def predict_load_data(
self,
csv_file: PATH_TYPE,
input_key: str,
root: Optional[PATH_TYPE] = None,
resolver: Optional[Callable[[Optional[PATH_TYPE], Any], PATH_TYPE]] = None,
clip_sampler: Union[str, "ClipSampler"] = "random",
clip_duration: float = 2,
clip_sampler_kwargs: Dict[str, Any] = None,
decode_audio: bool = False,
decoder: str = "pyav",
) -> List[str]:
data_frame = read_csv(csv_file)
if root is None:
root = os.path.dirname(csv_file)
return super().predict_load_data(
data_frame,
input_key,
root,
resolver,
clip_sampler=clip_sampler,
clip_duration=clip_duration,
clip_sampler_kwargs=clip_sampler_kwargs,
decode_audio=decode_audio,
decoder=decoder,
)
def load_data(
self,
csv_file: PATH_TYPE,
input_key: str,
target_keys: Optional[Union[str, List[str]]] = None,
root: Optional[PATH_TYPE] = None,
resolver: Optional[Callable[[Optional[PATH_TYPE], Any],
PATH_TYPE]] = None,
sampling_rate: int = 16000,
n_fft: int = 400,
target_formatter: Optional[TargetFormatter] = None,
) -> List[Dict[str, Any]]:
data_frame = read_csv(csv_file)
if root is None:
root = os.path.dirname(csv_file)
return super().load_data(
data_frame,
input_key,
target_keys,
root,
resolver,
sampling_rate=sampling_rate,
n_fft=n_fft,
target_formatter=target_formatter,
)
def load_data(
self,
csv_file: PATH_TYPE,
input_key: str,
target_keys: Optional[Union[str, List[str]]] = None,
root: Optional[PATH_TYPE] = None,
resolver: Optional[Callable[[Optional[PATH_TYPE], Any], PATH_TYPE]] = None,
clip_sampler: Union[str, "ClipSampler"] = "random",
clip_duration: float = 2,
clip_sampler_kwargs: Dict[str, Any] = None,
video_sampler: Type[Sampler] = torch.utils.data.RandomSampler,
decode_audio: bool = False,
decoder: str = "pyav",
target_formatter: Optional[TargetFormatter] = None,
) -> "LabeledVideoDataset":
data_frame = read_csv(csv_file)
if root is None:
root = os.path.dirname(csv_file)
return super().load_data(
data_frame,
input_key,
target_keys,
root,
resolver,
clip_sampler=clip_sampler,
clip_duration=clip_duration,
clip_sampler_kwargs=clip_sampler_kwargs,
video_sampler=video_sampler,
decode_audio=decode_audio,
decoder=decoder,
target_formatter=target_formatter,
)
def load_data(
self,
file: Optional[str],
categorical_fields: Optional[Union[str, List[str]]] = None,
numerical_fields: Optional[Union[str, List[str]]] = None,
target_field: Optional[str] = None,
parameters: Dict[str, Any] = None,
):
if file is not None:
return super().load_data(read_csv(file), categorical_fields,
numerical_fields, target_field,
parameters)
def serve_load_sample(self, data: str) -> Any:
parameters = self._parameters
df = read_csv(StringIO(data))
df = _pre_transform(
df,
parameters["numerical_fields"],
parameters["categorical_fields"],
parameters["codes"],
parameters["mean"],
parameters["std"],
)
cat_vars = _to_cat_vars_numpy(df, parameters["categorical_fields"])
num_vars = _to_num_vars_numpy(df, parameters["numerical_fields"])
cat_vars = np.stack(cat_vars, 1)
num_vars = np.stack(num_vars, 1)
return [{DataKeys.INPUT: [c, n]} for c, n in zip(cat_vars, num_vars)]