Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
Lucid 2.8.5 documentation
Lucid 2.8.5 documentation

Tensor

  • lucid.Tensor
  • lucid.tensor
  • Tensor Operations
    • lucid.exp
    • lucid.log
    • lucid.log2
    • lucid.sqrt
    • lucid.sin
    • lucid.cos
    • lucid.tan
    • lucid.arcsin
    • lucid.arccos
    • lucid.arctan
    • lucid.sinh
    • lucid.cosh
    • lucid.tanh
    • lucid.clip
    • lucid.abs
    • lucid.sign
    • lucid.reciprocal
    • lucid.square
    • lucid.cube
    • lucid.transpose
    • lucid.sum
    • lucid.trace
    • lucid.mean
    • lucid.var
    • lucid.min
    • lucid.max
    • lucid.swapaxes
    • lucid.round
    • lucid.floor
    • lucid.ceil
    • lucid.cumprod
    • lucid.cumsum
    • lucid.add
    • lucid.sub
    • lucid.multiply
    • lucid.div
    • lucid.minimum
    • lucid.maximum
    • lucid.power
    • lucid.dot
    • lucid.inner
    • lucid.outer
    • lucid.matmul
    • lucid.tensordot
    • lucid.zeros
    • lucid.zeros_like
    • lucid.ones
    • lucid.ones_like
    • lucid.eye
    • lucid.diag
    • lucid.arange
    • lucid.empty
    • lucid.empty_like
    • lucid.linspace
    • lucid.full
    • lucid.full_like
  • Tensor Utilities
    • lucid.reshape
    • lucid.squeeze
    • lucid.unsqueeze
    • lucid.ravel
    • lucid.stack
    • lucid.hstack
    • lucid.vstack
    • lucid.concatenate
    • lucid.pad
    • lucid.repeat
    • lucid.tile
    • lucid.flatten
    • lucid.meshgrid
    • lucid.split
    • lucid.tril
    • lucid.triu
    • lucid.broadcast_to
    • lucid.chunk
    • lucid.masked_fill
    • lucid.roll
    • lucid.unbind
    • lucid.sort
    • lucid.argsort
    • lucid.argmin
    • lucid.argmax
    • lucid.nonzero
    • lucid.unique
    • lucid.topk
    • lucid.histogram
    • lucid.histogram2d
    • lucid.histogramdd
    • lucid.where
    • lucid.diagonal

Linear Algebra

  • lucid.linalg
  • Linalg Operations
    • lucid.linalg.det
    • lucid.linalg.inv
    • lucid.linalg.solve
    • lucid.linalg.norm
    • lucid.linalg.cholesky
    • lucid.linalg.eig
    • lucid.linalg.qr
    • lucid.linalg.svd
    • lucid.linalg.matrix_power
    • lucid.linalg.pinv

Random

  • lucid.random
  • RNG Functions
    • lucid.random.seed
    • lucid.random.rand
    • lucid.random.randint
    • lucid.random.randn
    • lucid.random.uniform
    • lucid.random.bernoulli
    • lucid.random.permutation

Einstein Operations

  • lucid.einops
  • Einops Functions
    • lucid.einops.rearrange
    • lucid.einops.reduce
    • lucid.einops.repeat
    • lucid.einops.einsum

