torch-rb 0.1.2 → 0.1.7

data/lib/torch/nn/conv1d.rb

@@ -0,0 +1,22 @@
+module Torch
+  module NN
+    class Conv1d < ConvNd
+      def initialize(in_channels, out_channels, kernel_size, stride: 1,
+        padding: 0, dilation: 1, groups: 1, bias: true, padding_mode: "zeros")
+
+        kernel_size = _single(kernel_size)
+        stride = _single(stride)
+        padding = _single(padding)
+        dilation = _single(dilation)
+        super(in_channels, out_channels, kernel_size, stride, padding, dilation, false, _single(0), groups, bias, padding_mode)
+      end
+
+      def forward(input)
+        if @padding_mode == "circular"
+          raise NotImplementedError
+        end
+        F.conv1d(input, @weight, @bias, @stride, @padding, @dilation, @groups)
+      end
+    end
+  end
+end

data/lib/torch/nn/conv2d.rb

@@ -1,48 +1,26 @@
 module Torch
   module NN
-    class Conv2d < Module
-      attr_reader :bias, :weight
-
-      def initialize(in_channels, out_channels, kernel_size, stride: 1, padding: 0) #, dilation: 1, groups: 1)
-        @in_channels = in_channels
-        @out_channels = out_channels
-        @kernel_size = pair(kernel_size)
-        @stride = pair(stride)
-        @padding = pair(padding)
-        # @dilation = pair(dilation)
-
-        # TODO divide by groups
-        @weight = Parameter.new(Tensor.new(out_channels, in_channels, *@kernel_size))
-        @bias = Parameter.new(Tensor.new(out_channels))
-
-        reset_parameters
+    class Conv2d < ConvNd
+      def initialize(in_channels, out_channels, kernel_size, stride: 1,
+        padding: 0, dilation: 1, groups: 1, bias: true, padding_mode: "zeros")
+
+        kernel_size = _pair(kernel_size)
+        stride = _pair(stride)
+        padding = _pair(padding)
+        dilation = _pair(dilation)
+        super(in_channels, out_channels, kernel_size, stride, padding, dilation, false, _pair(0), groups, bias, padding_mode)
       end
 
-      def reset_parameters
-        Init.kaiming_uniform_(@weight, Math.sqrt(5))
-        if @bias
-          fan_in, _ = Init.calculate_fan_in_and_fan_out(@weight)
-          bound = 1 / Math.sqrt(fan_in)
-          Init.uniform_(@bias, -bound, bound)
+      def forward(input)
+        if @padding_mode == "circular"
+          raise NotImplementedError
         end
+        F.conv2d(input, @weight, @bias, @stride, @padding, @dilation, @groups)
       end
 
-      def call(input)
-        F.conv2d(input, @weight, @bias, stride: @stride, padding: @padding) #, @dilation, @groups)
-      end
-
-      def inspect
-        "Conv2d(#{@in_channels}, #{@out_channels}, kernel_size: #{@kernel_size.inspect}, stride: #{@stride.inspect})"
-      end
-
-      private
-
-      def pair(value)
-        if value.is_a?(Array)
-          value
-        else
-          [value] * 2
-        end
+      # TODO add more parameters
+      def extra_inspect
+        format("%s, %s, kernel_size: %s, stride: %s", @in_channels, @out_channels, @kernel_size, @stride)
       end
     end
   end

data/lib/torch/nn/conv3d.rb

@@ -0,0 +1,22 @@
+module Torch
+  module NN
+    class Conv3d < ConvNd
+      def initialize(in_channels, out_channels, kernel_size, stride: 1,
+        padding: 0, dilation: 1, groups: 1, bias: true, padding_mode: "zeros")
+
+        kernel_size = _triple(kernel_size)
+        stride = _triple(stride)
+        padding = _triple(padding)
+        dilation = _triple(dilation)
+        super(in_channels, out_channels, kernel_size, stride, padding, dilation, false, _triple(0), groups, bias, padding_mode)
+      end
+
+      def forward(input)
+        if @padding_mode == "circular"
+          raise NotImplementedError
+        end
+        F.conv3d(input, @weight, @bias, @stride, @padding, @dilation, @groups)
+      end
+    end
+  end
+end

