From 39ce5e32c2b228da3fed76fc29e4120626ef3dca Mon Sep 17 00:00:00 2001 From: liubiren Date: Mon, 19 Jan 2026 22:03:31 +0800 Subject: [PATCH] 1 --- 短视频合成自动化/export.py | 41 ++++++- 神经网络/main.py | 236 ++++++++++++++++++++++++++++++------ 神经网络/neural_network.npz | 0 3 files changed, 233 insertions(+), 44 deletions(-) create mode 100644 神经网络/neural_network.npz diff --git a/短视频合成自动化/export.py b/短视频合成自动化/export.py index 5e18102..d250852 100644 --- a/短视频合成自动化/export.py +++ b/短视频合成自动化/export.py @@ -98,7 +98,15 @@ class JianYingExport: "add_background_audio", "add_statement_video", "save", - ], + ], # 适用于淘宝闪购、存量抽手机 + "0002": [ + "add_subtitles_video", + "add_background_video", + "add_background_audio", + "add_logo_video", + "add_statement_video", + "save", + ], # 适用于视频号 } # 初始化工作配置 @@ -190,6 +198,13 @@ class JianYingExport: }, ], }, # 添加标识工作配置 + "add_logo_video": { + "material_path": self.materials["logo_video_material_path"], + "volume": [1.0], # 播放音量 + "clip_settings": [ + None, + ], # 图像调节设置 + }, # 添加标识视频工作配置 "add_statement": { "text": self.materials["statement_text"], "style": [ @@ -358,7 +373,18 @@ class JianYingExport: else: self.materials["logo_material_path"] = [] - # 声明（声明） + # 标识视频 + logo_video_folder_path = self.materials_folder_path / "标识视频" + if logo_video_folder_path.exists() and logo_video_folder_path.is_dir(): + self.materials["logo_video_material_path"] = [ + file_path + for file_path in logo_video_folder_path.rglob("*.mov") + if file_path.is_file() + ] + else: + self.materials["logo_video_material_path"] = [] + + # 声明文本 statement_path = self.materials_folder_path / "声明.txt" if statement_path.exists() and statement_path.is_file(): with open(statement_path, "r", encoding="utf-8") as file: @@ -369,7 +395,7 @@ class JianYingExport: else: self.materials["statement_text"] = [] - # 声明（视频） + # 声明视频 statement_video_folder_path = self.materials_folder_path / "声明视频" if ( statement_video_folder_path.exists() @@ -500,12 +526,17 @@ class JianYingExport: print("-> 正在添加标识...", end="") draft.add_video_segment(**parameters) print("已完成") - # 添加声明 + # 添加标识视频 + case "add_logo_video": + print("-> 正在添加标识视频...", end="") + draft.add_video_segment(**parameters) + print("已完成") + # 添加声明文本 case "add_statement": print("-> 正在添加声明...", end="") draft.add_text_segment(**parameters) print("已完成") - # 添加视频 + # 添加声明视频 case "add_statement_video": print("-> 正在添加声明视频...", end="") draft.add_video_segment(**parameters) diff --git a/神经网络/main.py b/神经网络/main.py index e0b299d..738e92d 100644 --- a/神经网络/main.py +++ b/神经网络/main.py @@ -4,7 +4,7 @@ """ # 导入模块 -from typing import List +from typing import List, Literal import numpy @@ -13,48 +13,206 @@ class NeuralNetwork: 神经网络 """ - # 激活函数和其导数函数 - FUNCTIONS = { - "relu": { - "activate": lambda x: numpy.maximum(0, x), - "derivative": lambda x: numpy.where(x > 0, 1, 0), - }, - "linear": { - "activate": lambda x: x, - "derivative": lambda x: numpy.ones_like(x), - }, # 适合回归任务的输出层 - "softmax": { - "activate": lambda