当前位置： 首页 > news >正文

做仓单的网站,为新公司取名,微网站免费创建平台,华升建设集团公司网站基于深度学习的花卉检测系统#xff08;含PyQt界面#xff09; 前言一、数据集1.1 数据集介绍1.2 数据预处理 二、模型搭建三、训练与测试3.1 模型训练3.2 模型测试 四、PyQt界面实现参考资料 前言 本项目是基于swin_transformer深度学习网络模型的花卉检测系统#xff0c;… 基于深度学习的花卉检测系统含PyQt界面 前言一、数据集1.1 数据集介绍1.2 数据预处理 二、模型搭建三、训练与测试3.1 模型训练3.2 模型测试 四、PyQt界面实现参考资料 前言 本项目是基于swin_transformer深度学习网络模型的花卉检测系统目前能够检测daisy、dandelion、roses、sunflowers、tulips五类花卉可以自己添加花卉种类进行训练。本文将详述数据集处理、模型构建、训练代码、以及基于PyQt5的应用界面设计。在应用中可以对花卉的图片进行识别输出花卉的类别和模型对其预测结果的置信度。本文附带了完整的应用界面设计、深度学习模型代码和训练数据集的下载链接。 完整资源下载链接博主在面包多网站上的完整资源下载页 项目演示视频 【项目分享】基于深度学习的花卉检测系统含PyQt界面 一、数据集 1.1 数据集介绍 本项目使用的数据集是由谷歌创建的一个用于机器学习和计算机视觉任务的图像数据集称为花卉数据集Flower Photos Dataset。它包含了来自五种不同花卉类别的图像每个类别大约有几百到一千张图像。这些花卉类别包括雏菊Daisy、蒲公英Dandelion、玫瑰Roses、向日葵Sunflowers、郁金香Tulips 。 下载链接http://download.tensorflow.org/example_images/flower_photos.tgz 下载后得到一个.tgr文件解压后文件夹下包含了5个子文件夹每个子文件夹都存储了一种类别的花的图片子文件夹的名称就是花的类别的名称如下图
1.2 数据预处理 使用MyDataSet类在 PyTorch 中加载图像数据并将其与相应的类别标签配对完成自定义数据集的生成。它包含初始化方法init来接收图像文件路径列表和对应的类别标签列表并提供了getitem方法来获取图像及其标签同时还可以使用collate_fn将多个样本进行批处理。 class MyDataSet(Dataset):自定义数据集def init(self, images_path: list, images_class: list, transformNone):self.images_path images_pathself.images_class images_classself.transform transformdef len(self):return len(self.images_path)def getitem(self, item):img Image.open(self.images_path[item])# RGB为彩色图片L为灰度图片if img.mode ! RGB:raise ValueError(image: {} isnt RGB mode..format(self.images_path[item]))label self.images_class[item]if self.transform is not None:img self.transform(img)return img, labelstaticmethoddef collate_fn(batch):# 官方实现的default_collate可以参考# https://github.com/pytorch/pytorch/blob/67b7e751e6b5931a9f45274653f4f653a4e6cdf6/torch/utils/data/_utils/collate.pyimages, labels tuple(zip(*batch))images torch.stack(images, dim0)labels torch.as_tensor(labels)return images, labels二、模型搭建 我们使用的是一种称为 Swin_Transformer 的新视觉 Transformer它可以作为 CV 的通用主干。将 Transformer 从语言适应到视觉方面的挑战来自两个域之间的差异例如视觉实体的规模以及相比于文本单词的高分辨率图像像素的巨大差异。为解决这些差异我们提出了一种 层次化 (hierarchical) Transformer其表示是用移位窗口 (Shifted Windows) 计算的。移位窗口方案通过将自注意力计算限制在不重叠的局部窗口的同时还允许跨窗口连接来提高效率。这种分层架构具有在各种尺度上建模的灵活性并且相对于图像大小具有线性计算复杂度。Swin Transformer 的这些特性使其与广泛的视觉任务兼容包括图像分类ImageNet-1K 的 87.3 top-1 Acc和密集预测任务例如目标检测COCO test dev 的 58.7 box AP 和 51.1 mask AP和语义分割ADE20K val 的 53.5 mIoU。它的性能在 COCO 上以 2.7 box AP 和 2.6 mask AP 以及在 ADE20K 上 3.2 mIoU 的大幅度超越了SOTA 技术证明了基于 Transformer 的模型作为视觉主干的潜力。分层设计和移位窗口方法也证明了其对全 MLP 架构是有益的。Swin_Transformer模型的整体架构如下图所示 而我们代码的模型具体实现主要包括如下几个模块PatchEmbed 模块、WindowAttention模块、SwinTransformerBlock模块 BasicLayer模块、SwinTransformer模块以及辅助函数drop_path_f等。 PatchEmbed 模块将输入图像划分为不重叠的图像块并将每个图像块转换为嵌入向量。 class PatchEmbed(nn.Module):2D Image to Patch Embeddingdef init(self, patch_size4, in_c3, embed_dim96, norm_layerNone):super().init()patch_size (patch_size, patch_size)self.patch_size patch_sizeself.in_chans in_cself.embed_dim embed_dimself.proj nn.Conv2d(in_c, embed_dim, kernel_sizepatch_size, stridepatch_size)self.norm norm_layer(embed_dim) if normlayer else nn.Identity()def forward(self, x):, _, H, W x.shape# padding# 如果输入图片的HW不是patch_size的整数倍需要进行paddingpad_input (H % self.patch_size[0] ! 