data/lib/torch/nn/convnd.rb

@@ -0,0 +1,41 @@
+module Torch
+  module NN
+    class ConvNd < Module
+      def initialize(in_channels, out_channels, kernel_size, stride, padding, dilation, transposed, output_padding, groups, bias, padding_mode)
+        super()
+        raise ArgumentError, "in_channels must be divisible by groups" if in_channels % groups != 0
+        raise ArgumentError, "out_channels must be divisible by groups" if out_channels % groups != 0
+        @in_channels = in_channels
+        @out_channels = out_channels
+        @kernel_size = kernel_size
+        @stride = stride
+        @padding = padding
+        @dilation = dilation
+        @transposed = transposed
+        @output_padding = output_padding
+        @groups = groups
+        @padding_mode = padding_mode
+        if transposed
+          @weight = Parameter.new(Tensor.new(in_channels, out_channels / groups, *kernel_size))
+        else
+          @weight = Parameter.new(Tensor.new(out_channels, in_channels / groups, *kernel_size))
+        end
+        if bias
+          @bias = Parameter.new(Tensor.new(out_channels))
+        else
+          raise NotImplementedError
+        end
+        reset_parameters
+      end
+
+      def reset_parameters
+        Init.kaiming_uniform!(@weight, a: Math.sqrt(5))
+        if @bias
+          fan_in, _ = Init._calculate_fan_in_and_fan_out(@weight)
+          bound = 1 / Math.sqrt(fan_in)
+          Init.uniform!(@bias, a: -bound, b: bound)
+        end
+      end
+    end
+  end
+end

data/lib/torch/nn/cosine_embedding_loss.rb

@@ -0,0 +1,14 @@
+module Torch
+  module NN
+    class CosineEmbeddingLoss < Loss
+      def initialize(margin: 0, reduction: "mean")
+        super(reduction)
+        @margin = margin
+      end
+
+      def forward(input1, input2, target)
+        F.cosine_embedding_loss(input1, input2, target, margin: @margin, reduction: @reduction)
+      end
+    end
+  end
+end

data/lib/torch/nn/cosine_similarity.rb

@@ -0,0 +1,15 @@
+module Torch
+  module NN
+    class CosineSimilarity < Module
+      def initialize(dim: 1, eps: 1e-8)
+        super()
+        @dim = dim
+        @eps = eps
+      end
+
+      def forward(x1, x2)
+        F.cosine_similarity(x1, x2, dim: @dim, eps: @eps)
+      end
+    end
+  end
+end

data/lib/torch/nn/cross_entropy_loss.rb

@@ -0,0 +1,14 @@
+module Torch
+  module NN
+    class CrossEntropyLoss < WeightedLoss
+      def initialize(weight: nil, ignore_index: -100, reduction: "mean")
+        super(weight, reduction)
+        @ignore_index = ignore_index
+      end
+
+      def forward(input, target)
+        F.cross_entropy(input, target, weight: @weight, ignore_index: @ignore_index, reduction: @reduction)
+      end
+    end
+  end
+end

data/lib/torch/nn/ctc_loss.rb

@@ -0,0 +1,15 @@
+module Torch
+  module NN
+    class CTCLoss < Loss
+      def initialize(blank: 0, reduction: "mean", zero_infinity: false)
+        super(reduction)
+        @blank = blank
+        @zero_infinity = zero_infinity
+      end
+
+      def forward(log_probs, targets, input_lengths, target_lengths)
+        F.ctc_loss(log_probs, targets, input_lengths, target_lengths, blank: @blank, reduction: @reduction, zero_infinity: @zero_infinity)
+      end
+    end
+  end
+end

data/lib/torch/nn/dropout.rb

@@ -0,0 +1,9 @@
+module Torch
+  module NN
+    class Dropout < DropoutNd
+      def forward(input)
+        F.dropout(input, p: @p, training: @training, inplace: @inplace)
+      end
+    end
+  end
+end

data/lib/torch/nn/dropout2d.rb

@@ -0,0 +1,9 @@
+module Torch
+  module NN
+    class Dropout2d < DropoutNd
+      def forward(input)
+        F.dropout2d(input, p: @p, training: @training, inplace: @inplace)
+      end
+    end
+  end
+end

