dew.config.jepa
The JEPA run, as one typed record: I-JEPA over images, V-JEPA over video.
The encoder is the run’s model. The predictor takes the encoder’s width and
heads, and its own depth and width from predictor: a predictor as wide as
the encoder makes the objective too easy. Optional probes score the frozen
encoder beside the validation loss.
| Name | Summary |
|---|---|
JepaRunConfig | A run, plus the JEPA objective’s own knobs. |
JepaRunConfig
Section titled “JepaRunConfig”class JepaRunConfig( model: ModelConfig = (lambda: ModelConfig('jepa_encoder', {'precision': 'default'}))(), data: Annotated[Dataset | None, tyro.conf.Suppress] = None, validation: Annotated[Dataset | Mapping[str, Dataset] | None, tyro.conf.Suppress] = None, data_seed: int = 0, source: SourceOptions = SourceOptions(), optim: OptimConfig = (lambda: OptimConfig(learning_rate=0.001))(), trainer: TrainerConfig = TrainerConfig(), objective: ObjectiveConfig = (lambda: ObjectiveConfig('jepa'))(), lora: Annotated[LoRA, tyro.conf.subcommand(lora)] | None = None, val_metrics: Annotated[tuple[MetricConfig, ...], record_argument(tuple[MetricConfig, ...])] | None = None, val_every: Annotated[Mapping[str, int | str], record_argument(Mapping[str, int | str])] = dict(), val_summaries: Annotated[tuple[Summary, ...], record_argument(tuple[Summary, ...])] = (), predictor: JsonDict = dict(), num_target_blocks: int = 4, target_scale: tuple[float, float] = (0.15, 0.2), target_aspect: tuple[float, float] = (0.75, 1.5), probe_classes: int | None = None, knn_k: int = 20,)A run, plus the JEPA objective’s own knobs.
objective: ObjectiveConfig-
The JEPA objective and its arguments (
--objective.momentum 0.996 1.0). Its EMA ramps over the whole run unless it namesmomentum_steps. predictor: JsonDict-
Predictor kwargs, over the encoder’s shared ones.
probe_classes: int | None-
Number of classes for optional frozen-encoder probes; without them, validation scores the loss alone.
JepaRunConfig.sample
Section titled “JepaRunConfig.sample”def sample() -> FieldThe batch field the encoder reads, at the resolution the data comes in (Dataset.facts): a
clip’s video, else an image. Raises ValueError for a chain of neither, or one whose size
varies.
JepaRunConfig.build_run
Section titled “JepaRunConfig.build_run”def build_run() -> PreparedThe encoder, the predictor over its patch grid, the target-block mask and the probes, around the objective the run names. The predictor computes as the encoder does, its compute dtype and attention kernel.
JepaRunConfig.default_metrics
Section titled “JepaRunConfig.default_metrics”def default_metrics(*, validates: bool) -> tuple[MetricConfig, ...]A linear and a k-nearest-neighbour probe of the frozen encoder, where probe_classes names
the labels’ classes, whatever the data holds: a run without a validation split refuses them.