eugene.models.zoo.BPNet

class eugene.models.zoo.BPNet(input_len, output_dim, n_filters=64, n_layers=8, n_outputs=2, n_control_tracks=2, alpha=1, profile_output_bias=True, count_output_bias=True, name=None, trimming=None, verbose=True)

This nn.Module was taken without permission from a Mr. Schreiber. Just kidding, he made it open source so, ipso facto, I do have permission. Anyway the documentation below is from him, so yell at him if it doesn’t work.

A basic BPNet model with stranded profile and total count prediction. This is a reference implementation for BPNet. The model takes in one-hot encoded sequence, runs it through: (1) a single wide convolution operation THEN (2) a user-defined number of dilated residual convolutions THEN (3a) profile predictions done using a very wide convolution layer that also takes in stranded control tracks AND (3b) total count prediction done using an average pooling on the output from 2 followed by concatenation with the log1p of the sum of the stranded control tracks and then run through a dense layer. This implementation differs from the original BPNet implementation in two ways: (1) The model concatenates stranded control tracks for profile prediction as opposed to adding the two strands together and also then smoothing that track (2) The control input for the count prediction task is the log1p of the strand-wise sum of the control tracks, as opposed to the raw counts themselves. (3) A single log softmax is applied across both strands such that the logsumexp of both strands together is 0. Put another way, the two strands are concatenated together, a log softmax is applied, and the MNLL loss is calculated on the concatenation. (4) The count prediction task is predicting the total counts across both strands. The counts are then distributed across strands according to the single log softmax from 3.

Parameters:

  • n_filters (int, optional) – The number of filters to use per convolution. Default is 64.

  • n_layers (int, optional) – The number of dilated residual layers to include in the model. Default is 8.

  • n_outputs (int, optional) – The number of profile outputs from the model. Generally either 1 or 2 depending on if the data is unstranded or stranded. Default is 2.

  • alpha (float, optional) – The weight to put on the count loss.

  • name (str or None, optional) – The name to save the model to during training.

  • trimming (int or None, optional) – The amount to trim from both sides of the input window to get the output window. This value is removed from both sides, so the total number of positions removed is 2*trimming.

  • verbose (bool, optional) – Whether to display statistics during training. Setting this to False will still save the file at the end, but does not print anything to screen during training. Default is True.

__init__(input_len, output_dim, n_filters=64, n_layers=8, n_outputs=2, n_control_tracks=2, alpha=1, profile_output_bias=True, count_output_bias=True, name=None, trimming=None, verbose=True)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

Methods

__init__(input_len, output_dim[, n_filters, ...])

Initializes internal Module state, shared by both nn.Module and ScriptModule.

add_module(name, module)

Adds a child module to the current module.

apply(fn)

Applies fn recursively to every submodule (as returned by .children()) as well as self.

bfloat16()

Casts all floating point parameters and buffers to bfloat16 datatype.

buffers([recurse])

Returns an iterator over module buffers.

children()

Returns an iterator over immediate children modules.

cpu()

Moves all model parameters and buffers to the CPU.

cuda([device])

Moves all model parameters and buffers to the GPU.

double()

Casts all floating point parameters and buffers to double datatype.

eval()

Sets the module in evaluation mode.

extra_repr()

Set the extra representation of the module

float()

Casts all floating point parameters and buffers to float datatype.

forward(X[, X_ctl])

A forward pass of the model.

get_buffer(target)

Returns the buffer given by target if it exists, otherwise throws an error.

get_extra_state()

Returns any extra state to include in the module's state_dict.

get_parameter(target)

Returns the parameter given by target if it exists, otherwise throws an error.

get_submodule(target)

Returns the submodule given by target if it exists, otherwise throws an error.

half()

Casts all floating point parameters and buffers to half datatype.

ipu([device])

Moves all model parameters and buffers to the IPU.

load_state_dict(state_dict[, strict])

Copies parameters and buffers from state_dict into this module and its descendants.

modules()

Returns an iterator over all modules in the network.

named_buffers([prefix, recurse])

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

named_children()

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

named_modules([memo, prefix, remove_duplicate])

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

named_parameters([prefix, recurse])

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

parameters([recurse])

Returns an iterator over module parameters.

register_backward_hook(hook)

Registers a backward hook on the module.

register_buffer(name, tensor[, persistent])

Adds a buffer to the module.

register_forward_hook(hook)

Registers a forward hook on the module.

register_forward_pre_hook(hook)

Registers a forward pre-hook on the module.

register_full_backward_hook(hook)

Registers a backward hook on the module.

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module's load_state_dict is called.

register_module(name, module)

Alias for add_module().

register_parameter(name, param)

Adds a parameter to the module.

requires_grad_([requires_grad])

Change if autograd should record operations on parameters in this module.

set_extra_state(state)

This function is called from load_state_dict() to handle any extra state found within the state_dict.

share_memory()

See torch.Tensor.share_memory_()

state_dict(*args[, destination, prefix, ...])

Returns a dictionary containing references to the whole state of the module.

to(*args, **kwargs)

Moves and/or casts the parameters and buffers.

to_empty(*, device)

Moves the parameters and buffers to the specified device without copying storage.

train([mode])

Sets the module in training mode.

type(dst_type)

Casts all parameters and buffers to dst_type.

xpu([device])

Moves all model parameters and buffers to the XPU.

zero_grad([set_to_none])

Sets gradients of all model parameters to zero.

Attributes

T_destination

dump_patches

training