agnapprox.nets

Subpackages

• agnapprox.nets.base
  • agnapprox.nets.base.lenet5
  • agnapprox.nets.base.resnet

Submodules

• agnapprox.nets.alexnet
• agnapprox.nets.approxnet
• agnapprox.nets.lenet5
• agnapprox.nets.mobilenetv2
• agnapprox.nets.resnet
• agnapprox.nets.vgg

Package Contents

Classes

ApproxNet	Base Class that provideds common functionality for approximate neural network training
AlexNet	Definition of training hyperparameters for
LeNet5	Definition of training hyperparameters for
MobileNetV2	Definition of training hyperparameters for
ResNet	Definition of training hyperparameters for
VGG	Definition of training hyperparameters for

class agnapprox.nets.ApproxNet(deterministic: bool = False)

Bases: pytorch_lightning.LightningModule

Base Class that provideds common functionality for approximate neural network training

gather_noisy_modules()

Replace regular Conv2d and Linear layer instances with derived approximate layer instances that provide additional functionality

property total_ops → torch.Tensor

Sum of the number of operations for all target layers in the model. This is calculated during inference for layers with dynamic input sizes like Convolutions.

Raises:

ValueError – Opcount has not yet been populated

Returns:

Tensor containing a single item with the total number of multiplications

property mode → str

The current mode of the network. This determines which optimizer and number of epochs are selected for optimization runs. Can be any of: - “baseline”: FP32 baseline model - “qat”: Quantization-aware training - “gradient_search”: Quantized model with added noise, noise injections

per layer is optimized together with other network parameters

• “approx”: Approximate Retraining with simulated approximate multipliers

forward(features) → torch.Tensor

Same as torch.nn.Module.forward().

Parameters:

• *args – Whatever you decide to pass into the forward method.

• **kwargs – Keyword arguments are also possible.

Returns:

Your model’s output

training_step(train_batch, _batch_idx) → torch.Tensor

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters:

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if truncated_bptt_steps > 0.

Returns:

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

validation_step(val_batch, _batch_idx) → torch.Tensor

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters:

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns:

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

test_step(test_batch, _batch_idx) → torch.Tensor

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters:

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns:

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

configure_optimizers()

Choose what optimizers and learning-rate schedulers to use in your optimization. Normally you’d need one. But in the case of GANs or similar you might have multiple.

Returns:

Any of these 6 options.

• Single optimizer.

• List or Tuple of optimizers.

• Two lists - The first list has multiple optimizers, and the second has multiple LR schedulers (or multiple lr_scheduler_config).

• Dictionary, with an "optimizer" key, and (optionally) a "lr_scheduler" key whose value is a single LR scheduler or lr_scheduler_config.

• Tuple of dictionaries as described above, with an optional "frequency" key.

• None - Fit will run without any optimizer.

The lr_scheduler_config is a dictionary which contains the scheduler and its associated configuration. The default configuration is shown below.

lr_scheduler_config = {
    # REQUIRED: The scheduler instance
    "scheduler": lr_scheduler,
    # The unit of the scheduler's step size, could also be 'step'.
    # 'epoch' updates the scheduler on epoch end whereas 'step'
    # updates it after a optimizer update.
    "interval": "epoch",
    # How many epochs/steps should pass between calls to
    # `scheduler.step()`. 1 corresponds to updating the learning
    # rate after every epoch/step.
    "frequency": 1,
    # Metric to to monitor for schedulers like `ReduceLROnPlateau`
    "monitor": "val_loss",
    # If set to `True`, will enforce that the value specified 'monitor'
    # is available when the scheduler is updated, thus stopping
    # training if not found. If set to `False`, it will only produce a warning
    "strict": True,
    # If using the `LearningRateMonitor` callback to monitor the
    # learning rate progress, this keyword can be used to specify
    # a custom logged name
    "name": None,
}

When there are schedulers in which the .step() method is conditioned on a value, such as the torch.optim.lr_scheduler.ReduceLROnPlateau scheduler, Lightning requires that the lr_scheduler_config contains the keyword "monitor" set to the metric name that the scheduler should be conditioned on.

Metrics can be made available to monitor by simply logging it using self.log('metric_to_track', metric_val) in your LightningModule.