data/lib/torch/nn/dropout3d.rb

@@ -0,0 +1,9 @@
+module Torch
+  module NN
+    class Dropout3d < DropoutNd
+      def forward(input)
+        F.dropout3d(input, p: @p, training: @training, inplace: @inplace)
+      end
+    end
+  end
+end

data/lib/torch/nn/dropoutnd.rb

@@ -0,0 +1,15 @@
+module Torch
+  module NN
+    class DropoutNd < Module
+      def initialize(p: 0.5, inplace: false)
+        super()
+        @p = p
+        @inplace = inplace
+      end
+
+      def extra_inspect
+        format("p: %s, inplace: %s", @p, @inplace)
+      end
+    end
+  end
+end

data/lib/torch/nn/embedding.rb

@@ -0,0 +1,52 @@
+# ported from https://github.com/pytorch/pytorch/blob/master/torch/nn/modules/sparse.py
+module Torch
+  module NN
+    class Embedding < Module
+      def initialize(num_embeddings, embedding_dim, padding_idx: nil, max_norm: nil,
+        norm_type: 2.0, scale_grad_by_freq: false, sparse: false, _weight: nil)
+
+        super()
+        @num_embeddings = num_embeddings
+        @embedding_dim = embedding_dim
+
+        if padding_idx
+          if padding_idx > 0
+            raise ArgumentError, "Padding_idx must be within num_embeddings" unless padding_idx < @num_embeddings
+          elsif padding_idx < 0
+            raise ArgumentError, "Padding_idx must be within num_embeddings" unless padding_idx >= -@num_embeddings
+            padding_idx = @num_embeddings + padding_idx
+          end
+        end
+        @padding_idx = padding_idx
+        @max_norm = max_norm
+        @norm_type = norm_type
+        @scale_grad_by_freq = scale_grad_by_freq
+        if _weight.nil?
+          @weight = Parameter.new(Tensor.new(num_embeddings, embedding_dim))
+          reset_parameters
+        else
+          raise ArgumentError, "Shape of weight does not match num_embeddings and embedding_dim" unless _weight.shape == [num_embeddings, embedding_dim]
+          @weight = Parameter.new(_weight)
+        end
+        @sparse = sparse
+      end
+
+      def reset_parameters
+        Init.normal!(@weight)
+        if @padding_idx
+          Torch.no_grad do
+            @weight[@padding_idx].fill!(0)
+          end
+        end
+      end
+
+      def forward(input)
+        F.embedding(input, @weight, padding_idx: @padding_idx, max_norm: @max_norm, norm_type: @norm_type, scale_grad_by_freq: @scale_grad_by_freq, sparse: @sparse)
+      end
+
+      def inspect
+        "Embedding(#{@num_embeddings}, #{@embedding_dim})"
+      end
+    end
+  end
+end

data/lib/torch/nn/embedding_bag.rb

@@ -0,0 +1,34 @@
+# ported from https://github.com/pytorch/pytorch/blob/master/torch/nn/modules/sparse.py
+module Torch
+  module NN
+    class EmbeddingBag < Module
+      def initialize(num_embeddings, embedding_dim, max_norm: nil, norm_type: 2.0,
+        scale_grad_by_freq: false, mode: "mean", sparse: false, _weight: nil)
+
+        super()
+        @num_embeddings = num_embeddings
+        @embedding_dim = embedding_dim
+        @max_norm = max_norm
+        @norm_type = norm_type
+        @scale_grad_by_freq = scale_grad_by_freq
+        if _weight.nil?
+          @weight = Parameter.new(Tensor.new(num_embeddings, embedding_dim))
+          reset_parameters
+        else
+          raise ArgumentError, "Shape of weight does not match num_embeddings and embedding_dim" unless _weight.shape == [num_embeddings, embedding_dim]
+          @weight = Parameter.new(_weight)
+        end
+        @mode = mode
+        @sparse = sparse
+      end
+
+      def reset_parameters
+        Init.normal!(@weight)
+      end
+
+      def forward(input, offsets: nil, per_sample_weights: nil)
+        F.embedding_bag(input, @weight, offsets: offsets, max_norm: @max_norm, norm_type: @norm_type, scale_grad_by_freq: @scale_grad_by_freq, mode: @mode, sparse: @sparse, per_sample_weights: per_sample_weights)
+      end
+    end
+  end
+end

