convert_batch_to_numpy_identity

convert_batch_to_numpy_identity(*args, **kwargs)

Corrects the shape of all zero-sized numpy arrays to be at least 1d

Parameters

• *args – positional arguments of potential arrays to be corrected

• **kwargs – keyword arguments of potential arrays to be corrected

float_to_pytorch_tensor

float_to_pytorch_tensor(f: float)

Converts a single float to a PyTorch Float-Tensor

Parameters

f (float) – float to convert

Returns

converted float

Return type

torch.Tensor

create_optims_default_pytorch

create_optims_default_pytorch(model, optim_cls, **optim_params)

Function to create a optimizer dictionary (in this case only one optimizer for the whole network)

Parameters

• model (AbstractPyTorchNetwork) – model whose parameters should be updated by the optimizer

• optim_cls – Class implementing an optimization algorithm

• **optim_params – Additional keyword arguments (passed to the optimizer class

Returns

dictionary containing all created optimizers

Return type

dict

create_optims_gan_pytorch

convert_torch_tensor_to_npy

convert_torch_tensor_to_npy(*args, **kwargs)

Function to convert all torch Tensors to numpy arrays and reshape zero-size tensors

Parameters

• *args – arbitrary positional arguments

• **kwargs – arbitrary keyword arguments

Returns

• Iterable – all given positional arguments (converted if necessary)

• dict – all given keyword arguments (converted if necessary)

create_optims_default_tf

create_optims_default_tf(optim_cls, **optim_params)

Function to create a optimizer dictionary (in this case only one optimizer)

Parameters

• optim_cls – Class implementing an optimization algorithm

• **optim_params – Additional keyword arguments (passed to the optimizer class)

Returns

dictionary containing all created optimizers

Return type

dict

initialize_uninitialized

initialize_uninitialized(sess)

Function to initialize only uninitialized variables in a session graph

Parameters

sess (tf.Session()) –

convert_tf_tensor_to_npy

convert_tf_tensor_to_npy(*args, **kwargs)

Function to convert all tf Tensors to numpy arrays and reshape zero-size tensors

Parameters

• *args – arbitrary positional arguments

• **kwargs – arbitrary keyword arguments

Returns

• Iterable – all given positional arguments (converted if necessary)

• dict – all given keyword arguments (converted if necessary)