def run_vmaf_cv_from_raw(train_dataset_raw_filepath, test_dataset_raw_filepath,
param_filepath, output_model_filepath, **kwargs):
if 'train_quality_wh' in kwargs and kwargs['train_quality_wh'] is not None:
train_quality_width, train_quality_height = kwargs['train_quality_wh']
else:
train_quality_width = None
train_quality_height = None
if 'test_quality_wh' in kwargs and kwargs['test_quality_wh'] is not None:
test_quality_width, test_quality_height = kwargs['test_quality_wh']
else:
test_quality_width = None
test_quality_height = None
if 'train_transform_final' in kwargs and kwargs[
'train_transform_final'] is not None:
train_transform_final = kwargs['train_transform_final']
else:
train_transform_final = None
if 'test_transform_final' in kwargs and kwargs[
'test_transform_final'] is not None:
test_transform_final = kwargs['test_transform_final']
else:
test_transform_final = None
train_output_dataset_filepath = VmafConfig.workspace_path(
'dataset', 'train_dataset.py')
generate_dataset_from_raw(
raw_dataset_filepath=train_dataset_raw_filepath,
output_dataset_filepath=train_output_dataset_filepath,
quality_width=train_quality_width,
quality_height=train_quality_height,
transform_final=train_transform_final,
**kwargs)
test_output_dataset_filepath = VmafConfig.workspace_path('dataset', 'test_dataset.py') \
if test_dataset_raw_filepath is not None else None
generate_dataset_from_raw(
raw_dataset_filepath=test_dataset_raw_filepath,
output_dataset_filepath=test_output_dataset_filepath,
quality_width=test_quality_width,
quality_height=test_quality_height,
transform_final=test_transform_final,
**kwargs)
run_vmaf_cv(train_dataset_filepath=train_output_dataset_filepath,
test_dataset_filepath=test_output_dataset_filepath,
param_filepath=param_filepath,
output_model_filepath=output_model_filepath,
**kwargs)
def run_vmaf_cv_from_raw(train_dataset_raw_filepath, test_dataset_raw_filepath,
param_filepath, output_model_filepath, **kwargs):
if 'train_quality_wh' in kwargs:
train_quality_width, train_quality_height = kwargs['train_quality_wh']
else:
train_quality_width = None
train_quality_height = None
if 'test_quality_wh' in kwargs:
test_quality_width, test_quality_height = kwargs['test_quality_wh']
else:
test_quality_width = None
test_quality_height = None
if 'train_transform_final' in kwargs:
train_transform_final = kwargs['train_transform_final']
else:
train_transform_final = None
if 'test_transform_final' in kwargs:
test_transform_final = kwargs['test_transform_final']
else:
test_transform_final = None
train_output_dataset_filepath = VmafConfig.workspace_path('dataset', 'train_dataset.py')
generate_dataset_from_raw(raw_dataset_filepath=train_dataset_raw_filepath,
output_dataset_filepath=train_output_dataset_filepath,
quality_width=train_quality_width,
quality_height=train_quality_height,
transform_final=train_transform_final,
**kwargs)
test_output_dataset_filepath = VmafConfig.workspace_path('dataset', 'test_dataset.py') \
if test_dataset_raw_filepath is not None else None
generate_dataset_from_raw(raw_dataset_filepath=test_dataset_raw_filepath,
output_dataset_filepath=test_output_dataset_filepath,
quality_width=test_quality_width,
quality_height=test_quality_height,
transform_final=test_transform_final,
**kwargs)
run_vmaf_cv(
train_dataset_filepath=train_output_dataset_filepath,
test_dataset_filepath=test_output_dataset_filepath,
param_filepath=param_filepath,
output_model_filepath=output_model_filepath,
**kwargs
)
def setUp(self):
train_dataset_path = VmafConfig.test_resource_path(
'test_image_dataset_diffdim.py')
train_dataset = import_python_file(train_dataset_path)
train_assets = read_dataset(train_dataset)
self.h5py_filepath = VmafConfig.workdir_path('test.hdf5')
self.h5py_file = DisYUVRawVideoExtractor.open_h5py_file(
self.h5py_filepath)
optional_dict2 = {'h5py_file': self.h5py_file}
_, self.features = run_executors_in_parallel(
DisYUVRawVideoExtractor,
train_assets,
fifo_mode=True,
delete_workdir=True,
parallelize=
False, # CAN ONLY USE SERIAL MODE FOR DisYRawVideoExtractor
result_store=None,
optional_dict=None,
optional_dict2=optional_dict2,
)
self.model_filename = VmafConfig.workspace_path(
"model", "test_save_load.pkl")
def setUp(self):
train_dataset_path = VmafConfig.test_resource_path(
'test_image_dataset.py')
train_dataset = import_python_file(train_dataset_path)
train_assets = read_dataset(train_dataset)
param = empty_object()
param.model_type = "NIQE"
param.model_param_dict = {
'patch_size': 96,
}
self.param = param
optional_dict = {'mode': 'train'}
