tag_constants
+from tensorflow.python.training import saver as saver_lib
+
+FLAGS = None
+
+
+def freeze_graph_with_def_protos(input_graph_def,
+                                 input_saver_def,
+                                 input_checkpoint,
+                                 output_node_names,
+                                 restore_op_name,
+                                 filename_tensor_name,
+                                 output_graph,
+                                 clear_devices,
+                                 initializer_nodes,
+                                 variable_names_blacklist="",
+                                 input_meta_graph_def=None,
+                                 input_saved_model_dir=None,
+                                 saved_model_tags=None):
+    """Converts all variables in a graph and checkpoint into constants."""
+    del restore_op_name, filename_tensor_name  # Unused by updated loading code.
+
+    # 'input_checkpoint' may be a prefix if we're using Saver V2 format
+    if (not input_saved_model_dir and
+            not saver_lib.checkpoint_exists(input_checkpoint)):
+        print("Input checkpoint '" + input_checkpoint + "' doesn't exist!")
+        return -1
+
+    if not output_node_names:
+        print("You need to supply the name of a node to --output_node_names.")
+        return -1
+
+    # Remove all the explicit device specifications for this node. This helps to
+    # make the graph more portable.
+    if clear_devices:
+        if input_meta_graph_def:
+            for node in input_meta_graph_def.graph_def.node:
+                node.device = ""
+        elif input_graph_def:
+            for node in input_graph_def.node:
+                node.device = ""
+
+    if input_graph_def:
+        _ = importer.import_graph_def(input_graph_def, name="")
+    with session.Session() as sess:
+        if input_saver_def:
+            saver = saver_lib.Saver(saver_def=input_saver_def)
+            saver.restore(sess, input_checkpoint)
+        elif input_meta_graph_def:
+            restorer = saver_lib.import_meta_graph(
+                    input_meta_graph_def, clear_devices=True)
+            restorer.restore(sess, input_checkpoint)
+            if initializer_nodes:
+                sess.run(initializer_nodes.split(","))
+        elif input_saved_model_dir:
+            if saved_model_tags is None:
+                saved_model_tags = []
+            loader.load(sess, saved_model_tags, input_saved_model_dir)
+        else:
+            var_list = {}
+            reader = pywrap_tensorflow.NewCheckpointReader(input_checkpoint)
+            var_to_shape_map = reader.get_variable_to_shape_map()
+            for key in var_to_shape_map:
+                try:
+                    tensor = sess.graph.get_tensor_by_name(key + ":0")
+                except KeyError:
+                    # This tensor doesn't exist in the graph (for example it's
+                    # 'global_step' or a similar housekeeping element) so skip it.
+                    continue
+                var_list[key] = tensor
+            saver = saver_lib.Saver(var_list=var_list)
+            saver.restore(sess, input_checkpoint)
+            if initializer_nodes:
+                sess.run(initializer_nodes.split(","))
+
+        variable_names_blacklist = (variable_names_blacklist.split(",")
+                                                                if variable_names_blacklist else None)
+
+        if input_meta_graph_def:
+            output_graph_def = graph_util.convert_variables_to_constants(
+                    sess,
+                    input_meta_graph_def.graph_def,
+                    output_node_names.split(","),
+                    variable_names_blacklist=variable_names_blacklist)
+        else:
+            output_graph_def = graph_util.convert_variables_to_constants(
+                    sess,
+                    input_graph_def,
+                    output_node_names.split(","),
+                    variable_names_blacklist=variable_names_blacklist)
+
+    # Write GraphDef to file if output path has been given.
+    if output_graph:
+        with gfile.GFile(output_graph, "wb") as f:
+            f.write(output_graph_def.SerializeToString())
+
+    return output_graph_def
+
+
+def _parse_input_graph_proto(input_graph, input_binary):
+    """Parser input tensorflow graph into GraphDef proto."""
+    if not gfile.Exists(input_graph):
+        print("Input graph file '" + input_graph + "' does not exist!")
+        return -1
+    input_graph_def = graph_pb2.GraphDef()
+    mode = "rb" if input_binary else "r"
+    with gfile.FastGFile(input_graph, mode) as f:
+        if input_binary:
+            input_graph_def.ParseFromString(f.read())
+        else:
+            text_format.Merge(f.read(), input_graph_def)
+    return input_graph_def
+
+
+def _parse_input_meta_graph_proto(input_graph, input_binary):
+    """Parser input tensorflow graph into MetaGraphDef proto."""
+    if not gfile.Exists(input_graph):
+        print("Input meta graph file '" + input_graph + "' does not exist!")
+        return -1
+    input_meta_graph_def = MetaGraphDef()
+    mode = "rb" if input_binary else "r"
+    with gfile.FastGFile(input_graph, mode) as f:
+        if input_binary:
+            input_meta_graph_def.ParseFromString(f.read())
+        else:
+            text_format.Merge(f.read(), input_meta_graph_def)
+    print("Loaded meta graph file '" + input_graph)
+    return input_meta_graph_def
+
+
+def _parse_input_saver_proto(input_saver, input_binary):
+        """Parser input tensorflow Saver into SaverDef proto."""
+        if not gfile.Exists(input_saver):
+                print("Input saver file '" + input_saver + "' does not exist!")
+                return -1
+        mode = "rb" if input_binary else "r"
+        with gfile.FastGFile(input_saver, mode) as f:
+                saver_def = saver_pb2.SaverDef()
+        if input_binary:
+                saver_def.ParseFromString(f.read())
+        else:
+                text_format.Merge(f.read(), saver_def)
+        return saver_def
+    
+
+def get_meta_graph_def(saved_model_dir, tag_set):
+    """Gets MetaGraphDef from SavedModel.
+
+    Returns the MetaGraphDef for the given tag-set and SavedModel directory.
+
+    Args:
+        saved_model_dir: Directory containing the SavedModel to inspect or execute.
+        tag_set: Group of tag(s) of the MetaGraphDef to load, in string format,
+                separated by ','. For tag-set contains multiple tags, all tags must be
+                passed in.
+
+    Raises:
+        RuntimeError: An error when the given tag-set does not exist in the
+        SavedModel.
+
+    Returns:
+        A MetaGraphDef corresponding to the tag-set.
+    """
+    saved_model = reader.read_saved_model(saved_model_dir)
+    set_of_tags = set(tag_set.split(','))
+    for meta_graph_def in saved_model.meta_graphs:
+        if set(meta_graph_def.meta_info_def.tags) == set_of_tags:
+            return meta_graph_def
+
+    raise RuntimeError('MetaGraphDef associated with tag-set ' + tag_set +
+                       ' could not be found in SavedModel')
+
+
+def freeze_graph(input_graph,
+                 input_saver,
+                 input_binary,
+                 input_checkpoint,
+                 output_node_names,
+                 restore_op_name,
+                 filename_tensor_name,
+                 output_graph,
+                 clear_devices,
+                 initializer_nodes,
+                 variable_names_blacklist="",
+                 input_meta_graph=None,
+                 input_saved_model_dir=None,
+                 saved_model_tags=tag_constants.SERVING):
+        """Converts all variables in a graph and checkpoint into constants."""
+        input_graph_def = None
+        if input_saved_model_dir:
+                input_graph_def = get_meta_graph_def(
+                        input_saved_model_dir, saved_model_tags).graph_def
+        elif input_graph:
+                input_graph_def = _parse_input_graph_proto(input_graph, input_binary)
+        input_meta_graph_def = None
+        if input_meta_graph:
+                input_meta_graph_def = _parse_input_meta_graph_proto(
+                        input_meta_graph, input_binary)
+        input_saver_def = None
+        if input_saver:
+                input_saver_def = _parse_input_saver_proto(input_saver, input_binary)
+        freeze_graph_with_def_protos(
+                input_graph_def, input_saver_def, input_checkpoint, output_node_names,
+                restore_op_name, filename_tensor_name, output_graph, clear_devices,
+                initializer_nodes, variable_names_blacklist, input_meta_graph_def,
+                input_saved_model_dir, saved_model_tags.split(","))

