brevitas.core.stats package#

Submodules#

brevitas.core.stats.stats_op module#

class brevitas.core.stats.stats_op.AbsAve(stats_reduce_dim=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.AbsMax(stats_reduce_dim=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.AbsMaxAve(stats_reduce_dim)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.AbsMaxL2(stats_reduce_dim)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.AbsMinMax(stats_reduce_dim=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.AbsPercentile(high_percentile_q, stats_reduce_dim, percentile_q=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.KLMinimizerThreshold(signed, bit_width_impl, num_bins=1001, smoothing_eps=0.0001)[source]#

Bases: Module

Based on: apache/incubator-mxnet

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

smooth_normalize_distribution(p, eps)[source]#
training: bool#
class brevitas.core.stats.stats_op.L1Norm(stats_reduce_dim=None)[source]#

Bases: Module

ScriptModule implementation to collect per-channel L1 normalization stats for weight normalization-based quantization.

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.L2Norm(stats_reduce_dim=None)[source]#

Bases: Module

ScriptModule implementation to collect per-channel L2 normalization stats for weight normalization-based quantization.

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.MeanLearnedSigmaStd(sigma, stats_output_shape, stats_reduce_dim=None, std_dev_epsilon=1e-08)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.MeanSigmaStd(sigma, stats_reduce_dim=None, std_dev_epsilon=1e-08)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

training: bool#
class brevitas.core.stats.stats_op.NegativeMinOrZero(stats_reduce_dim=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

Return type:

Tensor

training: bool#
class brevitas.core.stats.stats_op.NegativePercentileOrZero(low_percentile_q, stats_reduce_dim=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

Return type:

Tensor

training: bool#
class brevitas.core.stats.stats_op.PercentileInterval(low_percentile_q, high_percentile_q, stats_reduce_dim=None)[source]#

Bases: Module

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

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.

Return type:

Tensor

training: bool#

brevitas.core.stats.stats_wrapper module#

brevitas.core.stats.view_wrapper module#

Module contents#