_, self.features = run_executors_in_parallel(
NiqeNorefFeatureExtractor,
train_assets,
fifo_mode=True,
delete_workdir=True,
parallelize=True,
result_store=None,
optional_dict=optional_dict,
optional_dict2=None,
)
self.model_filename = VmafConfig.workspace_path(
'model', 'test_save_load.pkl')
def setUp(self):
train_dataset_path = VmafConfig.test_resource_path('test_image_dataset_diffdim.py')
train_dataset = import_python_file(train_dataset_path)
train_assets = read_dataset(train_dataset)
runner = MomentNorefFeatureExtractor(
train_assets,
None,
fifo_mode=True,
delete_workdir=True,
result_store=None,
optional_dict=None,
optional_dict2=None,
)
runner.run(parallelize=True)
self.features = runner.results
self.model_filename = VmafConfig.workspace_path("model", "test_save_load.pkl")
def setUp(self):
train_dataset_path = VmafConfig.test_resource_path('test_image_dataset_diffdim.py')
train_dataset = import_python_file(train_dataset_path)
train_assets = read_dataset(train_dataset)
self.h5py_filepath = VmafConfig.workdir_path('test.hdf5')
self.h5py_file = DisYUVRawVideoExtractor.open_h5py_file(self.h5py_filepath)
optional_dict2 = {'h5py_file': self.h5py_file}
runner = DisYUVRawVideoExtractor(
train_assets,
None,
fifo_mode=True,
delete_workdir=True,
result_store=None,
optional_dict=None,
optional_dict2=optional_dict2,
)
runner.run(parallelize=False) # CAN ONLY USE SERIAL MODE FOR DisYRawVideoExtractor
self.features = runner.results
self.model_filename = VmafConfig.workspace_path("model", "test_save_load.pkl")
def test_test_on_dataset_plot_per_content(self):
from vmaf.routine import run_test_on_dataset
test_dataset = import_python_file(
VmafConfig.test_resource_path('dataset_sample.py'))
import matplotlib.pyplot as plt
fig, ax = plt.subplots(1, 1, figsize=[20, 20])
run_test_on_dataset(test_dataset, VmafQualityRunner, ax,
None, VmafConfig.model_path("vmaf_float_v0.6.1.json"),
parallelize=False,
fifo_mode=False,
aggregate_method=None,
point_label='asset_id',
do_plot=['aggregate', # plots all contents in one figure
'per_content' # plots a separate figure per content
],
plot_linear_fit=True # adds linear fit line to each plot
)
output_dir = VmafConfig.workspace_path("output", "test_output")
DisplayConfig.show(write_to_dir=output_dir)
self.assertEqual(len(glob.glob(os.path.join(output_dir, '*.png'))), 3)
if os.path.exists(output_dir):
shutil.rmtree(output_dir)
def setUp(self):
self.output_model_filepath = VmafConfig.workspace_path(
"model", "test_output_model.pkl")
import matplotlib.pyplot as plt
import numpy as np
from vmaf.config import VmafConfig
from vmaf.routine import run_vmaf_cv, run_vmaf_kfold_cv
if __name__ == '__main__':
# ==== Run simple cross validation: one training and one testing dataset ====
run_vmaf_cv(
train_dataset_filepath=VmafConfig.resource_path('dataset', 'NFLX_dataset_public.py'),
test_dataset_filepath=VmafConfig.resource_path('dataset', 'VQEGHD3_dataset.py'),
param_filepath=VmafConfig.resource_path('param', 'vmaf_v3.py'),
output_model_filepath=VmafConfig.workspace_path('model', 'test_model1.pkl'),
)
# ==== Run cross validation across genres (tough test) ====
nflx_dataset_path = VmafConfig.resource_path('dataset', 'NFLX_dataset_public.py')
contentid_groups = [
[0, 5], # cartoon: BigBuckBunny, FoxBird
[1], # CG: BirdsInCage
[2, 6, 7], # complex: CrowdRun, OldTownCross, Seeking
[3, 4], # ElFuente: ElFuente1, ElFuente2
[8], # sports: Tennis
]
param_filepath = VmafConfig.resource_path('param', 'vmaf_v3.py')
aggregate_method = np.mean
import numpy as np
from vmaf.config import VmafConfig, DisplayConfig
from vmaf.routine import run_vmaf_cv, run_vmaf_kfold_cv
if __name__ == '__main__':
# ==== Run simple cross validation: one training and one testing dataset ====
run_vmaf_cv(
train_dataset_filepath=VmafConfig.resource_path(
'dataset', 'NFLX_dataset_public.py'),
test_dataset_filepath=VmafConfig.resource_path('dataset',
'VQEGHD3_dataset.py'),
param_filepath=VmafConfig.resource_path('param', 'vmaf_v3.py'),
output_model_filepath=VmafConfig.workspace_path(
'model', 'test_model1.pkl'),
)
# ==== Run cross validation across genres (tough test) ====
nflx_dataset_path = VmafConfig.resource_path('dataset',
'NFLX_dataset_public.py')
contentid_groups = [
[0, 5], # cartoon: BigBuckBunny, FoxBird
[1], # CG: BirdsInCage
[2, 6, 7], # complex: CrowdRun, OldTownCross, Seeking
[3, 4], # ElFuente: ElFuente1, ElFuente2
[8], # sports: Tennis
]
param_filepath = VmafConfig.resource_path('param', 'vmaf_v3.py')
def setUp(self):
self.output_model_filepath = VmafConfig.workspace_path("model", "test_output_model.pkl")