x: numpy.exp(x) / numpy.sum(numpy.exp(x), axis=1), - "derivative": lambda x: x * (1 - x), - }, # 适合分类任务的输出层 - } - def __init__( self, - hidden_layer_neurons: List[int] = [10], - hidden_layer_function: str = "relu", - output_layer_function: str = "softmax", + neurons: List[int], + hidden_activate: Literal["relu"] = "relu", + output_activate: Literal["linear", "softmax"] = "linear", ): """ 初始化 - :param hidden_layer_neurons: 隐含层神经元数量 - :param hidden_layer_function: 隐含层函数 - :param output_layer_function: 输出层函数 + :param neurons: 神经元结构，例如[2, 10, 1]表示输入层为2个神经元、第一层隐含层为10个神经元、输出层为1个神经元 + :param hidden_activate: 隐含层激活函数，默认为relu + :param output_activate: 输出层激活函数，默认为linear """ - # 检查函数是否存在 - if not ( - hidden_layer_function in self.FUNCTIONS - and output_layer_function in self.FUNCTIONS - ): - raise RuntimeError("所输入的隐含层或输出层函数未定义") + print("正在初始化神经网络...", end="") - # 初始化隐含层的激活函数和导数函数 - self.hidden_layer_activate, self.hidden_layer_derivative = ( - self.FUNCTIONS[hidden_layer_function]["activate"], - self.FUNCTIONS[hidden_layer_function]["derivative"], - ) - # 初始化输出层的激活函数和导数函数 - self.output_layer_activate, self.output_layer_derivative = ( - self.FUNCTIONS[output_layer_function]["activate"], - self.FUNCTIONS[output_layer_function]["derivative"], - ) + # 初始化神经元结构 + self.neurons = neurons + + # 初始化隐含层激活函数 + self.hidden_activate = hidden_activate + # 初始化输出层激活函数 + self.output_activate = output_activate + + # 初始化神经网络结构 + self.neural_network = {} + # 初始化神经网络所有层权重和偏置 + self._init_neural_network() + + print("已完成") + + def _init_neural_network(self): + """ + 初始化神经网络所有层权重和偏置 + """ + for idx in range(1, len(self.neurons)): + # 若为隐含层则根据隐含层激活函数计算当前层权重的标准偏差，若为输出层则根据输出层激活函数计算当前层权重的标准偏差 + if idx != len(self.neurons) - 1: + # 激活函数 + activate = self.hidden_activate + match self.hidden_activate: + case "relu": + # 当前层权重的标准偏差 + standard_deviation = numpy.sqrt( + 2 / self.neurons[idx - 1] + ) # 使用He方差公式 + case _: + raise RuntimeError( + f"暂不支持该隐含层激活函数 {self.hidden_activate}" + ) + else: + # 激活函数 + activate = self.output_activate + match self.output_activate: + case "linear": + # 当前层权重的标准偏差 + standard_deviation = numpy.sqrt(1 / self.neurons[idx - 1]) + case "softmax": + # 当前层权重的标准偏差 + standard_deviation = numpy.sqrt( + 2 / (self.neurons[idx - 1] + self.neurons[idx]) + ) # 使用Xavier方差公式 + case _: + raise RuntimeError( + f"暂不支持该输出层激活函数 {self.output_activate}" + ) + + self.neural_network[f"layer:{idx:03d}"] = { + "weight": numpy.random.randn(self.neurons[idx - 1], self.neurons[idx]) + * standard_deviation, # 当前层权重 + "bias": numpy.zeros((1, self.neurons[idx])), # 当前层偏置 + "activate": activate, # 当前层激活函数 + "gamma": numpy.ones((1, self.neurons[idx])), # 当前层批标准化的缩放因子 + "beta": numpy.zeros((1, self.neurons[idx])), # 当前层批标准化的偏移因子 + } + + def _forward_propagate(self, x: numpy.ndarray) -> numpy.ndarray: + """ + 前向传播 + :param x: 输入层输入 + :return: 输出层预测值 + """ + activation = x # 将输入层输入作为第0层的激活值 + for layer_name, layer in