0) or (W % self.patch_size[1] ! 0)if pad_input:# to pad the last 3 dimensions,# (W_left, W_right, H_top,H_bottom, C_front, C_back)x F.pad(x, (0, self.patch_size[1] - W % self.patch_size[1],0, self.patch_size[0] - H % self.patch_size[0],0, 0))# 下采样patchsize倍x self.proj(x), _, H, W x.shape# flatten: [B, C, H, W] - [B, C, HW]# transpose: [B, C, HW] - [B, HW, C]x x.flatten(2).transpose(1, 2)x self.norm(x)return x, H, WWindowAttention 模块基于窗口的多头自注意力机制用于捕获图像块之间的全局关系。 class WindowAttention(nn.Module):r Window based multi-head self attention (W-MSA) module with relative position bias.It supports both of shifted and non-shifted window.Args:dim (int): Number of input channels.window_size (tuple[int]): The height and width of the window.num_heads (int): Number of attention heads.qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: Trueattn_drop (float, optional): Dropout ratio of attention weight. Default: 0.0proj_drop (float, optional): Dropout ratio of output. Default: 0.0def init(self, dim, window_size, num_heads, qkv_biasTrue, attn_drop0., proj_drop0.):super().init()self.dim dimself.window_size window_size # [Mh, Mw]self.num_heads num_headshead_dim dim // num_headsself.scale head_dim ** -0.5# define a parameter table of relative position biasself.relative_position_bias_table nn.Parameter(torch.zeros((2 * window_size[0] - 1) * (2 * window_size[1] - 1), num_heads)) # [2*Mh-1 * 2*Mw-1, nH]# get pair-wise relative position index for each token inside the windowcoords_h torch.arange(self.window_size[0])coords_w torch.arange(self.window_size[1])coords torch.stack(torch.meshgrid([coords_h, coords_w])) # [2, Mh, Mw]coords_flatten torch.flatten(coords, 1) # [2, Mh*Mw]# [2, Mh*Mw, 1] - [2, 1, Mh*Mw]relative_coords coords_flatten[:, :, None] - coords_flatten[:, None, :] # [2, Mh*Mw, Mh*Mw]relative_coords relative_coords.permute(1, 2, 0).contiguous() # [Mh*Mw, Mh*Mw, 2]relative_coords[:, :, 0] self.window_size[0] - 1 # shift to start from 0relative_coords[:, :, 1] self.window_size[1] - 1relative_coords[:, :, 0] * 2 * self.window_size[1] - 1relative_position_index relative_coords.sum(-1) # [Mh*Mw, Mh*Mw]self.register_buffer(relative_position_index, relative_position_index)self.qkv nn.Linear(dim, dim * 3, biasqkv_bias)self.attn_drop nn.Dropout(attn_drop)self.proj nn.Linear(dim, dim)self.proj_drop nn.Dropout(proj_drop)nn.init.truncnormal(self.relative_position_bias_table, std.02)self.softmax nn.Softmax(dim-1)def forward(self, x, mask: Optional[torch.Tensor] None):Args:x: input features with shape of (num_windows*B, Mh*Mw, C)mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh*Ww) or None# [batch_size*num_windows, Mh*Mw, total_embeddim]B, N, C x.shape# qkv(): - [batch_size*num_windows, Mh*Mw, 3 * total_embed_dim]# reshape: - [batch_size*num_windows, Mh*Mw, 3, num_heads, embed_dim_per_head]# permute: - [3, batch_size*num_windows, num_heads, Mh*Mw, embed_dim_perhead]qkv self.qkv(x).reshape(B, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)# [batch_size*num_windows, num_heads, Mh*Mw, embed_dim_per_head]q, k, v qkv.unbind(0) # make torchscript happy (cannot use tensor as tuple)# transpose: - [batch_size*num_windows, num_heads, embed_dim_per_head, Mh*Mw]# : multiply - [batch_size*num_windows, num_heads, Mh*Mw, Mh*Mw]q q * self.scaleattn (q k.transpose(-2, -1))# relative_position_bias_table.view: [Mh*Mw*Mh*Mw,nH] - [Mh*Mw,Mh*Mw,nH]relative_position_bias self.relative_position_bias_table[self.relative_position_index.view(-1)].view(self.window_size[0] * self.window_size[1], self.window_size[0] * self.window_size[1], -1)relative_position_bias relative_position_bias.permute(2, 0, 1).contiguous() # [nH, Mh*Mw, Mh*Mw]attn attn relative_position_bias.unsqueeze(0)if mask is not None:# mask: [nW, Mh*Mw, Mh*Mw]nW mask.shape[0] # num_windows# attn.view: [batch_size, num_windows, numheads, Mh*Mw, Mh*Mw]# mask.unsqueeze: [1, nW, 1, Mh*Mw, Mh*Mw]attn attn.view(B // nW, nW, self.num_heads, N, N) mask.unsqueeze(1).unsqueeze(0)attn attn.view(-1, self.num_heads, N, N)attn self.softmax(attn)else:attn self.softmax(attn)attn self.attn_drop(attn)# : multiply - [batch_size*num_windows, num_heads, Mh*Mw, embed_dim_per_head]# transpose: - [batch_size*num_windows, Mh*Mw, num_heads, embed_dim_per_head]# reshape: - [batch_size*num_windows, Mh*Mw, total_embeddim]x (attn v).transpose(1, 2).reshape(B, N, C)x self.proj(x)x self.proj_drop(x)return xSwinTransformerBlock 模块Swin Transformer 的基本模块包含了窗口注意力机制和MLP前馈网络。 class SwinTransformerBlock(nn.Module):r Swin Transformer Block.Args:dim (int): Number of input channels.num_heads (int): Number of attention heads.window_size (int): Window size.shift_size (int): Shift size for SW-MSA.mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: Truedrop (float, optional): Dropout rate. Default: 0.0attn_drop (float, optional): Attention dropout rate. Default: 0.0drop_path (float, optional): Stochastic depth rate. Default: 0.0act_layer (nn.Module, optional): Activation layer. Default: nn.GELUnorm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNormdef init(self, dim, num_heads, window_size7, shift_size0,mlp_ratio4., qkv_biasTrue, drop0., attn_drop0., drop_path0.,act_layernn.GELU, norm_layernn.LayerNorm):super().init()self.dim dimself.num_heads num_headsself.window_size window_sizeself.shift_size shift_sizeself.mlp_ratio mlp_ratioassert 0 self.shift_size self.window_size, shift_size must in 0-window_sizeself.norm1 norm_layer(dim)self.attn WindowAttention(dim, window_size(self.window_size, self.window_size), num_headsnum_heads, qkv_biasqkv_bias,attn_dropattn_drop, proj_dropdrop)self.drop_path DropPath(drop_path) if drop_path 0. else nn.Identity()self.norm2 norm_layer(dim)mlp_hidden_dim int(dim * mlp_ratio)self.mlp Mlp(in_featuresdim, hidden_featuresmlp_hidden_dim, act_layeract_layer, dropdrop)def forward(self, x, attn_mask):H, W self.H, self.WB, L, C x.shapeassert L H * W, input feature has wrong sizeshortcut xx self.norm1(x)x x.view(B, H, W, C)# pad feature maps to multiples of window size# 把feature map给pad到window size的整数倍pad_l pad_t 0pad_r (self.window_size - W % self.window_size) % self.window_sizepad_b (self.window_size - H % self.window_size) % self.window_sizex F.pad(x, (0, 0, pad_l, pad_r, pad_t, padb)), Hp, Wp, _ x.shape# cyclic shiftif