data/lib/torch/nn/feature_alpha_dropout.rb

@@ -0,0 +1,9 @@
+module Torch
+  module NN
+    class FeatureAlphaDropout < DropoutNd
+      def forward(input)
+        F.feature_alpha_dropout(input, p: @p, training: @training, inplace: @inplace)
+      end
+    end
+  end
+end

data/lib/torch/nn/fold.rb

@@ -0,0 +1,20 @@
+module Torch
+  module NN
+    class Fold < Module
+      def initialize(output_size, kernel_size, dilation: 1, padding: 0, stride: 1)
+        super()
+        @output_size = output_size
+        @kernel_size = kernel_size
+        @dilation = dilation
+        @padding = padding
+        @stride = stride
+      end
+
+      def forward(input)
+        F.fold(input, @output_size, @kernel_size, dilation: @dilation, padding: @padding, stride: @stride)
+      end
+
+      # TODO add extra_inspect
+    end
+  end
+end

data/lib/torch/nn/functional.rb

@@ -2,52 +2,441 @@ module Torch
   module NN
     class Functional
       class << self
-        def relu(input)
-          Torch.relu(input)
+        include Utils
+
+        # convolution layers
+
+        def conv1d(*args, **options)
+          Torch.conv1d(*args, **options)
+        end
+
+        def conv2d(*args, **options)
+          Torch.conv2d(*args, **options)
+        end
+
+        def conv3d(*args, **options)
+          Torch.conv3d(*args, **options)
+        end
+
+        def unfold(input, kernel_size, dilation: 1, padding: 0, stride: 1)
+          if input.dim == 4
+            NN.im2col(input, _pair(kernel_size), _pair(dilation), _pair(padding), _pair(stride))
+          else
+            raise Error, "Input Error: Only 4D input Tensors are supported (got #{input.dim}D)"
+          end
+        end
+
+        def fold(input, output_size, kernel_size, dilation: 1, padding: 0, stride: 1)
+          if input.dim == 3
+            NN.col2im(input, _pair(output_size), _pair(kernel_size), _pair(dilation), _pair(padding), _pair(stride))
+          else
+            raise Error, "Input Error: Only 3D input Tensors are supported (got #{input.dim}D)"
+          end
+        end
+
+        # pooling layers
+
+        def max_pool1d(*args, **options)
+          return_indices = args.pop if args.size == 7
+          if return_indices
+            Torch.max_pool1d_with_indices(*args, **options)
+          else
+            Torch.max_pool1d(*args, **options)
+          end
+        end
+
+        def max_pool2d(*args, **options)
+          return_indices = args.pop if args.size == 7
+          if return_indices
+            NN.max_pool2d_with_indices(*args, **options)
+          else
+            Torch.max_pool2d(*args, **options)
+          end
+        end
+
+        def max_pool3d(*args, **options)
+          return_indices = args.pop if args.size == 7
+          if return_indices
+            NN.max_pool3d_with_indices(*args, **options)
+          else
+            Torch.max_pool3d(*args, **options)
+          end
+        end
+
+        def max_unpool1d(input, indices, kernel_size, stride: nil, padding: 0, output_size: nil)
+          raise NotImplementedYet
+          kernel_size = _single(kernel_size)
+          if !stride.nil?
+            _stride = _single(stride)
+          else
+            _stride = kernel_size
+          end
+          padding = _single(padding)
+          output_size = _unpool_output_size(input, kernel_size, _stride, padding, output_size)
+          output_size = output_size + [1]
+          NN.max_unpool2d(input.unsqueeze(3), indices.unsqueeze(3), output_size).squeeze(3)
+        end
+
+        def max_unpool2d(*args, **options)
+          raise NotImplementedYet
+          NN.max_unpool2d(*args, **options)
+        end
+
+        def max_unpool3d(*args, **options)
+          raise NotImplementedYet
+          NN.max_unpool3d(*args, **options)
+        end
+
+        def avg_pool1d(*args, **options)
+          Torch.avg_pool1d(*args, **options)
+        end
+
+        def avg_pool2d(*args, **options)
+          NN.avg_pool2d(*args, **options)
+        end
+
+        def avg_pool3d(*args, **options)
+          NN.avg_pool3d(*args, **options)
+        end
+
+        # padding layers
+
+        def pad(input, pad, mode: "constant", value: 0)
+          raise ArgumentError, "Padding length must be divisible by 2" unless pad.size % 2 == 0
+          raise ArgumentError, "Padding length too large" unless pad.size / 2 <= input.dim
+
+          if mode == "constant"
+            return Torch.constant_pad_nd(input, pad, value)
+          else
+            raise ArgumentError, "Padding mode doesn't take in value argument" unless value == 0
+
+            if input.dim == 3
+              raise ArgumentError, "3D tensors expect 2 values for padding" unless pad.size == 2
+              case mode
+              when "reflect"
+                NN.reflection_pad1d(input, pad)
+              when "replicate"
+                NN.replication_pad1d(input, pad)
+              else
+                raise NotImplementedYet
+              end
+            elsif input.dim == 4
+              raise ArgumentError, "4D tensors expect 4 values for padding" unless pad.size == 4
+              case mode
+              when "reflect"
+                NN.reflection_pad2d(input, pad)
+              when "replicate"
+                NN.replication_pad2d(input, pad)
+              else
+                raise NotImplementedYet
+              end
+            elsif input.dim == 5
+              raise ArgumentError, "5D tensors expect 6 values for padding" unless pad.size == 6
+              case mode
+              when "replicate"
+                NN.replication_pad3d(input, pad)
+              else
+                raise NotImplementedYet
+              end
+            else
+              raise ArgumentError, "Only 3D, 4D, 5D padding with non-constant padding are supported for now"
+            end
+          end
+        end
+
+        # activation layers
+
+        def hardshrink(input, lambd = 0.5)
+          Torch.hardshrink(input, lambd)
+        end
+
+        def leaky_relu(input, negative_slope = 0.01)
+          NN.leaky_relu(input, negative_slope)
         end
 
