def build_playback_archive():
filename = _G.PlotPlaybackFilename
if os.path.exists(filename):
print(f"Archive {filename} already exists")
return _G.load_data(filename)
print(f"Building archive playback for '{filename}'")
files = sorted(glob_plots(filename))
cur_timestamp = 0
data = []
_len = len(files)
for i, file in enumerate(files):
print(f"Processing {i}/{_len}")
dat = PlotPlaybackRecord(cur_timestamp)
dat.sx, dat.ex = find_plot_window_length(file)
data.append(dat)
cur_timestamp += _G.TimeWindowSize
_G.dump_data(data, filename)
print("Archived dumped")
return data
def start_analyze():
global data
print(f"Analyzing stream file index of {_G.StreamFileIndex}")
data = []
if _G.FLAG_POSITIVE_PROC:
print(f"Analyzing Positive Samples")
_G.ensure_dir_exist(f"{_G.PositiveSamplePath}/.")
_G.wait(1)
files = _G.positive_audios()
for i, file in enumerate(files):
analyze_and_plot_audio(file, _G.positive_plot_filename(i), True)
_G.dump_data(data, _G.make_positive_dataname(i))
elif _G.FLAG_NEGATIVE_PROC:
print(f"Analyzing Negative Samples")
_G.ensure_dir_exist(f"{_G.NegativeSamplePath}/.")
_G.wait(1)
files = _G.negative_audios()
for i, file in enumerate(files):
analyze_and_plot_audio(file, _G.negative_plot_filename(i), True)
_G.dump_data(data, _G.make_negative_dataname(i))
else:
_G.ensure_dir_exist(_G.plot_filename(0))
files = get_audio_files(_G.StreamFilePrefix, _G.StreamFileSuffix)
flen = len(files)
for i, file in enumerate(files):
print(f"Analyzing {i+1}/{flen}")
analyze_and_plot_audio(file, _G.plot_filename(i), True)
# if i >= 2:
# break
_G.dump_data(data, _G.get_stream_adump_filename())
verbose=VERBOSE,
n_jobs=N_JOBS)
x_train = np.array(x_train, dtype=object)
x_train = x_train.reshape(x_train.shape[0],
x_train.shape[1] * x_train.shape[2])
print(f"Reshaped size: {x_train.shape}")
if TRAIN_SVM:
print("Training SVM")
clsier_svm.fit(x_train, y_train)
print("Best params: ", clsier_svm.best_params_)
print("Result:")
pprint(clsier_svm.cv_results_)
print("Dumping SVM data")
_G.dump_data(clsier_svm, f"svm_zcr.mod")
if TRAIN_KNN:
print("Training KNN")
clsier_knn.fit(x_train, y_train)
print("Best params: ", clsier_knn.best_params_)
print("Result:")
pprint(clsier_knn.cv_results_)
print("Dumping KNN data")
_G.dump_data(clsier_knn, f"knn_zcr.mod")
if TRAIN_RFR:
print("Training Random Forest")
clsier_rfr.fit(x_train, y_train)
print("Best params: ", clsier_rfr.best_params_)
print("Result:")
clsier_svm = GridSearchCV(estimator=svm.SVC(), param_grid=parm_svm, scoring='accuracy',cv=GRID_CV,verbose=VERBOSE,n_jobs=N_JOBS)
clsier_knn = GridSearchCV(estimator=KNeighborsClassifier(), param_grid=parm_knn, scoring='accuracy',cv=GRID_CV,verbose=VERBOSE,n_jobs=N_JOBS)
clsier_rfr = GridSearchCV(estimator=RandomForestRegressor(), param_grid=parm_rfr, scoring='explained_variance',cv=GRID_CV,verbose=VERBOSE,n_jobs=N_JOBS)
x_train = np.array(x_train, dtype=object)
x_train = x_train.reshape(x_train.shape[0], x_train.shape[1]*x_train.shape[2])
print(f"x_traine reshped: {x_train.shape}")
if TRAIN_SVM:
print("Training SVM")
clsier_svm.fit(x_train, y_train)
print("Best params: ", clsier_svm.best_params_)
print("Result:")
pprint(clsier_svm.cv_results_)
print("Dumping SVM")
_G.dump_data(clsier_svm, "svm_mfcc.mod")
if TRAIN_KNN:
print("Training KNN")
clsier_knn.fit(x_train, y_train)
print("Best params: ", clsier_knn.best_params_)
print("Result:")
pprint(clsier_knn.cv_results_)
_G.dump_data(clsier_knn, f"knn_mfcc.mod")
if TRAIN_RFR:
print("Training Random Forest")
clsier_rfr.fit(x_train, y_train)
print("Best params: ", clsier_rfr.best_params_)
print("Result:")
pprint(clsier_rfr.cv_results_)
verbose=VERBOSE,
n_jobs=N_JOBS)
x_train = np.array(x_train, dtype=object)
x_train = x_train.reshape(x_train.shape[0],
x_train.shape[1] * x_train.shape[2])
print(f"Reshaped size: {x_train.shape}")
if TRAIN_SVM:
print("Training SVM")
clsier_svm.fit(x_train, y_train)
print("Best params: ", clsier_svm.best_params_)
print("Result:")
pprint(clsier_svm.cv_results_)
print("Dumping SVM data")
_G.dump_data(clsier_svm, f"svm_rolloff.mod")
if TRAIN_KNN:
print("Training KNN")
clsier_knn.fit(x_train, y_train)
print("Best params: ", clsier_knn.best_params_)
print("Result:")
pprint(clsier_knn.cv_results_)
print("Dumping KNN data")
_G.dump_data(clsier_knn, f"knn_rolloff.mod")
if TRAIN_RFR:
print("Training Random Forest")
clsier_rfr.fit(x_train, y_train)
print("Best params: ", clsier_rfr.best_params_)
print("Result:")