add some code

managed_components/78__esp-opus/scripts/dump_rnn.py

@@ -0,0 +1,57 @@
+#!/usr/bin/python
+
+from __future__ import print_function
+
+from keras.models import Sequential
+from keras.layers import Dense
+from keras.layers import LSTM
+from keras.layers import GRU
+from keras.models import load_model
+from keras import backend as K
+
+import numpy as np
+
+def printVector(f, vector, name):
+    v = np.reshape(vector, (-1));
+    #print('static const float ', name, '[', len(v), '] = \n', file=f)
+    f.write('static const opus_int16 {}[{}] = {{\n   '.format(name, len(v)))
+    for i in range(0, len(v)):
+        f.write('{}'.format(int(round(8192*v[i]))))
+        if (i!=len(v)-1):
+            f.write(',')
+        else:
+            break;
+        if (i%8==7):
+            f.write("\n   ")
+        else:
+            f.write(" ")
+    #print(v, file=f)
+    f.write('\n};\n\n')
+    return;
+
+def binary_crossentrop2(y_true, y_pred):
+        return K.mean(2*K.abs(y_true-0.5) * K.binary_crossentropy(y_pred, y_true), axis=-1)
+
+
+model = load_model("weights.hdf5", custom_objects={'binary_crossentrop2': binary_crossentrop2})
+
+weights = model.get_weights()
+
+f = open('rnn_weights.c', 'w')
+
+f.write('/*This file is automatically generated from a Keras model*/\n\n')
+f.write('#ifdef HAVE_CONFIG_H\n#include "config.h"\n#endif\n\n#include "mlp.h"\n\n')
+
+printVector(f, weights[0], 'layer0_weights')
+printVector(f, weights[1], 'layer0_bias')
+printVector(f, weights[2], 'layer1_weights')
+printVector(f, weights[3], 'layer1_recur_weights')
+printVector(f, weights[4], 'layer1_bias')
+printVector(f, weights[5], 'layer2_weights')
+printVector(f, weights[6], 'layer2_bias')
+
+f.write('const DenseLayer layer0 = {\n   layer0_bias,\n   layer0_weights,\n   25, 16, 0\n};\n\n')
+f.write('const GRULayer layer1 = {\n   layer1_bias,\n   layer1_weights,\n   layer1_recur_weights,\n   16, 12\n};\n\n')
+f.write('const DenseLayer layer2 = {\n   layer2_bias,\n   layer2_weights,\n   12, 2, 1\n};\n\n')
+
+f.close()

