aihwkit.simulator.rpu_base.tiles

Bindings for the simulator analog tiles.

class aihwkit.simulator.rpu_base.tiles.AnalogTile

Bases: FloatingPointTile

Analog tile.

Parameters:

x_size – X size of the tile.
d_size – D size of the tile.

get_meta_parameters(self: aihwkit.simulator.rpu_base.tiles.AnalogTile) → aihwkit.simulator.rpu_base.devices.AnalogTileParameter

class aihwkit.simulator.rpu_base.tiles.FloatingPointTile

Bases: pybind11_object

Floating point tile.

Parameters:

x_size – X size of the tile.
d_size – D size of the tile.

backward(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, d_input: torch.Tensor, bias: bool = False, d_trans: bool = False, x_trans: bool = False, non_blocking: bool = False) → torch.Tensor

Compute the transposed dot product (backward pass).

Compute the transposed dot product: .. math:

mathbf{y} = Wmathbf{d}

where \(\mathbf{d}\) is the input and \(W\) is the current weight matrix (of size [d_size, x_size]).

An analog tile will have a possible non-ideal version of this backward pass.

Parameters:

d_input – [N, *, d_size] input \(\mathbf{d}\) torch::Tensor.
bias – whether to use bias.
d_trans – whether the d_input matrix is transposed. That is of size [d_size, *, N]
x_trans – whether the x output matrix is transposed.

Returns:

[N, *, x_size (-1)] or [x_size (-1), *, N] torch::Tensor.

Return type:

torch::Tensor

backward_indexed(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, d_input: torch.Tensor, x_tensor: torch.Tensor, non_blocking: bool = False) → torch.Tensor

Compute the dot product using an index matrix (backward pass).

Caution

Internal use for convolutions only.

Parameters:

d_input – 4D torch::tensor in order N,C,H,W
x_tensor – torch:tensor with convolution dimensions

Returns:

4D (5D) torch::tensor in order N,C, (x_depth,) x_height, x_width

Return type:

x_output

clip_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, weight_clipper_params: aihwkit.simulator.rpu_base.tiles.WeightClipParameter) → None

Clips the weights for use of hardware-aware training.

Several clipping types are available, see WeightClipParameter.

Parameters:: weight_clipper_params – parameters of the clipping.

decay_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, alpha: float = 1.0) → None

Decays the weights:

W *= (1 - alpha / life_time)

An analog tile will have possible non-ideal version of this decay.

Parameters:: alpha – decay scale

diffuse_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → None

Diffuse the weights.

Diffuse the weights:

W += diffusion_rate * Gaussian noise

An analog tile will have a possible non-ideal version of this diffusion.

drift_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, time_since_last_call: float) → None

Drift weights according to a power law:

W = W0*(delta_t/t0)^(-nu_actual)

Applies the weight drift to all unchanged weight elements (judged by reset_tol) and resets the drift for those that have changed (nu is not re-drawn, however). Each device might have a different version of this drift.

Parameters:: time_since_last_call – This is the time between the calls (delta_t), typically the time to process a mini-batch for the network.

dump_extra(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → Dict[str, List[float]]

Return additional state vraiables for pickling.

Returns:: dictionary of extra variables states
Return type:: state

forward(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, x_input: torch.Tensor, bias: bool = False, x_trans: bool = False, d_trans: bool = False, is_test: bool = False, non_blocking: bool = False) → torch.Tensor

Compute the dot product (forward pass).

Compute the dot product: .. math:

mathbf{y} = Wmathbf{x} [+ mathbf{b}]

where \(\mathbf{x}\) is the input and \(W\) is the [d_size, x_size] current weight matrix. If bias is True, then it is assumes that a bias row is added to the analog tile weights. The input \(\mathbf{x}\) is then expected to be of size x_size -1 , as internally it will be expanded by a 1, to match the bias row in the tile weights.

An analog tile will have a possible non-ideal version of this forward pass.

Parameters:

x_input – [N,*, x_size (- 1)] input \(\mathbf{x}\) torch::Tensor.
bias – whether to use bias.
x_trans – whether the x_input matrix is transposed. That is of size [x_size (- 1), *, N]
d_trans – whether the d matrix is transposed.
is_test – whether inference (true) mode or training (false)

Returns:

[N, *, d_size] or [d_size, *, N] matrix.

Return type:

torch::tensor

forward_indexed(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, x_input: torch.Tensor, d_tensor: torch.Tensor, is_test: bool = False, non_blocking: bool = False) → torch.Tensor

Compute the dot product using an index matrix (forward pass).