-        def conv2d(input, weight, bias, stride: 1, padding: 0)
-          # TODO pair stride and padding when needed
-          Torch.conv2d(input, weight, bias, stride, padding)
+        def log_sigmoid(input)
+          NN.log_sigmoid(input)
         end
 
         def prelu(input, weight)
           Torch.prelu(input, weight)
         end
 
-        def leaky_relu(input, negative_slope = 0.01)
-          Torch.leaky_relu(input, negative_slope)
+        def relu(input, inplace: false)
+          if inplace
+            input.relu!
+          else
+            input.relu
+          end
+        end
+
+        def softplus(input, beta: 1, threshold: 20)
+          NN.softplus(input, beta, threshold)
+        end
+
+        def softshrink(*args, **options)
+          NN.softshrink(*args, **options)
+        end
+
+        def softsign(input)
+          input / (input.abs + 1)
+        end
+
+        def tanhshrink(input)
+          input - input.tanh
+        end
+
+        # other activation layers
+
+        def softmin(input, dim: nil)
+          dim ||= softmax_dim(input.dim)
+          (-input).softmax(dim)
+        end
+
+        def softmax(input, dim: nil)
+          dim ||= softmax_dim(input.dim)
+          input.softmax(dim)
+        end
+
+        # TODO make dim keyword argument and update examples
+        def log_softmax(input, dim = nil)
+          dim ||= softmax_dim(input.dim)
+          input.log_softmax(dim)
+        end
+
+        # normalization layers
+
+        def batch_norm(input, running_mean, running_var, weight: nil, bias: nil,
+          training: false, momentum: 0.1, eps: 1e-5)
+
+          if training
+            size = input.size
+            size_prods = size[0]
+            (size.length - 2).times do |i|
+              size_prods *= size[i + 2]
+            end
+            if size_prods == 1
+              raise ArgumentError, "Expected more than 1 value per channel when training, got input size #{size.inspect}"
+            end
+          end
+
+          Torch.batch_norm(
+            input, weight, bias, running_mean, running_var,
+            training, momentum, eps, false
+          )
+        end
+
+        def group_norm(input, num_groups, weight: nil, bias: nil, eps: 1e-5)
+          Torch.group_norm(input, num_groups, weight, bias, eps, false)
         end
 