models/gap-clas/A-RNN/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "model"
+all_model_checkpoint_paths: "model"

models/gap-clas/A-RNN/model.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb9340ba3f9156f671ba3e78d61f280d32195f31d287e2cb4b2e8121a39ef3f2
+size 20459744

models/gap-clas/A-RNN/model.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87cd2cb0b4e89914887127c607bb3240f272feb966afa7635647945c45a73f54
+size 2161008

models/gap-clas/Bi-RNN.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e3e67304611c52f34efd64d27ace70ddfa8ea094be4b9e0a95754afcc050029
+size 63168532

models/gap-clas/CNN-CG.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d753209958f41092716721614ad07f29a8bab2957d23efdf30b8d5cae792c55
+size 36324872

models/gap-clas/RNN-A/Bi-RNN.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:840283cecc2cd25a4285cd2146852bdb91f44a8f5676c6cec2ef9dc2e038bc45
+size 35047968

models/gap-clas/RNN-A/Bi-RNN.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4f61f67c5ff6ed1888228d204539626c75cf4422b14ddf543f46ec57a43b352
+size 1946207

models/gap-clas/RNN-A/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "Bi-RNN"
+all_model_checkpoint_paths: "Bi-RNN"

models/gap-clas/RNN/Bi-RNN-new.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01cebebe4e81fc511b6224b2a003d35968aa25179e2095cf5412dd7f8df416d4
+size 66272288