self.shift_size 0:shifted_x torch.roll(x, shifts(-self.shift_size, -self.shift_size), dims(1, 2))else:shifted_x xattn_mask None# partition windowsx_windows window_partition(shifted_x, self.window_size) # [nW*B, Mh, Mw, C]x_windows x_windows.view(-1, self.window_size * self.window_size, C) # [nW*B, Mh*Mw, C]# W-MSA/SW-MSAattn_windows self.attn(x_windows, maskattn_mask) # [nW*B, Mh*Mw, C]# merge windowsattn_windows attn_windows.view(-1, self.window_size, self.window_size, C) # [nW*B, Mh, Mw, C]shifted_x window_reverse(attn_windows, self.window_size, Hp, Wp) # [B, H, W, C]# reverse cyclic shiftif self.shift_size 0:x torch.roll(shifted_x, shifts(self.shift_size, self.shift_size), dims(1, 2))else:x shifted_xif pad_r 0 or pad_b 0:# 把前面pad的数据移除掉x x[:, :H, :W, :].contiguous()x x.view(B, H * W, C)# FFNx shortcut self.drop_path(x)x x self.drop_path(self.mlp(self.norm2(x)))return xBasicLayer 模块用于构建 Swin Transformer 的一个阶段可以包含多个 SwinTransformerBlock 模块。 class BasicLayer(nn.Module):A basic Swin Transformer layer for one stage.Args:dim (int): Number of input channels.depth (int): Number of blocks.num_heads (int): Number of attention heads.window_size (int): Local window size.mlp_ratio (float): Ratio of mlp hidden dim to embedding dim.qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: Truedrop (float, optional): Dropout rate. Default: 0.0attn_drop (float, optional): Attention dropout rate. Default: 0.0drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNormdownsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: Noneuse_checkpoint (bool): Whether to use checkpointing to save memory. Default: False.def init(self, dim, depth, num_heads, window_size,mlp_ratio4., qkv_biasTrue, drop0., attn_drop0.,drop_path0., norm_layernn.LayerNorm, downsampleNone, use_checkpointFalse):super().init()self.dim dimself.depth depthself.window_size window_sizeself.use_checkpoint use_checkpointself.shift_size window_size // 2# build blocksself.blocks nn.ModuleList([SwinTransformerBlock(dimdim,num_headsnum_heads,window_sizewindow_size,shift_size0 if (i % 2 0) else self.shift_size,mlp_ratiomlp_ratio,qkv_biasqkv_bias,dropdrop,attn_dropattn_drop,drop_pathdrop_path[i] if isinstance(drop_path, list) else drop_path,norm_layernorm_layer)for i in range(depth)])# patch merging layerif downsample is not None:self.downsample downsample(dimdim, norm_layernorm_layer)else:self.downsample Nonedef create_mask(self, x, H, W):# calculate attention mask for SW-MSA# 保证Hp和Wp是window_size的整数倍Hp int(np.ceil(H / self.window_size)) * self.window_sizeWp int(np.ceil(W / self.window_size)) * self.window_size# 拥有和feature map一样的通道排列顺序方便后续window_partitionimg_mask torch.zeros((1, Hp, Wp, 1), devicex.device) # [1, Hp, Wp, 1]h_slices (slice(0, -self.window_size),slice(-self.window_size, -self.shift_size),slice(-self.shift_size, None))w_slices (slice(0, -self.window_size),slice(-self.window_size, -self.shift_size),slice(-self.shift_size, None))cnt 0for h in h_slices:for w in w_slices:img_mask[:, h, w, :] cntcnt 1mask_windows window_partition(img_mask, self.window_size) # [nW, Mh, Mw, 1]mask_windows mask_windows.view(-1, self.window_size * self.window_size) # [nW, Mh*Mw]attn_mask mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2) # [nW, 1, Mh*Mw] - [nW, Mh*Mw, 1]# [nW, Mh*Mw, Mh*Mw]attn_mask attn_mask.masked_fill(attn_mask ! 