masskit_ai.spectrum.small_mol.models package

Submodules

masskit_ai.spectrum.small_mol.models.small_mol_models module

class masskit_ai.spectrum.small_mol.models.small_mol_models.AIMSNet(*args: Any, **kwargs: Any)

Bases: SpectrumModel

generates fingerprint for AIMS hybrid search

forward(x)
class masskit_ai.spectrum.small_mol.models.small_mol_models.BasicBlock(*args: Any, **kwargs: Any)

Bases: Module

ResNet Basic Block

forward(x)
class masskit_ai.spectrum.small_mol.models.small_mol_models.Conv1dSamePadding(*args: Any, **kwargs: Any)

Bases: Conv1d

Represents the “Same” padding functionality from Tensorflow. See: https://github.com/pytorch/pytorch/issues/3867 Note that the padding argument in the initializer doesn’t do anything now

forward(input)
class masskit_ai.spectrum.small_mol.models.small_mol_models.ConvBlock(*args: Any, **kwargs: Any)

Bases: Module

forward(x: torch.Tensor) → torch.Tensor
class masskit_ai.spectrum.small_mol.models.small_mol_models.LocalLinear(*args: Any, **kwargs: Any)

Bases: Module

forward(x: torch.Tensor)
class masskit_ai.spectrum.small_mol.models.small_mol_models.MyConv1dPadSame(*args: Any, **kwargs: Any)

Bases: Module

extend nn.Conv1d to support SAME padding

forward(x)
class masskit_ai.spectrum.small_mol.models.small_mol_models.MyMaxPool1dPadSame(*args: Any, **kwargs: Any)

Bases: Module

extend nn.MaxPool1d to support SAME padding

forward(x)
class masskit_ai.spectrum.small_mol.models.small_mol_models.ResNetBaseline(*args: Any, **kwargs: Any)

Bases: SpectrumModel

A PyTorch implementation of the ResNet Baseline From https://arxiv.org/abs/1909.04939 Attributes ———- sequence_length: The size of the input sequence self.config.ml.model.ResNetBaseline.mid_channels:The 3 residual blocks will have as output channels: [self.config.ml.model.ResNetBaseline.mid_channels, self.config.ml.model.ResNetBaseline.mid_channels * 2, self.config.ml.model.ResNetBaseline.mid_channels * 2] self.config.ml.model.ResNetBaseline.fp_size:The number of output classes

forward(x: torch.Tensor) → torch.Tensor
class masskit_ai.spectrum.small_mol.models.small_mol_models.ResNetBaseline_new(*args: Any, **kwargs: Any)

Bases: SpectrumModel

Input:

X: (n_samples, n_channel, n_length) Y: (n_samples)

Output:

out: (n_samples)

Pararmetes:

in_channels: dim of input, the same as n_channel base_filters: number of filters in the first several Conv layer, it will double at every 4 layers kernel_size: width of kernel stride: stride of kernel moving groups: set larget to 1 as ResNeXt n_block: number of blocks n_classes: number of classes

forward(x)
class masskit_ai.spectrum.small_mol.models.small_mol_models.ResNetBlock(*args: Any, **kwargs: Any)

Bases: Module

forward(x: torch.Tensor) → torch.Tensor
class masskit_ai.spectrum.small_mol.models.small_mol_models.SimpleNet(*args: Any, **kwargs: Any)

Bases: SpectrumModel

simple n layer dense network

forward(x)
masskit_ai.spectrum.small_mol.models.small_mol_models.conv1d_same_padding(input, weight, bias, stride, dilation, groups)

Module contents