FFmpeg
dnn_backend_native.h
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2018 Sergey Lavrushkin
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU Lesser General Public
8  * License as published by the Free Software Foundation; either
9  * version 2.1 of the License, or (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14  * Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with FFmpeg; if not, write to the Free Software
18  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
21 /**
22  * @file
23  * DNN inference functions interface for native backend.
24  */
25 
26 
27 #ifndef AVFILTER_DNN_DNN_BACKEND_NATIVE_H
28 #define AVFILTER_DNN_DNN_BACKEND_NATIVE_H
29 
30 #include "../dnn_interface.h"
31 #include "libavformat/avio.h"
32 #include "libavutil/opt.h"
33 
34 /**
35  * the enum value of DNNLayerType should not be changed,
36  * the same values are used in convert_from_tensorflow.py
37  * and, it is used to index the layer execution/load function pointer.
38  */
39 typedef enum {
40  DLT_INPUT = 0,
48  DLT_DENSE = 8,
50 } DNNLayerType;
51 
55 
56 typedef struct Layer{
58  /**
59  * a layer can have multiple inputs and one output.
60  * 4 is just a big enough number for input operands (increase it if necessary),
61  * do not use 'int32_t *input_operand_indexes', so we don't worry about mem leaks.
62  */
65  void *params;
66 } Layer;
67 
68 typedef struct DnnOperand{
69  /**
70  * there are two memory layouts, NHWC or NCHW, so we use dims,
71  * dims[0] is Number.
72  */
74 
75  /**
76  * input/output/intermediate operand of the network
77  */
79 
80  /**
81  * support different kinds of data type such as float, half float, int8 etc,
82  * first support float now.
83  */
85 
86  /**
87  * NHWC if 1, otherwise NCHW.
88  * let's first support NHWC only, this flag is for extensive usage.
89  */
90  int8_t isNHWC;
91 
92  /**
93  * to avoid possible memory leak, do not use char *name
94  */
95  char name[128];
96 
97  /**
98  * data pointer with data length in bytes.
99  * usedNumbersLeft is only valid for intermediate operand,
100  * it means how many layers still depend on this operand,
101  * todo: the memory can be reused when usedNumbersLeft is zero.
102  */
103  void *data;
106 }DnnOperand;
107 
108 typedef struct InputParams{
110 } InputParams;
111 
112 typedef struct NativeOptions{
113  uint32_t conv2d_threads;
114 } NativeOptions;
115 
116 typedef struct NativeContext {
117  const AVClass *class;
119 } NativeContext;
120 
121 // Represents simple feed-forward convolutional network.
122 typedef struct NativeModel{
129 } NativeModel;
130 
131 DNNModel *ff_dnn_load_model_native(const char *model_filename, DNNFunctionType func_type, const char *options, AVFilterContext *filter_ctx);
132 
134 
135 void ff_dnn_free_model_native(DNNModel **model);
136 
137 // NOTE: User must check for error (return value <= 0) to handle
138 // case like integer overflow.
141 #endif
DLT_COUNT
@ DLT_COUNT
Definition: dnn_backend_native.h:49
InputParams
Definition: dnn_backend_native.h:108
opt.h
filter_ctx
static FilteringContext * filter_ctx
Definition: transcoding.c:48
ff_calculate_operand_dims_count
int32_t ff_calculate_operand_dims_count(const DnnOperand *oprd)
Definition: dnn_backend_native.c:389
DnnOperand::isNHWC
int8_t isNHWC
NHWC if 1, otherwise NCHW.
Definition: dnn_backend_native.h:90
DNNFunctionType
DNNFunctionType
Definition: dnn_interface.h:51
DLT_AVG_POOL
@ DLT_AVG_POOL
Definition: dnn_backend_native.h:47
NONE
@ NONE
Definition: dnn_backend_native.h:54
NativeModel::operands
DnnOperand * operands
Definition: dnn_backend_native.h:127
DLT_INPUT
@ DLT_INPUT
Definition: dnn_backend_native.h:40
DLT_MATH_BINARY
@ DLT_MATH_BINARY
Definition: dnn_backend_native.h:45
SIGMOID
@ SIGMOID
Definition: dnn_backend_native.h:54
DNNLayerType
DNNLayerType
the enum value of DNNLayerType should not be changed, the same values are used in convert_from_tensor...
Definition: dnn_backend_native.h:39
DnnOperand::type
DNNOperandType type
input/output/intermediate operand of the network
Definition: dnn_backend_native.h:78
NativeModel::ctx
NativeContext ctx
Definition: dnn_backend_native.h:123
NativeModel::layers_num
int32_t layers_num
Definition: dnn_backend_native.h:126
DLT_MAXIMUM
@ DLT_MAXIMUM
Definition: dnn_backend_native.h:44
TANH
@ TANH
Definition: dnn_backend_native.h:54
ff_dnn_load_model_native
DNNModel * ff_dnn_load_model_native(const char *model_filename, DNNFunctionType func_type, const char *options, AVFilterContext *filter_ctx)
Definition: dnn_backend_native.c:116
Layer::type
DNNLayerType type
Definition: dnn_backend_native.h:57
DLT_CONV2D
@ DLT_CONV2D
Definition: dnn_backend_native.h:41
DnnOperand::name
char name[128]
to avoid possible memory leak, do not use char *name
Definition: dnn_backend_native.h:95
DnnOperand::data
void * data
data pointer with data length in bytes.