0, float(-100.0)).masked_fill(attn_mask 0, float(0.0))return attn_maskdef forward(self, x, H, W):attn_mask self.create_mask(x, H, W) # [nW, Mh*Mw, Mh*Mw]for blk in self.blocks:blk.H, blk.W H, Wif not torch.jit.is_scripting() and self.use_checkpoint:x checkpoint.checkpoint(blk, x, attn_mask)else:x blk(x, attn_mask)if self.downsample is not None:x self.downsample(x, H, W)H, W (H 1) // 2, (W 1) // 2return x, H, WSwinTransformer 模块整个 Swin Transformer 模型的主体结构包含了多个 BasicLayer 模块。 class SwinTransformer(nn.Module):r Swin TransformerA PyTorch impl of : Swin Transformer: Hierarchical Vision Transformer using Shifted Windows -https://arxiv.org/pdf/2103.14030Args:patch_size (int | tuple(int)): Patch size. Default: 4in_chans (int): Number of input image channels. Default: 3num_classes (int): Number of classes for classification head. Default: 1000embed_dim (int): Patch embedding dimension. Default: 96depths (tuple(int)): Depth of each Swin Transformer layer.num_heads (tuple(int)): Number of attention heads in different layers.window_size (int): Window size. Default: 7mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4qkv_bias (bool): If True, add a learnable bias to query, key, value. Default: Truedrop_rate (float): Dropout rate. Default: 0attn_drop_rate (float): Attention dropout rate. Default: 0drop_path_rate (float): Stochastic depth rate. Default: 0.1norm_layer (nn.Module): Normalization layer. Default: nn.LayerNorm.patch_norm (bool): If True, add normalization after patch embedding. Default: Trueuse_checkpoint (bool): Whether to use checkpointing to save memory. Default: Falsedef init(self, patch_size4, in_chans3, num_classes1000,embed_dim96, depths(2, 2, 6, 2), num_heads(3, 6, 12, 24),window_size7, mlp_ratio4., qkv_biasTrue,drop_rate0., attn_drop_rate0., drop_path_rate0.1,norm_layernn.LayerNorm, patch_normTrue,use_checkpointFalse, **kwargs):super().init()self.num_classes num_classesself.num_layers len(depths)self.embed_dim embed_dimself.patch_norm patch_norm# stage4输出特征矩阵的channelsself.num_features int(embed_dim * 2 ** (self.num_layers - 1))self.mlp_ratio mlp_ratio# split image into non-overlapping patchesself.patch_embed PatchEmbed(patch_sizepatch_size, in_cin_chans, embed_dimembed_dim,norm_layernorm_layer if self.patch_norm else None)self.pos_drop nn.Dropout(pdrop_rate)# stochastic depthdpr [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))] # stochastic depth decay rule# build layersself.layers nn.ModuleList()for i_layer in range(self.num_layers):# 注意这里构建的stage和论文图中有些差异# 这里的stage不包含该stage的patch_merging层包含的是下个stage的layers BasicLayer(dimint(embed_dim * 2 ** i_layer),depthdepths[i_layer],num_headsnum_heads[i_layer],window_sizewindow_size,mlp_ratioself.mlp_ratio,qkv_biasqkv_bias,dropdrop_rate,attn_dropattn_drop_rate,drop_pathdpr[sum(depths[:i_layer]):sum(depths[:i_layer 1])],norm_layernorm_layer,downsamplePatchMerging if (i_layer self.num_layers - 1) else