Neural Networks

  • lucid.nn
  • nn.Parameter
  • nn.Buffer
  • nn.Module
  • Neural Functions
    • Linear Functions
      • nn.functional.linear
      • nn.functional.bilinear
    • Activation Functions
      • nn.functional.relu
      • nn.functional.leaky_relu
      • nn.functional.elu
      • nn.functional.selu
      • nn.functional.gelu
      • nn.functional.sigmoid
      • nn.functional.tanh
      • nn.functional.softmax
    • Attention Functions
      • nn.functional.scaled_dot_product_attention
    • Convolution Functions
      • nn.functional.unfold
      • nn.functional.conv1d
      • nn.functional.conv2d
      • nn.functional.conv3d
      • nn.functional.conv_transpose1d
      • nn.functional.conv_transpose2d
      • nn.functional.conv_transpose3d
    • Pooling Functions
      • nn.functional.avg_pool1d
      • nn.functional.avg_pool2d
      • nn.functional.avg_pool3d
      • nn.functional.max_pool1d
      • nn.functional.max_pool2d
      • nn.functional.max_pool3d
      • nn.functional.adaptive_avg_pool1d
      • nn.functional.adaptive_avg_pool2d
      • nn.functional.adaptive_avg_pool3d
      • nn.functional.adaptive_max_pool1d
      • nn.functional.adaptive_max_pool2d
      • nn.functional.adaptive_max_pool3d
    • Dropout Functions
      • nn.functional.dropout
      • nn.functional.dropout1d
      • nn.functional.dropout2d
      • nn.functional.dropout3d
      • nn.functional.alpha_dropout
      • nn.functional.drop_block
      • nn.functional.drop_path
    • Normalization Functions
      • nn.functional.normalize
      • nn.functional.batch_norm
      • nn.functional.layer_norm
      • nn.functional.instance_norm
      • nn.functional.group_norm
      • nn.functional.global_response_norm
    • Loss Functions
      • nn.functional.mse_loss
      • nn.functional.binary_cross_entropy
      • nn.functional.binary_cross_entropy_with_logits
      • nn.functional.cross_entropy
      • nn.functional.nll_loss
      • nn.functional.huber_loss
    • Spatial Functions
      • nn.functional.affine_grid
      • nn.functional.grid_sample
    • Utility Functions
      • nn.functional.interpolate
      • nn.functional.rotate
      • nn.functional.embedding
      • nn.functional.one_hot
  • Weight Initializations
    • nn.init.uniform
    • nn.init.normal
    • nn.init.constant
    • nn.init.xavier_uniform
    • nn.init.xavier_normal
    • nn.init.kaiming_uniform
    • nn.init.kaiming_normal
  • Modules
    • Linear Layers
      • nn.Identity
      • nn.Flatten
      • nn.Linear
      • nn.Bilinear
    • Convolution Layers
      • nn.Unfold
      • nn.Conv1d
      • nn.Conv2d
      • nn.Conv3d
      • nn.ConvTranspose1d
      • nn.ConvTranspose2d
      • nn.ConvTranspose3d
    • Activation Layers
      • nn.ReLU
      • nn.ReLU6
      • nn.LeakyReLU
      • nn.ELU
      • nn.SELU
      • nn.GELU
      • nn.Sigmoid
      • nn.HardSigmoid
      • nn.Tanh
      • nn.Softmax
      • nn.Swish
      • nn.HardSwish
      • nn.Mish
    • Pooling Layers
      • nn.AvgPool1d
      • nn.AvgPool2d
      • nn.AvgPool3d
      • nn.MaxPool1d
      • nn.MaxPool2d
      • nn.MaxPool3d
      • nn.AdaptiveAvgPool1d
      • nn.AdaptiveAvgPool2d
      • nn.AdaptiveAvgPool3d
      • nn.AdaptiveMaxPool1d
      • nn.AdaptiveMaxPool2d
      • nn.AdaptiveMaxPool3d
    • Normalization Layers
      • nn.BatchNorm1d
      • nn.BatchNorm2d
      • nn.BatchNorm3d
      • nn.InstanceNorm1d
      • nn.InstanceNorm2d
      • nn.InstanceNorm3d
      • nn.LayerNorm
      • nn.GroupNorm
      • nn.GlobalResponseNorm
    • Dropout Layers
      • nn.Dropout
      • nn.Dropout1d
      • nn.Dropout2d
      • nn.Dropout3d
      • nn.AlphaDropout
      • nn.DropBlock
      • nn.DropPath
    • Loss Layers
      • nn.MSELoss
      • nn.BCELoss
      • nn.BCEWithLogitsLoss
      • nn.CrossEntropyLoss
      • nn.NLLLoss
      • nn.HuberLoss
    • Vision Layers
      • nn.Upsample
    • Sparse Layers
      • nn.Embedding
    • Recurrent Layers
      • nn.RNNBase
      • nn.RNN
      • nn.LSTM
      • nn.GRU
      • nn.RNNCell
      • nn.LSTMCell
      • nn.GRUCell
    • Attention Layers
      • nn.ScaledDotProductAttention
      • nn.MultiHeadAttention
    • Transformer Layers
      • nn.TransformerEncoderLayer
      • nn.TransformerDecoderLayer
      • nn.TransformerEncoder
      • nn.TransformerDecoder
      • nn.Transformer
    • Einops Layers
      • nn.Rearrange
  • Fused Modules
    • nn.ConvBNReLU1d
    • nn.ConvBNReLU2d
    • nn.ConvBNReLU3d
    • nn.DepthSeparableConv1d
    • nn.DepthSeparableConv2d
    • nn.DepthSeparableConv3d
    • nn.SEModule
    • nn.SelectiveKernel
  • Containers
    • nn.Sequential
    • nn.ModuleList
    • nn.ModuleDict
    • nn.ParameterList
    • nn.ParameterDict
  • Utilities
    • nn.util.grad_norm
    • nn.util.clip_grad_norm
    • nn.util.clip_grad_value