Note

The frequency value specified in a dict along with the optimizer key is an int corresponding to the number of sequential batches optimized with the specific optimizer. It should be given to none or to all of the optimizers. There is a difference between passing multiple optimizers in a list, and passing multiple optimizers in dictionaries with a frequency of 1:

• In the former case, all optimizers will operate on the given batch in each optimization step.

• In the latter, only one optimizer will operate on the given batch at every step.

This is different from the frequency value specified in the lr_scheduler_config mentioned above.

def configure_optimizers(self):
    optimizer_one = torch.optim.SGD(self.model.parameters(), lr=0.01)
    optimizer_two = torch.optim.SGD(self.model.parameters(), lr=0.01)
    return [
        {"optimizer": optimizer_one, "frequency": 5},
        {"optimizer": optimizer_two, "frequency": 10},
    ]

In this example, the first optimizer will be used for the first 5 steps, the second optimizer for the next 10 steps and that cycle will continue. If an LR scheduler is specified for an optimizer using the lr_scheduler key in the above dict, the scheduler will only be updated when its optimizer is being used.

Examples:

# most cases. no learning rate scheduler
def configure_optimizers(self):
    return Adam(self.parameters(), lr=1e-3)

# multiple optimizer case (e.g.: GAN)
def configure_optimizers(self):
    gen_opt = Adam(self.model_gen.parameters(), lr=0.01)
    dis_opt = Adam(self.model_dis.parameters(), lr=0.02)
    return gen_opt, dis_opt

# example with learning rate schedulers
def configure_optimizers(self):
    gen_opt = Adam(self.model_gen.parameters(), lr=0.01)
    dis_opt = Adam(self.model_dis.parameters(), lr=0.02)
    dis_sch = CosineAnnealing(dis_opt, T_max=10)
    return [gen_opt, dis_opt], [dis_sch]

# example with step-based learning rate schedulers
# each optimizer has its own scheduler
def configure_optimizers(self):
    gen_opt = Adam(self.model_gen.parameters(), lr=0.01)
    dis_opt = Adam(self.model_dis.parameters(), lr=0.02)
    gen_sch = {
        'scheduler': ExponentialLR(gen_opt, 0.99),
        'interval': 'step'  # called after each training step
    }
    dis_sch = CosineAnnealing(dis_opt, T_max=10) # called every epoch
    return [gen_opt, dis_opt], [gen_sch, dis_sch]

# example with optimizer frequencies
# see training procedure in `Improved Training of Wasserstein GANs`, Algorithm 1
# https://arxiv.org/abs/1704.00028
def configure_optimizers(self):
    gen_opt = Adam(self.model_gen.parameters(), lr=0.01)
    dis_opt = Adam(self.model_dis.parameters(), lr=0.02)
    n_critic = 5
    return (
        {'optimizer': dis_opt, 'frequency': n_critic},
        {'optimizer': gen_opt, 'frequency': 1}
    )

Note

Some things to know:

• Lightning calls .backward() and .step() on each optimizer as needed.

• If learning rate scheduler is specified in configure_optimizers() with key "interval" (default “epoch”) in the scheduler configuration, Lightning will call the scheduler’s .step() method automatically in case of automatic optimization.

• If you use 16-bit precision (precision=16), Lightning will automatically handle the optimizers.

• If you use multiple optimizers, training_step() will have an additional optimizer_idx parameter.

• If you use torch.optim.LBFGS, Lightning handles the closure function automatically for you.

• If you use multiple optimizers, gradients will be calculated only for the parameters of current optimizer at each training step.

• If you need to control how often those optimizers step or override the default .step() schedule, override the optimizer_step() hook.

_train(datamodule: pytorch_lightning.LightningDataModule, run_name: str, epochs: Optional[int] = None, log_mlflow: bool = False, test: bool = False, **kwargs)

Internal Trainer function. This function is called by the different training stage functions.

Parameters:

• datamodule – The dataset to train on

• run_name – Run name passed to MLFlow

• epochs – Optional number of epochs to train for. If not set, number of epochs defined in the network definition will be used. Defaults to None.

• log_mlflow – Log training data to MLFlow. Defaults to False.