Definition: dnn_backend_native.h:103
DNNReturnType
DNNReturnType
Definition: dnn_interface.h:33
DnnOperand::data_type
DNNDataType data_type
support different kinds of data type such as float, half float, int8 etc, first support float now.
Definition: dnn_backend_native.h:84
InputParams::height
int height
Definition: dnn_backend_native.h:109
SAME_CLAMP_TO_EDGE
@ SAME_CLAMP_TO_EDGE
Definition: dnn_backend_native.h:53
ff_dnn_execute_model_native
DNNReturnType ff_dnn_execute_model_native(const DNNModel *model, DNNExecBaseParams *exec_params)
Definition: dnn_backend_native.c:376
DOT_OUTPUT
@ DOT_OUTPUT
Definition: dnn_backend_native.h:52
AVClass
Describe the class of an AVClass context structure.
Definition: log.h:67
Layer::params
void * params
Definition: dnn_backend_native.h:65
NativeModel
Definition: dnn_backend_native.h:122
DnnOperand::dims
int32_t dims[4]
there are two memory layouts, NHWC or NCHW, so we use dims, dims[0] is Number.
Definition: dnn_backend_native.h:73
SAME
@ SAME
Definition: dnn_backend_native.h:53
DOT_INTERMEDIATE
@ DOT_INTERMEDIATE
Definition: dnn_backend_native.h:52
DNNOperandType
DNNOperandType
Definition: dnn_backend_native.h:52
DnnOperand::length
int32_t length
Definition: dnn_backend_native.h:104
DOT_INPUT
@ DOT_INPUT
Definition: dnn_backend_native.h:52
NativeModel::model
DNNModel * model
Definition: dnn_backend_native.h:124
options
const OptionDef options[]
Layer::output_operand_index
int32_t output_operand_index
Definition: dnn_backend_native.h:64
DLT_MIRROR_PAD
@ DLT_MIRROR_PAD
Definition: dnn_backend_native.h:43
NativeContext
Definition: dnn_backend_native.h:116
Layer
Definition: dnn_backend_native.h:56
Layer::input_operand_indexes
int32_t input_operand_indexes[4]
a layer can have multiple inputs and one output.
Definition: dnn_backend_native.h:63
DnnOperand::usedNumbersLeft
int32_t usedNumbersLeft
Definition: dnn_backend_native.h:105
NativeModel::layers
Layer * layers
Definition: dnn_backend_native.h:125
avio.h
VALID
@ VALID
Definition: dnn_backend_native.h:53
DNNPaddingParam
DNNPaddingParam
Definition: dnn_backend_native.h:53
DNNDataType
DNNDataType
Definition: dnn_interface.h:37
RELU
@ RELU
Definition: dnn_backend_native.h:54
NativeModel::operands_num
int32_t operands_num
Definition: dnn_backend_native.h:128
ff_calculate_operand_data_length
int32_t ff_calculate_operand_data_length(const DnnOperand *oprd)
Definition: dnn_backend_native.c:398
DNNActivationFunc
DNNActivationFunc
Definition: dnn_backend_native.h:54
NativeContext::options
NativeOptions options
Definition: dnn_backend_native.h:118
NativeOptions
Definition: dnn_backend_native.h:112
NativeOptions::conv2d_threads
uint32_t conv2d_threads
Definition: dnn_backend_native.h:113
DnnOperand
Definition: dnn_backend_native.h:68
DLT_MATH_UNARY
@ DLT_MATH_UNARY
Definition: dnn_backend_native.h:46
ff_dnn_free_model_native
void ff_dnn_free_model_native(DNNModel **model)
Definition: dnn_backend_native.c:410
AVFilterContext
An instance of a filter.
Definition: avfilter.h:333
DNNModel
Definition: dnn_interface.h:83
LEAKY_RELU
@ LEAKY_RELU
Definition: dnn_backend_native.h:54
DLT_DENSE
@ DLT_DENSE
Definition: dnn_backend_native.h:48
int32_t
int32_t
Definition: audioconvert.c:56
DNNExecBaseParams
Definition: dnn_interface.h:66
DLT_DEPTH_TO_SPACE
@ DLT_DEPTH_TO_SPACE
Definition: dnn_backend_native.h:42
InputParams::channels
int channels
Definition: dnn_backend_native.h:109
InputParams::width
int width
Definition: dnn_backend_native.h:109