managed_components/78__esp-opus/scripts/local_build.py

@@ -0,0 +1,220 @@
+#!/usr/bin/python3
+
+"""
+This script tries to build a project with CMake, Meson, and Autotools.
+It checks if CMake, Meson, and Autotools are installed, and performs
+the configure, build, and optionally test steps for each build system.
+"""
+
+import multiprocessing
+import os
+import random
+import string
+import subprocess
+import sys
+import shutil
+
+
+if sys.platform == "win32":
+    REPO_DIR = "\\".join(os.path.abspath(__file__).split("\\")[:-2])
+else:
+    REPO_DIR = "/".join(os.path.abspath(__file__).split("/")[:-2])
+
+
+def main():
+    import argparse
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--test", action="store_true", help="Run tests")
+    parser.add_argument("--cmake", action="store_true", help="Only run CMake")
+    parser.add_argument("--meson", action="store_true", help="Only run Meson")
+    if sys.platform != "win32":
+        parser.add_argument("--autotools", action="store_true", help="Only run Autotools")
+    args = parser.parse_args()
+
+    test = args.test
+    cmake_only = args.cmake
+    meson_only = args.meson
+    if sys.platform != "win32":
+        autotools_only = args.autotools
+    autotools_only = False
+
+    result = []
+    os.chdir(REPO_DIR)
+
+    # download model
+    if sys.platform == "win32":
+        run_command("autogen.bat")
+    else:
+        run_command("./autogen.sh")
+
+    if sys.platform != "win32" and not cmake_only and not meson_only:
+        result += autotools_build(test)
+
+    if not autotools_only and not meson_only:
+        result += cmake_build(test)
+        result += cmake_build(test, extra_options="-DOPUS_NEURAL_FEC=ON")
+
+    if not autotools_only and not cmake_only:
+        result += meson_build(test)
+
+    print_result(result, test)
+
+def print_result(result, test=False):
+    if len(result) == 0:
+        print("No results available")
+        return
+
+    headers = ["Name", "Build Passed"]
+    if test:
+        headers.append("Test Passed")
+
+    # Calculate the maximum width for each column
+    column_widths = [max(len(str(row[i])) for row in result) for i in range(len(headers))]
+
+    # Print the headers
+    header_row = " | ".join(f"{header: <{column_widths[i]}}" for i, header in enumerate(headers))
+    print(header_row)
+    print("-" * len(header_row))
+
+    # Print the data rows
+    for row in result:
+        row_values = [str(value) for value in row[:len(headers)]]  # Include values up to the last column to be printed
+        row_string = " | ".join(f"{row_values[i]: <{column_widths[i]}}" for i in range(len(headers)))
+        print(row_string)
+
+def autotools_build(test=False):
+    build = "Autotools"
+    autotools_build_succeeded = False
+    autotools_test_succeeded = False
+
+    if not check_tool_installed("autoreconf") or not check_tool_installed("automake") or not check_tool_installed("autoconf"):
+        print("Autotools dependencies are not installed. Aborting.")
+        print("Install with: sudo apt-get install git autoconf automake libtool gcc make")
+        return [(build, autotools_build_succeeded, autotools_test_succeeded)]
+
+    run_command("./configure")
+    if run_command("make -j {}".format(get_cpu_core_count())) == 0:
+        autotools_build_succeeded = True
+    if test:
+        if run_command("make check -j {}".format(get_cpu_core_count())) == 0:
+            autotools_test_succeeded = True
+
+    return [(build, autotools_build_succeeded, autotools_test_succeeded)]
+
+
+def cmake_build(test=False, extra_options=""):
+    cmake_build_succeeded = False
+    cmake_test_succeeded = False
+    build = "CMake"
+
+    if not check_tool_installed("cmake"):
+        print("CMake is not installed. Aborting.")
+        if sys.platform != "win32":
+            print("Install with: sudo apt install cmake")
+        else:
+            print("Download and install from: https://cmake.org/download/")
+        return [(build, cmake_build_succeeded, cmake_test_succeeded)]
+
+    if not check_tool_installed("ninja"):
+        print("Ninja is not installed. Aborting.")
+        if sys.platform != "win32":
+            print("Install with: sudo apt ninja-build ")
+        else:
+            print("Download and install from: https://ninja-build.org/")
+        return [(build, cmake_build_succeeded, cmake_test_succeeded)]
+
+    cmake_build_dir = create_dir_with_random_postfix("cmake-build")
+    cmake_cfg_cmd = ["cmake", "-S" ".", "-B", cmake_build_dir, "-G", '"Ninja"', "-DCMAKE_BUILD_TYPE=Release", "-DOPUS_BUILD_TESTING=ON", "-DOPUS_BUILD_PROGRAMS=ON", "-DOPUS_FAST_MATH=ON", "-DOPUS_FLOAT_APPROX=ON"]
+    cmake_cfg_cmd += [option for option in extra_options.split(" ")]
+    run_command(" ".join(cmake_cfg_cmd))
+    cmake_build_cmd = ["cmake", "--build", cmake_build_dir, "-j", "{}".format(get_cpu_core_count())]
+    cmake_test_cmd = ["ctest", "-j", "{}".format(get_cpu_core_count())]
+    if sys.platform == "win32":
+        cmake_build_cmd += ["--config", "Release"]
+        cmake_test_cmd += ["-C", "Release"]
+
+    if run_command(" ".join(cmake_build_cmd)) == 0:
+        cmake_build_succeeded = True
+
+    if test:
+        os.chdir(cmake_build_dir)
+        if run_command(" ".join(cmake_test_cmd)) == 0:
+            cmake_test_succeeded = True
+        os.chdir(REPO_DIR)
+
+    shutil.rmtree(cmake_build_dir)
+
+    if extra_options != "":
+        build += " with options: "
+        build += extra_options.replace("-D", "")
+    return [(build, cmake_build_succeeded, cmake_test_succeeded)]
+
+def meson_build(test=False, extra_options=""):
+    build = "Meson"
+    meson_build_succeeded = False
+    meson_test_succeeded = False
+
+    if not check_tool_installed("meson"):
+        print("Meson is not installed. Aborting.")
+        print("Install with: pip install meson")
+        return [(build, meson_build_succeeded, meson_test_succeeded)]
+
+    if not check_tool_installed("ninja"):
+        print("Ninja is not installed. Aborting.")
+        if sys.platform != "win32":
+            print("Install with: sudo apt ninja-build ")
+        else:
+            print("Download and install from: https://ninja-build.org/")
+        return [(build, meson_build_succeeded, meson_test_succeeded)]
+
+    meson_build_dir = create_dir_with_random_postfix("meson-build")
+    meson_cfg_cmd = ["meson", "setup", meson_build_dir]
+    meson_cfg_cmd += [option for option in extra_options.split(" ")]
+    run_command(" ".join(meson_cfg_cmd))
+    meson_build_cmd = ["meson", "compile", "-C", meson_build_dir]
+    meson_test_cmd = ["meson", "test", "-C", meson_build_dir]
+
+    if run_command(" ".join(meson_build_cmd)) == 0:
+        meson_build_succeeded = True
+
+    if test:
+        os.chdir(meson_build_dir)
+        if run_command(" ".join(meson_test_cmd)) == 0:
+            meson_test_succeeded = True
+        os.chdir(REPO_DIR)
+
+    shutil.rmtree(meson_build_dir)
+
+    if extra_options != "":
+        build += " with options: "
+        build += extra_options.replace("-D", "")
+    return [(build, meson_build_succeeded, meson_test_succeeded)]
+
+def create_dir_with_random_postfix(dir):
+    random_chars = "".join(random.choices(string.ascii_letters, k=3))
+    dir = dir + "-" + random_chars
+    dir = os.path.join(os.getcwd(), dir)
+    os.makedirs(dir)
+    return dir
+
+def check_tool_installed(tool):
+    if sys.platform == "win32":
+        try:
+            # Use the "where" command to search for the executable
+            subprocess.check_output(["where", tool], shell=True)
+            return True
+        except subprocess.CalledProcessError:
+            return False
+    else:
+        return run_command(f"command -v {tool}") == 0
+
+def run_command(command):
+    process = subprocess.Popen(command, shell=True)
+    process.communicate()
+    return process.returncode
+
+def get_cpu_core_count():
+    return int(max(multiprocessing.cpu_count(), 1))
+
+if __name__ == "__main__":
+    main()

