ml4gw.nn.streaming package

Submodules

ml4gw.nn.streaming.online_average module

class ml4gw.nn.streaming.online_average.OnlineAverager(update_size, batch_size, num_updates, num_channels, offset=None)

Bases: Module

Module for performing stateful online averaging of batches of overlapping timeseries. At present, the first num_updates predictions produced by this model will underestimate the true average.

Parameters:

• update_size (int) -- The number of samples separating the timestamps of subsequent inputs.

• batch_size (int) -- The number of batched inputs to expect at inference time.

• num_updates (int) -- The number of steps over which to average predictions before returning them.

• num_channels (int) -- The expected channel dimension of the input passed to the module at inference time.

• offset (Optional[int]) -- Number of samples to throw away from the front edge of the kernel when averaging.

forward(update, state=None)

Define the computation performed at every call.

Should be overridden by all subclasses. :rtype: Tuple[Tensor, Tensor]

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

get_initial_state()

ml4gw.nn.streaming.snapshotter module

class ml4gw.nn.streaming.snapshotter.Snapshotter(num_channels, snapshot_size, stride_size, batch_size, channels_per_snapshot=None)

Bases: Module

Model for converting streaming state updates into a batch of overlapping snaphots of a multichannel timeseries. Can support multiple timeseries in a single state update via the channels_per_snapshot kwarg.

Specifically, maps tensors of shape (num_channels, batch_size * stride_size) to a tensor of shape (batch_size, num_channels, snapshot_size). If channels_per_snapshot is specified, it will return len(channels_per_snapshot) tensors of this shape, with the channel dimension replaced by the corresponding value of channels_per_snapshot. The last tensor returned at call time will be the current state that can be passed to the next forward call.

Parameters:

• num_channels (int) -- Number of channels in the timeseries. If channels_per_snapshot is not None, this should be equal to sum(channels_per_snapshot).

• snapshot_size (int) -- The size of the output snapshot windows in number of samples

• stride_size (int) -- The number of samples in between each output snapshot

• batch_size (int) -- The number of snapshots to produce at each update. The last dimension of the input tensor should have size batch_size * stride_size.

• channels_per_snapshot (Optional[Sequence[int]]) -- How to split up the channels in the timeseries for different tensors. If left as None, all the channels will be returned in a single tensor. Otherwise, the channels will be split up into len(channels_per_snapshot) tensors, with each tensor's channel dimension being equal to the corresponding value in channels_per_snapshot. Therefore, if specified, these values should add up to num_channels.

forward(update, snapshot=None)

Define the computation performed at every call.

Should be overridden by all subclasses. :rtype: Tuple[Tensor, ...]

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

get_initial_state()

Module contents