activations.h¶

Namespaces¶

Name
tenzr
tenzr::nn

Classes¶

	Name
class	tenzr::nn::ReLU Rectified linear unit activation module.
class	tenzr::nn::Sigmoid Logistic sigmoid activation module.
class	tenzr::nn::Tanh Hyperbolic tangent activation module.
class	tenzr::nn::SiLU SiLU (swish) activation module.
class	tenzr::nn::GELU Gaussian error linear unit activation module.
class	tenzr::nn::Softmax Softmax module normalizing over a configurable axis.
class	tenzr::nn::LogSoftmax Log-softmax module normalizing over a configurable axis.

Source code¶

// Copyright 2026 The Tenzr Project Authors
//
// SPDX-License-Identifier: Apache-2.0
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//      https://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

#ifndef TENZR_MODULES_NN_ACTIVATIONS_H_
#define TENZR_MODULES_NN_ACTIVATIONS_H_

#include <string>

#include "tenzr/frontend/tensor.h"
#include "tenzr/modules/nn/module.h"

namespace tenzr::nn {

  class ReLU : public Module {
   public:
    Tensor Forward(const Tensor& input) override { return input.Relu(); }
    std::string Name() const override { return "ReLU"; }
  };

  class Sigmoid : public Module {
   public:
    Tensor Forward(const Tensor& input) override { return input.sigmoid(); }
    std::string Name() const override { return "Sigmoid"; }
  };

  class Tanh : public Module {
   public:
    Tensor Forward(const Tensor& input) override { return input.tanh(); }
    std::string Name() const override { return "Tanh"; }
  };

  class SiLU : public Module {
   public:
    Tensor Forward(const Tensor& input) override { return input.Silu(); }
    std::string Name() const override { return "SiLU"; }
  };

  class GELU : public Module {
   public:
    Tensor Forward(const Tensor& input) override { return input.Gelu(); }
    std::string Name() const override { return "GELU"; }
  };

  Array softmax(const Array& input, int64_t axis = -1);

  Array log_softmax(const Array& input, int64_t axis = -1);

  class Softmax : public Module {
   public:
    explicit Softmax(int64_t axis = -1) : axis_(axis) {}
    Tensor Forward(const Tensor& input) override { return softmax(input, axis_); }
    std::string Name() const override { return "Softmax"; }

   private:
    int64_t axis_;
  };

  class LogSoftmax : public Module {
   public:
    explicit LogSoftmax(int64_t axis = -1) : axis_(axis) {}
    Tensor Forward(const Tensor& input) override { return log_softmax(input, axis_); }
    std::string Name() const override { return "LogSoftmax"; }

   private:
    int64_t axis_;
  };

}  // namespace tenzr::nn

#endif  // TENZR_MODULES_NN_ACTIVATIONS_H_