sources/libs/nvdsinferserver/infer_base_backend.h

Go to the documentation of this file.

/*
 * SPDX-FileCopyrightText: Copyright (c) 2019-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 * 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
 *
 * http://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 __NVDSINFER_BASE_BACKEND_H__
#define __NVDSINFER_BASE_BACKEND_H__
 
#include <stdarg.h>
#include <condition_variable>
#include <memory>
#include <mutex>
#include <queue>
 
#include "infer_datatypes.h"
#include "infer_ibackend.h"
#include "infer_utils.h"
 
namespace nvdsinferserver {
 
class BaseBackend : public IBackend {
public:
    ~BaseBackend() override = default;
    InferTensorOrder getInputTensorOrder() const final { return m_InputOrder; }
 
    void setUniqueId(uint32_t id) { m_UniqueId = id; }
 
    int uniqueId() const { return m_UniqueId; }
 
    void setFirstDimBatch(bool flag) { m_IsFirstDimBatch = flag; }
 
    bool isFirstDimBatch() const final { return m_IsFirstDimBatch; }
 
    uint32_t getLayerSize() const final {
        assert(!m_AllLayers.empty());
        return (int)m_AllLayers.size();
    }
 
    uint32_t getInputLayerSize() const final {
        return m_InputSize;
    }
 
    const LayerDescription* getLayerInfo(
        const std::string& bindingName) const final;
 
    LayersTuple getInputLayers() const final;
 
    LayersTuple getOutputLayers() const final;
 
    bool checkInputDims(const InputShapes& shapes) const;
 
    const LayerDescriptionList& allLayers() const { return m_AllLayers; }
 
    void setKeepInputs(bool enable) {
        m_KeepInputs = enable;
    }
 
    int32_t maxBatchSize() const final { return m_MaxBatchSize; }
 
    bool isNonBatching() const { return isNonBatch(maxBatchSize()); }
 
protected:
    using LayerIdxMap = std::unordered_map<std::string, int>;
 
    void resetLayers(LayerDescriptionList layers, int inputSize);
 
    LayerDescription* mutableLayerInfo(const std::string& bindingName)
    {
        const LayerDescription* info = getLayerInfo(bindingName);
        return const_cast<LayerDescription*>(info);
    }
 
    void setInputTensorOrder(InferTensorOrder order) { m_InputOrder = order; }
 
    bool needKeepInputs() const {
        return m_KeepInputs;
    }
 
    void setMaxBatchSize(uint32_t size) { m_MaxBatchSize = size; }
 
private:
    LayerDescriptionList m_AllLayers;
    LayerIdxMap m_LayerNameToIdx;
    uint32_t m_InputSize = 0;
    int32_t m_MaxBatchSize = 0;
    bool m_IsFirstDimBatch = false;
 
    InferTensorOrder m_InputOrder = InferTensorOrder::kNone;
    uint32_t m_UniqueId = 0;
    bool m_KeepInputs = false;
};
 
using UniqBackend = std::unique_ptr<BaseBackend>;
 
} // namespace nvdsinferserver
#endif