Caution

Internal use for convolutions only.

Parameters:

x_input – 4D or 5D torch::tensor in order N,C,(D),H,W
d_tensor – torch:tensor with convolution dimensions
is_test – whether inference (true) mode or training (false)

Returns:

4D 5D torch::tensor in order N, C, (d_depth,) d_height, d_width

Return type:

d_output

get_brief_info(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → str

get_d_size(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → int

Return the tile output dimensions (d-size).

Returns:: the tile number of rows
Return type:: int

get_hidden_parameter_names(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → List[str]

Get the hidden parameters of the tile.

Returns:: list of hidden parameter names.
Return type:: list

get_hidden_parameters(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → torch.Tensor

Get the hidden parameters of the tile.

Returns:

Each 2D slice tensor is of size [d_size, x_size] (in row-major order): corresponding to the parameter name.

Return type:

3D tensor

get_hidden_update_index(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → int

Get the current device index that is updated (in case multiple devices per cross-point).

Parameters:: idx – index of the (unit cell) devices, returns 0 in all other cases.

get_info(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → str

get_learning_rate(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → float

Return the tile learning rate.

Returns:: the tile learning rate.
Return type:: float

get_meta_parameters(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → aihwkit.simulator.rpu_base.devices.FloatingPointTileParameter: Returns the current meta parameter structure.

get_pulse_counters(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → torch.Tensor

Get the pulse counters if available.

Returns:: Pulse counters: pos, neg (and for each sub-device)
Return type:: 3D tensor

get_shared_weights_if(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → bool: Returns whether weight is shared.

get_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → torch.Tensor

Return the exact tile weights.

Return the tile weights by producing and exact copy.

Note

This is not hardware realistic, and is used for debug purposes only.

Returns:: the [d_size, x_size] weight matrix.
Return type:: tensor

get_x_size(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → int

Return the tile input dimensions (x-size).

Returns:: the tile number of columns (including bias if available)
Return type:: int

has_matrix_indices(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → bool

Returns whether the index matrix necessary for the *_indexed functionality has been set.

Caution

Internal use only.

Returns:: whether it was set or not.
Return type:: bool

load_extra(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, state: Dict[str, List[float]], strict: bool) → None

Load the state dictionary generated by dump_extra.

Parameters:: strict – Whether to throw a runtime error when a field is not found.

modify_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, weight_modifier_params: aihwkit.simulator.rpu_base.tiles.WeightModifierParameter) → None

Modifies the weights in forward and backward (but not update) pass for use of hardware-aware training.

Several modifier types are available, see WeightModifierParameter.

Parameters:: weight_modifier_params – parameters of the modifications.

remap_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, weight_remap_params: aihwkit.simulator.rpu_base.tiles.WeightRemapParameter, scales: torch.Tensor) → torch.Tensor

Remaps the weights for use of hardware-aware training.

Several remap types are available, see WeightRemapParameter.

Parameters:

weight_remap_params – parameters of the remapping.
scales – scales that will be used and updated during remapping

Returns:

[d_size] of scales

Return type:

torch::tensor

reset_columns(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, start_column_idx: int = 0, num_columns: int = 1, reset_prob: float = 1.0) → None

Resets the weights with device-to-device and cycle-to-cycle variability (depending on device type), typically:

W_ij = xi*reset_std + reset_bias_ij

Parameters:

start_col_idx – a start index of columns (0..x_size-1)
num_columns – how many consecutive columns to reset (with circular warping)
reset_prob – individual probability of reset.

reset_delta_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile) → None

set_delta_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, delta_weights: torch.Tensor) → None

set_hidden_parameters(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, arg0: torch.Tensor) → None

Sets the hidden parameters of the tile.

Parameters:: tensor (3D) – Each 2D slice tensor is of size [d_size, x_size] (in row-major order) corresponding to the parameter name.

set_hidden_update_index(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, arg0: int) → None

Set the updated device index (in case multiple devices per cross-point).

Note

Only used for vector unit cells, so far. Ignored in other cases.

Parameters:: idx – index of the (unit cell) devices

set_learning_rate(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, learning_rate: float) → None

Set the tile learning rate.

Set the tile learning rate to -learning_rate. Please note that the learning rate is always taken to be negative (because of the meaning in gradient descent) and positive learning rates are not supported.

Parameters:: learning_rate – the desired learning rate.

set_matrix_indices(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, indices: torch.Tensor) → None

Sets the index vector for the *_indexed functionality.

Caution

Internal use only.