self.neural_network.items(): + self.neural_network[layer_name].update( + { + "weighted_sum": ( + weighted_sum := numpy.dot(activation, layer["weight"]) + + layer["bias"] + ), # 当前层加权和 + "batch_normalized_weighted_sum": ( + batch_normalized_weighted_sum := layer["gamma"] + * ( + weighted_sum + - numpy.mean(weighted_sum, axis=0, keepdims=True) + ) + / numpy.sqrt( + numpy.var( + weighted_sum, ddof=0, axis=0, keepdims=True + ) # 使用有偏方差公式 + + 1e-8 + ) + + layer["beta"] + ), # 当前层批标准化加权和 + "activation": ( + activation := self._activate( + activate=layer["activate"], + weighted_sum=batch_normalized_weighted_sum, + ) + ), # 当前层激活值 + } + ) + + y_predict = activation # 将第L-1层（最后一层）的激活值作为输出层预测值（L为神经网络层数） + return y_predict + + def _activate( + self, + activate: Literal["relu", "linear", "softmax"], + weighted_sum: numpy.ndarray, + ) -> numpy.ndarray: + """ + 激活函数 + :param activate: 激活函数 + :param weighted_sum: 加权和 + :return: 激活值 + """ + match activate: + case "relu": + return numpy.maximum(0, weighted_sum) + case "linear": + return weighted_sum + case "softmax": + # 加权和指数项 + e_weighted_sum = numpy.exp( + weighted_sum - numpy.max(weighted_sum, axis=1, keepdims=True) + ) + return e_weighted_sum / numpy.sum(e_weighted_sum, axis=1, keepdims=True) + + def _calculate_loss( + self, + y_true: numpy.ndarray, + y_predict: numpy.ndarray, + ) -> numpy.floating: + """ + 计算损失 + :param y_true: 输出层真实值 + :param y_predict: 输出层预测值 + :return: 损失值 + """ + # 第L-1层（最后一层）的层名 + layer_name = list(self.neural_network.keys())[-1] + # 根据第L-1层（最后一层）的激活函数计算损失 + match activate := self.neural_network[layer_name]["activate"]: + case "linear": + loss = 0.5 * numpy.mean( + numpy.square(y_true - y_predict) + ) # 使用均方误差公式 + case "softmax": + loss = -1 * numpy.mean( + numpy.sum(y_true * numpy.log(y_predict + 1e-8), axis=1) + ) # 使用交叉熵损失公式 + case _: + raise RuntimeError(f"暂不支持该输出层激活函数 {activate}") + + return loss + + def _backward_propagate( + self, + x: numpy.ndarray, + y_true: numpy.ndarray, + y_predict: numpy.ndarray, + ) -> None: + """ + 后向传播 + :param x: 输入层输入 + :param y_true: 输出层真实值 + :param y_predict: 输出层预测值 + :return: 无 + """ + # 所有层的层名 + layer_names = list(self.neural_network.keys()) + + for idx, layer_name in enumerate(reversed(layer_names)): + # 当前层激活函数、加权和、批标准化加权和和激活值 + activate, weighted_sum, batch_normalized_weighted_sum, activation = ( + self.neural_network[layer_name]["activate"], + self.neural_network[layer_name]["weighted_sum"], + self.neural_network[layer_name]["batch_normalized_weighted_sum"], + self.neural_network[layer_name]["activation"], + ) + + # 输出层的误差项 + if idx == 0: + match activate: + case "linear" | "softmax": + delta = y_predict - y_true # 损失函数对第L-1层（最后一层）激活值的梯度 + case _: + raise RuntimeError(f"暂不支持该输出层激活函数 {activate}") + # 隐含层的误差项 + else: + delta = numpy.dot(delta, self.neural_network[layer_names[idx - 1]]["weight"].T) + + delta = 0 + diff --git a/神经网络/neural_network.npz b/神经网络/neural_network.npz new file mode 100644 index 0000000..e69de29