Optimization

  • lucid.optim
  • optim.Optimizer
  • optim.lr_scheduler
  • Optimizers
    • optim.SGD
    • optim.ASGD
    • optim.RMSprop
    • optim.Rprop
    • optim.Adam
    • optim.AdamW
    • optim.NAdam
    • optim.RAdam
    • optim.Adamax
    • optim.Adagrad
    • optim.Adadelta
  • LR Schedulers
    • lr_scheduler.LRScheduler
    • lr_scheduler.LambdaLR
    • lr_scheduler.StepLR
    • lr_scheduler.MultiStepLR
    • lr_scheduler.ExponentialLR
    • lr_scheduler.CosineAnnealingLR
    • lr_scheduler.ReduceLROnPlateau
    • lr_scheduler.CyclicLR
    • lr_scheduler.NoamScheduler

Data

  • lucid.data
  • data.Dataset
  • data.Subset
  • data.TensorDataset
  • data.ConcatDataset
  • data.DataLoader
  • Utilities
    • data.random_split

Datasets

  • lucid.datasets
  • Image Datasets
    • MNIST
    • FashionMNIST
    • CIFAR10
    • CIFAR100

Models

  • lucid.models
  • Image Classification
    • LeNet
      • lenet_1
      • lenet_4
      • lenet_5
    • AlexNet
      • alexnet
    • ZFNet
      • zfnet
    • VGGNet
      • vggnet_11
      • vggnet_13
      • vggnet_16
      • vggnet_19
    • Inception
      • inception_v1
      • inception_v3
      • inception_v4
    • Inception-ResNet
      • inception_resnet_v1
      • inception_resnet_v2
    • ResNet
      • resnet_18
      • resnet_34
      • resnet_50
      • resnet_101
      • resnet_152
      • resnet_200
      • resnet_269
      • resnet_1001
      • wide_resnet_50
      • wide_resnet_101
    • ResNeXt
      • resnext_50_32x4d
      • resnext_101_32x4d
      • resnext_101_32x8d
      • resnext_101_32x16d
      • resnext_101_32x32d
      • resnext_101_64x4d
    • ResNeSt
      • resnest_14
      • resnest_26
      • resnest_50
      • resnest_101
      • resnest_200
      • resnest_269
      • resnest_50_4s2x40d
      • resnest_50_1s4x24d
    • SENet
      • se_resnet_18
      • se_resnet_34
      • se_resnet_50
      • se_resnet_101
      • se_resnet_152
      • se_resnext_50_32x4d
      • se_resnext_101_32x4d
      • se_resnext_101_32x8d
      • se_resnext_101_64x4d
    • SKNet
      • sk_resnet_18
      • sk_resnet_34
      • sk_resnet_50
      • sk_resnext_50_32x4d
    • DenseNet
      • densenet_121
      • densenet_169
      • densenet_201
      • densenet_264
    • Xception
      • xception
    • MobileNet
      • mobilenet
    • MobileNet-v2
      • mobilenet_v2
    • MobileNet-v3
      • mobilenet_v3_small
      • mobilenet_v3_large
    • MobileNet-v4
      • mobilenet_v4_conv_small
      • mobilenet_v4_conv_medium
      • mobilenet_v4_conv_large
      • mobilenet_v4_hybrid_medium
      • mobilenet_v4_hybrid_large
    • EfficientNet
      • efficientnet_b0
      • efficientnet_b1
      • efficientnet_b2
      • efficientnet_b3
      • efficientnet_b4
      • efficientnet_b5
      • efficientnet_b6
      • efficientnet_b7
    • EfficientNet-v2
      • efficientnet_v2_s
      • efficientnet_v2_m
      • efficientnet_v2_l
      • efficientnet_v2_xl
    • ConvNeXt
      • convnext_tiny
      • convnext_small
      • convnext_base
      • convnext_large
      • convnext_xlarge
    • ConvNeXt-v2
      • convnext_v2_atto
      • convnext_v2_femto
      • convnext_v2_pico
      • convnext_v2_nano
      • convnext_v2_tiny
      • convnext_v2_base
      • convnext_v2_large