models/gap-clas/RNN/Bi-RNN-new.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43da4e220fb7e11b9c244167acd975611c771c645c006943a2f4241448608867
+size 1812295

models/gap-clas/RNN/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "Bi-RNN-new"
+all_model_checkpoint_paths: "Bi-RNN-new"

models/gap-clas/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "CNN-CG"
+all_model_checkpoint_paths: "CNN-CG"

models/gap-clas/large/CNN-CG.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1ba8a17b60e3155406ea2a06d0dfcf13b3a51e4b219cf94258d41ba8ba6446b
+size 108104072

models/gap-clas/large/checkpoint

@@ -0,0 +1,2 @@
+model_checkpoint_path: "CNN-CG"
+all_model_checkpoint_paths: "CNN-CG"

models/graph_optimizer.py

@@ -0,0 +1,87 @@
+"""
+Usage:
+python graph_optimizer.py \
+--tf_path ../../tensorflow/ \
+--model_folder "path_to_the_model_folder" \
+--output_names "activation, accuracy" \
+--input_names "x"
+"""
+
+import os, argparse
+from subprocess import call
+
+import freeze_graph
+import tensorflow as tf
+
+dir = os.path.dirname(os.path.realpath(__file__))
+
+fr_name = "_frozen.pb"
+op_name = "_optimized.pb"
+
+
+def graph_freez(model_folder, output_names):
+    print("Model folder", model_folder)
+    checkpoint = tf.train.get_checkpoint_state(model_folder)
+    print(checkpoint)
+    checkpoint_path = checkpoint.model_checkpoint_path
+    output_graph_filename = checkpoint_path + fr_name
+
+    input_saver_def_path = ""
+    input_binary = True
+    output_node_names = output_names
+    restore_op_name = "save/restore_all"
+    filename_tensor_name = "save/Const:0"
+    clear_devices = False
+    input_meta_graph = checkpoint_path + ".meta"
+
+    freeze_graph.freeze_graph(
+        "", input_saver_def_path, input_binary, checkpoint_path,
+        output_node_names, restore_op_name, filename_tensor_name,
+        output_graph_filename, clear_devices, "", "", input_meta_graph)
+
+    return output_graph_filename
+
+
+def graph_optimization(tf_path, graph_file, input_names, output_names):
+    output_file = graph_file[:-len(fr_name)] + op_name
+    tf_path += "bazel-bin/tensorflow/tools/graph_transforms/transform_graph"
+
+    call([tf_path,
+          "--in_graph=" + graph_file,
+          "--out_graph=" + output_file,
+          "--inputs=" + input_names,
+          "--outputs=" + output_names,
+          """--transforms=
+          strip_unused_nodes(type=float, shape="1,299,299,3")
+          fold_constants(ignore_errors=true)
+          fold_batch_norms
+          fold_old_batch_norms"""])
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(
+            "Script freezes graph and optimize it for mobile usage")
+    parser.add_argument(
+        "--model",
+        type=str,
+        help="Path of folder + model name (folder_path/model_name)")
+    parser.add_argument(
+        "--input_names",
+        type=str,
+        default="",
+        help="Input node names, comma separated.")
+    parser.add_argument(
+        "--output_names",
+        type=str,
+        default="",
+        help="Output node names, comma separated.")
+    parser.add_argument(
+        "--tf_path",
+        type=str,
+        default="../../tensorflow/",
+        help="Path to the folder with tensorflow (requires bazel build of graph_transforms)")
+
+    args = parser.parse_args()
+
+    graph = graph_freez(args.model, args.output_names)
+    graph_optimization(args.tf_path, graph, args.input_names, args.output_names)

models/word-clas/CTC/Classifier1.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8257c216932dd878023da0a28719f3aa09ba5cf1ff6d9187fe09a9488f0f784
+size 304709420

models/word-clas/CTC/Classifier1.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64a0626ed8bb9aab20faf54b1b8a697ddd932c3ce0dd9d7428d0455c7a784350
+size 1844874

models/word-clas/CTC/checkpoint

@@ -0,0 +1,6 @@
+model_checkpoint_path: "Classifier2-16500"
+all_model_checkpoint_paths: "Classifier2-14500"
+all_model_checkpoint_paths: "Classifier2-15000"
+all_model_checkpoint_paths: "Classifier2-15500"
+all_model_checkpoint_paths: "Classifier2-16000"
+all_model_checkpoint_paths: "Classifier2-16500"

models/word-clas/en/CTC/Classifier2.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f40c80e41ea158725a366ebbed70ddd31d367eabee3334792d71c434e2518f7
+size 180905808

models/word-clas/en/CTC/Classifier2.meta

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89c4a6dfd6ed2eed76259a9b162e7b75cb45a397447c449185bb98e37d6c06b7
+size 1165772