mendevi.models.linear_regression¶

Predicting absolute values using linear regression.

Classes

EncodeLinear(*args, **kwargs)

Biaised linear regression to predict parameters on encoding.

Details

class mendevi.models.linear_regression.EncodeLinear(*args: tuple, **kwargs: dict)[source]

Biaised linear regression to predict parameters on encoding.

Examples¶

>>> import pprint
>>> import cutcutcodec
>>> from mendevi.models.linear_regression import EncodeLinear
>>> model = EncodeLinear().fit("x264_vs_openh264.db", table="t_enc_encode")
>>> media = cutcutcodec.utils.get_project_root() / "media" / "video" / "intro.webm"
>>> pred = model.predict_from_video(
...     media, effort="medium", encoder="libx264", quality=0.5, threads=8, mode="vbr",
... )
>>> pprint.pprint(pred)
{'log_act_duration_per_frame': [-0.9902404546737671],
 'log_energy_per_frame': [0.5947314500808716],
 'log_rate': [6.39665412902832],
 'psnr': [37.500999450683594],
 'ssim': [0.8862448334693909],
 'vmaf': [82.45337677001953]}
>>>

Initialise the model.