      • convnext_v2_huge
    • InceptionNeXt
      • inception_next_atto
      • inception_next_tiny
      • inception_next_small
      • inception_next_base
    • CoAtNet
      • coatnet_0
      • coatnet_1
      • coatnet_2
      • coatnet_3
      • coatnet_4
      • coatnet_5
      • coatnet_6
      • coatnet_7
    • CSPNet
      • csp_resnet_50
      • csp_resnext_50_32x4d
      • csp_darknet_53
    • ViT
      • vit_tiny
      • vit_small
      • vit_base
      • vit_large
      • vit_huge
    • Swin Transformer
      • swin_tiny
      • swin_small
      • swin_base
      • swin_large
    • Swin Transformer-v2
      • swin_v2_tiny
      • swin_v2_small
      • swin_v2_base
      • swin_v2_large
      • swin_v2_huge
      • swin_v2_giant
    • CvT
      • cvt_13
      • cvt_21
      • cvt_w24
    • PVT
      • pvt_tiny
      • pvt_small
      • pvt_medium
      • pvt_large
      • pvt_huge
    • PVT-v2
      • pvt_v2_b0
      • pvt_v2_b1
      • pvt_v2_b2
      • pvt_v2_b2_li
      • pvt_v2_b3
      • pvt_v2_b4
      • pvt_v2_b5
    • CrossViT
      • crossvit_tiny
      • crossvit_small
      • crossvit_base
      • crossvit_9
      • crossvit_15
      • crossvit_18
      • crossvit_9_dagger
      • crossvit_15_dagger
      • crossvit_18_dagger
    • MaxViT
      • maxvit_tiny
      • maxvit_small
      • maxvit_base
      • maxvit_base
      • maxvit_xlarge
    • EfficientFormer
      • efficientformer_l1
      • efficientformer_l3
      • efficientformer_l7
  • Image Generation
    • VAE
    • DDPM
  • Object Detection
    • Utilities
      • util.iou
      • util.nms
      • util.bbox_to_delta
      • util.apply_deltas
      • util.clip_boxes
      • util.SelectiveSearch
      • util.ROIAlign
      • util.MultiScaleROIAlign
      • util.FPN
    • R-CNN
    • Fast R-CNN
    • Faster R-CNN
      • faster_rcnn_resnet_50_fpn
      • faster_rcnn_resnet_101_fpn
    • YOLO
      • YOLO-v1
        • yolo_v1
        • yolo_v1_tiny
      • YOLO-v2
        • yolo_v2
        • yolo_v2_tiny
      • YOLO-v3
        • yolo_v3
        • yolo_v3_tiny
      • YOLO-v4
        • yolo_v4
    • EfficientDet
      • efficientdet_d0
      • efficientdet_d1
      • efficientdet_d2
      • efficientdet_d3
      • efficientdet_d4
      • efficientdet_d5
      • efficientdet_d6
      • efficientdet_d7
    • DETR
      • detr_r50
      • detr_r101
  • Sequence-to-Sequence
    • Transformer
      • transformer_base
      • transformer_big

Weights

  • lucid.weights
  • Pre-Trained Weights

Transformation

  • lucid.transforms
  • transforms.Compose
  • transforms.ToTensor
  • Image Transforms
    • transforms.Normalize
    • transforms.Resize
    • transforms.RandomHorizontalFlip
    • transforms.RandomVerticalFlip
    • transforms.RandomCrop
    • transforms.CenterCrop
    • transforms.RandomRotation
    • transforms.RandomGrayscale

Visualization

  • lucid.visual
  • visual.draw_tensor_graph

Porting

  • lucid.save
  • lucid.load

Others

  • lucid.Numeric
  • lucid.no_grad
  • lucid.grad_enabled
  • lucid.count_flops
  • lucid.newaxis
  • lucid.register_model
Back to top
View this page
Edit this page

Sparse LayersΒΆ

  • nn.Embedding
Next
nn.Embedding
Previous
nn.Upsample
Copyright © 2025, ChanLumerico
Made with Sphinx and @pradyunsg's Furo