-        def max_pool2d(input, kernel_size)
-          kernel_size = [kernel_size, kernel_size] if kernel_size.is_a?(Integer)
-          Torch.max_pool2d(input, kernel_size)
+        def instance_norm(input, running_mean: nil, running_var: nil, weight: nil,
+          bias: nil, use_input_stats: true, momentum: 0.1, eps: 1e-5)
+
+          Torch.instance_norm(
+              input, weight, bias, running_mean, running_var,
+              use_input_stats, momentum, eps, false
+          )
+        end
+
+        def layer_norm(input, normalized_shape, weight: nil, bias: nil, eps: 1e-5)
+          Torch.layer_norm(input, normalized_shape, weight, bias, eps, false)
         end
 
-        def avg_pool2d(input, kernel_size)
-          kernel_size = [kernel_size, kernel_size] if kernel_size.is_a?(Integer)
-          Torch.avg_pool2d(input, kernel_size)
+        def local_response_norm(input, size, alpha: 1e-4, beta: 0.75, k: 1.0)
+          dim = input.dim
+          if dim < 3
+            raise ArgumentError, "Expected 3D or higher dimensionality input (got #{dim} dimensions)"
+          end
+          div = input.mul(input).unsqueeze(1)
+          if dim == 3
+            div = pad(div, [0, 0, size / 2, (size - 1) / 2])
+            div = avg_pool2d(div, [size, 1], stride: 1).squeeze(1)
+          else
+            sizes = input.size
+            div = div.view(sizes[0], 1, sizes[1], sizes[2], -1)
+            div = pad(div, [0, 0, 0, 0, size / 2, (size - 1) / 2])
+            div = avg_pool3d(div, [size, 1, 1], stride: 1).squeeze(1)
+            div = div.view(sizes)
+          end
+          div = div.mul(alpha).add(k).pow(beta)
+          input / div
         end
 
+        # linear layers
+
         def linear(input, weight, bias)
-          Torch.linear(input, weight, bias)
+          NN.linear(input, weight, bias)
+        end
+
+        def bilinear(input1, input2, weight, bias)
+          Torch.bilinear(input1, input2, weight, bias)
+        end
+
+        # dropout layers
+
+        def dropout(input, p: 0.5, training: true, inplace: false)
+          if inplace
+            Torch.dropout!(input, p, training)
+          else
+            Torch.dropout(input, p, training)
+          end
+        end
+
+        def dropout2d(input, p: 0.5, training: true, inplace: false)
+          raise ArgumentError, "dropout probability has to be between 0 and 1, but got #{p}" if p < 0 || p > 1
+
+          if inplace
+            Torch.feature_dropout!(input, p, training)
+          else
+            Torch.feature_dropout(input, p, training)
+          end
+        end
+
+        def dropout3d(input, p: 0.5, training: true, inplace: false)
+          if inplace
+            Torch.feature_dropout!(input, p, training)
+          else
+            Torch.feature_dropout(input, p, training)
+          end
+        end
+
+        def alpha_dropout(input, p: 0.5, training: true, inplace: false)
+          if inplace
+            Torch.alpha_dropout!(input, p, training)
+          else
+            Torch.alpha_dropout(input, p, training)
+          end
+        end
+
+        def feature_alpha_dropout(input, p: 0.5, training: true, inplace: false)
+          if inplace
+            Torch.feature_alpha_dropout!(input, p, training)
+          else
+            Torch.feature_alpha_dropout(input, p, training)
+          end
+        end
+
+        # sparse layers
+
+        def embedding(input, weight, padding_idx: nil, max_norm: nil, norm_type: 2.0, scale_grad_by_freq: false, sparse: false)
+          # TODO handle max_norm and norm_type
+          raise NotImplementedYet unless max_norm.nil? && norm_type == 2.0
+
+          padding_idx ||= -1
+          # weight and indices are swapped from Python interface
+          Torch.embedding(weight, input, padding_idx, scale_grad_by_freq, sparse)
+        end
+
+        def embedding_bag(input, weight, offsets: nil, max_norm: nil, norm_type: 2, scale_grad_by_freq: false, mode: "mean", sparse: false, per_sample_weights: nil)
+          # TODO handle max_norm and norm_type
+          raise NotImplementedYet unless max_norm.nil? && norm_type == 2.0
+
+          mode_enum =
+            case mode
+            when "sum"
+              0
+            when "mean"
+              1
+            when "max"
+              2
+            else
+              raise ArgumentError, "Unknown mode: #{mode}"
+            end
+
+          # weight and input swapped
+          Torch.embedding_bag(weight, input, offsets, scale_grad_by_freq, mode_enum, sparse, per_sample_weights)
+        end
+
+        # distance functions
+
+        def cosine_similarity(x1, x2, dim: 1, eps: 1e-8)
+          Torch.cosine_similarity(x1, x2, dim, eps)
+        end
+
+        def pairwise_distance(x1, x2, p: 2.0, eps: 1e-6, keepdim: false)
+          Torch.pairwise_distance(x1, x2, p, eps, keepdim)
+        end
+
+        # loss functions
+
+        def binary_cross_entropy(input, target, weight: nil, reduction: "mean")
+          NN.binary_cross_entropy(input, target, weight, reduction)
+        end
+
+        def binary_cross_entropy_with_logits(input, target, weight: nil, reduction: "mean", pos_weight: nil)
+          Torch.binary_cross_entropy_with_logits(input, target, weight, pos_weight, reduction)
+        end
+
+        def cosine_embedding_loss(input1, input2, target, margin: 0, reduction: "mean")
+          raise NotImplementedYet
+        end
+
+        def cross_entropy(input, target, weight: nil, ignore_index: -100, reduction: "mean")
+          nll_loss(log_softmax(input, 1), target, weight: weight, ignore_index: ignore_index, reduction: reduction)
+        end
+
+        def ctc_loss(log_probs, targets, input_lengths, target_lengths, blank: 0, reduction: "mean", zero_infinity: false)
+          # call to_a on input_lengths and target_lengths for C++
+          Torch.ctc_loss(log_probs, targets, input_lengths.to_a, target_lengths.to_a, blank, reduction, zero_infinity)
+        end
+
+        def hinge_embedding_loss(input, target, margin: 1.0, reduction: "mean")
+          Torch.hinge_embedding_loss(input, target, margin, reduction)
+        end
+
+        def kl_div(input, target, reduction: "mean")
+          Torch.kl_div(input, target, reduction)
+        end
+
+        def l1_loss(input, target, reduction: "mean")
+          NN.l1_loss(input, target, reduction)
+        end
+
+        def margin_ranking_loss(input1, input2, target, margin: 0, reduction: "mean")
+          raise NotImplementedYet
         end
 