Parameters:: indices – int torch::Tensor

set_shared_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, weights: torch.Tensor) → None

set_verbosity_level(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, verbose: int) → None

Sets the verbosity level for debugging.

Parameters:: verbose – verbosity level

set_weights(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, weights: torch.Tensor) → None

Set the tile weights exactly.

Set the tile weights to the exact values of the weights parameter.

Note

This is not hardware realistic, and is used for debug purposes only.

Parameters:: weights – [d_size, x_size] weight matrix.

set_weights_uniform_random(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, min_value: float, max_value: float) → None

Sets weights uniformlay in the range min_value to max_value.

Parameters:

min_value – lower bound of uniform distribution
max_value – upper bound

update(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, x_input: torch.Tensor, d_input: torch.Tensor, bias: bool, d_trans: bool = False, x_trans: bool = False, non_blocking: bool = False) → None

Compute an n-rank update.

Compute an n-rank update: .. math:

W leftarrow W - lambda mathbf{x}mathbf{d}^T

where \(\lambda\) is the learning rate.

An analog tile will have a possible non-ideal version of this update pass.

Note

The learning rate is always positive, and thus scaling is negative.

Parameters:

x_input – [N, *, x_size (-1)] input \(\mathbf{x}\) torch::Tensor.
d_input – [N, *, d_size] input \(\mathbf{d}\) torch::Tensor.
bias – whether to use bias.
x_trans – whether the x_input matrix is transposed, ie. [x_size (-1), *, N]
d_trans – whether the d matrix is transposed, ie. [d_size, *, N]

update_indexed(self: aihwkit.simulator.rpu_base.tiles.FloatingPointTile, x_input: torch.Tensor, d_input: torch.Tensor, non_blocking: bool = False) → None

Compute the dot product using an index matrix (backward pass).

Caution

Internal use for convolutions only.

Parameters:

x_input – 4D torch::tensor input in order N,C,H,W
d_input – 4D torch::tensor (grad_output) in order N,C,oH,oW

class aihwkit.simulator.rpu_base.tiles.WeightClipParameter

Bases: pybind11_object

property fixed_value

property sigma

property type

class aihwkit.simulator.rpu_base.tiles.WeightClipType

Bases: pybind11_object

Members:

None

FixedValue

LayerGaussian

AverageChannelMax

AverageChannelMax = <WeightClipType.AverageChannelMax: 3>

FixedValue = <WeightClipType.FixedValue: 1>

LayerGaussian = <WeightClipType.LayerGaussian: 2>

None = <WeightClipType.None: 0>

property name

property value

class aihwkit.simulator.rpu_base.tiles.WeightModifierParameter

Bases: pybind11_object

property assumed_wmax

property coeffs

property copy_last_column

property dorefa_clip

property enable_during_test

property g_max

property pcm_prob_at_gmax

property pcm_prob_at_random

property pcm_prob_at_reset

property pcm_t0

property pcm_t_inference

property pcm_zero_thres

property pdrop

property per_batch_sample

property rel_to_actual_wmax

property res

property std_dev

property sto_round

property type

class aihwkit.simulator.rpu_base.tiles.WeightModifierType

Bases: pybind11_object

Members:

Copy

Discretize

MultNormal

AddNormal

DiscretizeAddNormal

DoReFa

Poly

PCMNoise

ProgNoise

DropConnect

None

AddNormal = <WeightModifierType.AddNormal: 3>

Copy = <WeightModifierType.Copy: 0>

Discretize = <WeightModifierType.Discretize: 1>

DiscretizeAddNormal = <WeightModifierType.DiscretizeAddNormal: 4>

DoReFa = <WeightModifierType.DoReFa: 5>

DropConnect = <WeightModifierType.DropConnect: 8>

MultNormal = <WeightModifierType.MultNormal: 2>

None = <WeightModifierType.Copy: 0>

PCMNoise = <WeightModifierType.PCMNoise: 7>

Poly = <WeightModifierType.Poly: 6>

ProgNoise = <WeightModifierType.ProgNoise: 9>

property name

property value

class aihwkit.simulator.rpu_base.tiles.WeightRemapParameter

Bases: pybind11_object

property max_scale_range

property max_scale_ref

property remapped_wmax

property type

class aihwkit.simulator.rpu_base.tiles.WeightRemapType

Bases: pybind11_object

Members:

None

LayerwiseSymmetric

ChannelwiseSymmetric

ChannelwiseSymmetric = <WeightRemapType.ChannelwiseSymmetric: 2>

LayerwiseSymmetric = <WeightRemapType.LayerwiseSymmetric: 1>

None = <WeightRemapType.None: 0>

property name

property value