managed_components/78__esp-opus/scripts/rnn_train.py

@@ -0,0 +1,67 @@
+#!/usr/bin/python
+
+from __future__ import print_function
+
+from keras.models import Sequential
+from keras.models import Model
+from keras.layers import Input
+from keras.layers import Dense
+from keras.layers import LSTM
+from keras.layers import GRU
+from keras.layers import SimpleRNN
+from keras.layers import Dropout
+from keras import losses
+import h5py
+
+from keras import backend as K
+import numpy as np
+
+def binary_crossentrop2(y_true, y_pred):
+    return K.mean(2*K.abs(y_true-0.5) * K.binary_crossentropy(y_pred, y_true), axis=-1)
+
+print('Build model...')
+#model = Sequential()
+#model.add(Dense(16, activation='tanh', input_shape=(None, 25)))
+#model.add(GRU(12, dropout=0.0, recurrent_dropout=0.0, activation='tanh', recurrent_activation='sigmoid', return_sequences=True))
+#model.add(Dense(2, activation='sigmoid'))
+
+main_input = Input(shape=(None, 25), name='main_input')
+x = Dense(16, activation='tanh')(main_input)
+x = GRU(12, dropout=0.1, recurrent_dropout=0.1, activation='tanh', recurrent_activation='sigmoid', return_sequences=True)(x)
+x = Dense(2, activation='sigmoid')(x)
+model = Model(inputs=main_input, outputs=x)
+
+batch_size = 64
+
+print('Loading data...')
+with h5py.File('features.h5', 'r') as hf:
+    all_data = hf['features'][:]
+print('done.')
+
+window_size = 1500
+
+nb_sequences = len(all_data)/window_size
+print(nb_sequences, ' sequences')
+x_train = all_data[:nb_sequences*window_size, :-2]
+x_train = np.reshape(x_train, (nb_sequences, window_size, 25))
+
+y_train = np.copy(all_data[:nb_sequences*window_size, -2:])
+y_train = np.reshape(y_train, (nb_sequences, window_size, 2))
+
+all_data = 0;
+x_train = x_train.astype('float32')
+y_train = y_train.astype('float32')
+
+print(len(x_train), 'train sequences. x shape =', x_train.shape, 'y shape = ', y_train.shape)
+
+# try using different optimizers and different optimizer configs
+model.compile(loss=binary_crossentrop2,
+              optimizer='adam',
+              metrics=['binary_accuracy'])
+
+print('Train...')
+model.fit(x_train, y_train,
+          batch_size=batch_size,
+          epochs=200,
+          validation_data=(x_train, y_train))
+model.save("newweights.hdf5")

managed_components/78__esp-opus/scripts/shrink_model.sh

@@ -0,0 +1,7 @@
+#!/bin/sh
+
+for i in fargan_data.c pitchdnn_data.c dred_rdovae_dec_data.c dred_rdovae_enc_data.c plc_data.c lace_data.c nolace_data.c
+do
+    cat dnn/$i | perl -ne 'if (/DEBUG/ || /#else/) {$skip=1} if (!$skip && !/ifdef DOT_PROD/) {s/^ *//; s/, /,/g; print $_} elsif (/endif/) {$skip=0}' > tmp_data.c
+    mv tmp_data.c dnn/$i
+done