• test – Run on test set after training. Defaults to False.

train_baseline(datamodule: pytorch_lightning.LightningDataModule, **kwargs)

Train an FP32 baseline model

Parameters:

datamodule – Dataset provider

train_quant(datamodule: pytorch_lightning.LightningDataModule, **kwargs)

Train a quantized model using Quantization-Aware training

Parameters:

datamodule – Dataset provider

train_gradient(datamodule: pytorch_lightning.LightningDataModule, lmbd: float = 0.2, initial_noise: float = 0.1, **kwargs)

Run Gradient Search algorithm to optimize layer robustness parameters

Parameters:

• datamodule – Dataset provider

• lmdb – Lambda parameter that controls weighing of task loss and noise loss in the overall loss function. Defaults to 0.2

• initial_noise – The initial value to set for the noise parameter. Defaults to 0.1.

train_approx(datamodule: pytorch_lightning.LightningDataModule, name_ext: Optional[str] = None, **kwargs)

Train model with simulated approximate multipliers

Parameters:

• datamodule – Dataset provider

• name_ext – Optional extension to add to experiment tracking name. Helpful for distinguishing different multiplier configurations (i.e. signed/unsigned, uniform/non-uniform, etc.). Defaults to None.

on_fit_start()

Called at the very beginning of fit.

If on DDP it is called on every process

_baseline_optimizers()

Baseline Optimizer and Scheduler definition

_qat_optimizers()

Quantization-Aware Training Optimizer and Scheduler definition

_approx_optimizers()

Approximate Retraining Training Optimizer and Scheduler definition

_gs_optimizers()

Gradient Search Optimizer and Scheduler definition

class agnapprox.nets.AlexNet(num_classes: int = 200, pretrained: bool = True)

Bases: agnapprox.nets.approxnet.ApproxNet

Definition of training hyperparameters for approximate AlexNet

_baseline_optimizers()

Baseline Optimizer and Scheduler definition

_qat_optimizers()

Quantization-Aware Training Optimizer and Scheduler definition

_approx_optimizers()

Approximate Retraining Training Optimizer and Scheduler definition

_gs_optimizers()

Gradient Search Optimizer and Scheduler definition

class agnapprox.nets.LeNet5

Bases: agnapprox.nets.approxnet.ApproxNet

Definition of training hyperparameters for approximate LeNet5

_baseline_optimizers()

Baseline Optimizer and Scheduler definition

_qat_optimizers()

Quantization-Aware Training Optimizer and Scheduler definition

_approx_optimizers()

Approximate Retraining Training Optimizer and Scheduler definition

_gs_optimizers()

Gradient Search Optimizer and Scheduler definition

class agnapprox.nets.MobileNetV2(num_classes: int = 200, pretrained: bool = True)

Bases: agnapprox.nets.approxnet.ApproxNet

Definition of training hyperparameters for approximate MobileNetV2

_baseline_optimizers()

Baseline Optimizer and Scheduler definition

_qat_optimizers()

Quantization-Aware Training Optimizer and Scheduler definition

_approx_optimizers()

Approximate Retraining Training Optimizer and Scheduler definition

_gs_optimizers()

Gradient Search Optimizer and Scheduler definition

class agnapprox.nets.ResNet(resnet_size: Optional[str] = 'ResNet8', **kwargs)

Bases: agnapprox.nets.approxnet.ApproxNet

Definition of training hyperparameters for approximate ResNet

_baseline_optimizers()

Baseline Optimizer and Scheduler definition

_qat_optimizers()

Quantization-Aware Training Optimizer and Scheduler definition

_approx_optimizers()

Approximate Retraining Training Optimizer and Scheduler definition

_gs_optimizers()

Gradient Search Optimizer and Scheduler definition

class agnapprox.nets.VGG(vgg_size: Optional[str] = 'VGG11', num_classes: int = 200, pretrained: bool = True)

Bases: agnapprox.nets.approxnet.ApproxNet

Definition of training hyperparameters for approximate VGG

_baseline_optimizers()

Baseline Optimizer and Scheduler definition

_qat_optimizers()

Quantization-Aware Training Optimizer and Scheduler definition

_approx_optimizers()

Approximate Retraining Training Optimizer and Scheduler definition

_gs_optimizers()

Gradient Search Optimizer and Scheduler definition