         def mse_loss(input, target, reduction: "mean")
-          Torch.mse_loss(input, target, reduction)
+          NN.mse_loss(input, target, reduction)
         end
 
-        def cross_entropy(input, target)
-          nll_loss(log_softmax(input, 1), target)
+        def multilabel_margin_loss(input, target, reduction: "mean")
+          NN.multilabel_margin_loss(input, target, reduction)
         end
 
-        def nll_loss(input, target)
-          # TODO fix for non-1d
-          Torch.nll_loss(input, target)
+        def multilabel_soft_margin_loss(input, target, weight: nil)
+          raise NotImplementedYet
         end
 
-        def log_softmax(input, dim)
-          input.log_softmax(dim)
+        def multi_margin_loss(input, target, p: 1, margin: 1.0, weight: nil, reduction: "mean")
+          NN.multi_margin_loss(input, target, p, margin, weight, reduction)
+        end
+
+        def nll_loss(input, target, weight: nil, ignore_index: -100, reduction: "mean")
+          NN.nll_loss(input, target, weight, reduction, ignore_index)
+        end
+
+        def poisson_nll_loss(input, target, log_input: true, full: false, eps: 1e-8, reduction: "mean")
+          Torch.poisson_nll_loss(input, target, log_input, full, eps, reduction)
+        end
+
+        def soft_margin_loss(input, target, reduction: "mean")
+          NN.soft_margin_loss(input, target, reduction)
+        end
+
+        def smooth_l1_loss(input, target, reduction: "mean")
+          NN.smooth_l1_loss(input, target, reduction)
+        end
+
+        def triplet_margin_loss(anchor, positive, negative, margin: 1.0, p: 2, eps: 1e-06, swap: false, reduction: "mean")
+          Torch.triplet_margin_loss(anchor, positive, negative, margin, p, eps, swap, reduction)
+        end
+
+        private
+
+        def softmax_dim(ndim)
+          ndim == 0 || ndim == 1 || ndim == 3 ? 0 : 1
         end
       end
     end