None,use_checkpointuse_checkpoint)self.layers.append(layers)self.norm norm_layer(self.num_features)self.avgpool nn.AdaptiveAvgPool1d(1)self.head nn.Linear(self.num_features, num_classes) if num_classes 0 else nn.Identity()self.apply(self._init_weights)def _init_weights(self, m):if isinstance(m, nn.Linear):nn.init.truncnormal(m.weight, std.02)if isinstance(m, nn.Linear) and m.bias is not None:nn.init.constant(m.bias, 0)elif isinstance(m, nn.LayerNorm):nn.init.constant(m.bias, 0)nn.init.constant_(m.weight, 1.0)def forward(self, x):# x: [B, L, C]x, H, W self.patch_embed(x)x self.pos_drop(x)for layer in self.layers:x, H, W layer(x, H, W)x self.norm(x) # [B, L, C]x self.avgpool(x.transpose(1, 2)) # [B, C, 1]x torch.flatten(x, 1)x self.head(x)return x辅助函数drop_path_f 用于实现随机深度路径Stochastic Depth以及一些用于处理窗口的辅助函数。 def drop_path_f(x, drop_prob: float 0., training: bool False):Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).This is the same as the DropConnect impl I created for EfficientNet, etc networks, however,the original name is misleading as Drop Connect is a different form of dropout in a separate paper…See discussion: https://github.com/tensorflow/tpu/issues/494#issuecomment-532968956 … Ive opted forchanging the layer and argument names to drop path rather than mix DropConnect as a layer name and usesurvival rate as the argument.if drop_prob 0. or not training:return xkeep_prob 1 - drop_probshape (x.shape[0],) (1,) * (x.ndim - 1) # work with diff dim tensors, not just 2D ConvNetsrandom_tensor keep_prob torch.rand(shape, dtypex.dtype, devicex.device)randomtensor.floor() # binarizeoutput x.div(keep_prob) * random_tensorreturn output三、训练与测试 3.1 模型训练 我们训练的模型是在通用的预训练模型swin_base_patch4_window7_224.pth上再次训练的通过模型训练微调能给得到一个效果更好的花卉检测模型。 首先设置模型训练的关键参数如检测目标类别数目可以按照自己的数据集和检测种类进行设置、批量大小、训练周期、输入数据的维度等参数。 parser argparse.ArgumentParser()parser.add_argument(–num_classes, typeint, default5)parser.add_argument(–epochs, typeint, default100)parser.add_argument(–batch-size, typeint, default16)parser.add_argument(–lr, typefloat, default0.0001)# 数据集所在根目录# http://download.tensorflow.org/example_images/flower_photos.tgzparser.add_argument(–data-path, typestr,defaultflower_photos)# 预训练权重路径如果不想载入就设置为空字符parser.add_argument(–weights, typestr, default./swin_base_patch4_window7_224.pth,helpinitial weights path)# 是否冻结权重parser.add_argument(–freeze-layers, typebool, defaultFalse)parser.add_argument(–device, defaultcuda:0, helpdevice id (i.e. 0 or 0,1 or cpu))然后通过下面代码设置模型训练设备和文件夹路径。接着对数据进行预处理并创建数据集和数据加载器。并根据命令行参数配置模型并加载预训练权重可选择性地冻结部分模型参数。最后使用AdamW优化器进行训练并在每个epoch结束时保存模型权重。整个训练过程可以记录损失、准确率等指标并将其写入TensorBoard。 def main(args):device torch.device(args.device if torch.cuda.is_available() else cpu)if os.path.exists(./weights) is False:os.makedirs(./weights)tb_writer SummaryWriter()train_images_path, train_images_label, val_images_path, val_images_label read_split_data(args.data_path)img_size 224data_transform {train: transforms.Compose([transforms.RandomResizedCrop(img_size),transforms.RandomHorizontalFlip(),transforms.ToTensor(),transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])]),val: transforms.Compose([transforms.Resize(int(img_size * 1.143)),transforms.CenterCrop(img_size),transforms.ToTensor(),transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])}# 实例化训练数据集train_dataset MyDataSet(images_pathtrain_images_path,images_classtrain_images_label,transformdata_transform[train])# 实例化验证数据集val_dataset MyDataSet(images_pathval_images_path,images_classval_images_label,transformdata_transform[val])batch_size args.batch_sizenw min([os.cpu_count(), batch_size if batch_size 1 else 0, 8]) # number of workersprint(Using {} dataloader workers every process.format(nw))train_loader torch.utils.data.DataLoader(train_dataset,batch_sizebatch_size,shuffleTrue,pin_memoryTrue,num_workersnw,collate_fntrain_dataset.collate_fn)val_loader torch.utils.data.DataLoader(val_dataset,batch_sizebatch_size,shuffleFalse,pin_memoryTrue,num_workersnw,collate_fnval_dataset.collate_fn)model create_model(num_classesargs.num_classes).to(device)if args.weights ! :assert os.path.exists(args.weights), weights file: {} not exist..format(args.weights)weights_dict torch.load(args.weights, map_locationdevice)[model]# 删除有关分类类别的权重for k in list(weights_dict.keys()):if head in k:del weights_dict[k]print(model.load_state_dict(weights_dict, strictFalse))if args.freeze_layers:for name, para in model.named_parameters():# 除head外其他权重全部冻结if head not in name:para.requiresgrad(False)else:print(training {}.format(name))pg [p for p in model.parameters() if p.requires_grad]optimizer optim.AdamW(pg, lrargs.lr, weight_decay5E-2)for epoch in range(args.epochs):# traintrain_loss, train_acc train_one_epoch(modelmodel,optimizeroptimizer,data_loadertrain_loader,devicedevice,epochepoch)# validateval_loss, val_acc evaluate(modelmodel,data_loaderval_loader,devicedevice,epochepoch)train_acc_list.append(train_acc)train_loss_list.append(train_loss)val_acc_list.append(val_acc)val_loss_list.append(val_loss)tags [train_loss, train_acc, val_loss, val_acc, learning_rate]tb_writer.add_scalar(tags[0], train_loss, epoch)tb_writer.add_scalar(tags[1], train_acc, epoch)tb_writer.add_scalar(tags[2], val_loss, epoch)tb_writer.add_scalar(tags[3], val_acc, epoch)tb_writer.add_scalar(tags[4], optimizer.param_groups[0][lr], epoch)torch.save(model.state_dict(), ./weights/model-{}.pth.format(epoch))整个训练过程可以记录损失、准确率等指标
3.2 模型测试 可以分别使用predict.py对单张花卉图片和predict-batch.py批量进行检测。

predict.py

def main(img_path):import osos.environ[KMP_DUPLICATE_LIB_OK] TRUEdevice torch.device(cuda:0 if torch.cuda.is_available() else cpu)img_size 224data_transform transforms.Compose([transforms.Resize(int(img_size * 1.143)),transforms.CenterCrop(img_size),transforms.ToTensor(),transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])# load image# img_path ./tulip.jpgassert os.path.exists(img_path), file: {} dose not exist..format(img_path)img Image.open(img_path)plt.imshow(img)# [N, C, H, W]img data_transform(img)# expand batch dimensionimg torch.unsqueeze(img, dim0)# read class_indictjson_path ./class_indices.jsonassert os.path.exists(json_path), file: {} dose not exist..format(json_path)json_file open(json_path, r)class_indict json.load(json_file)# create modelmodel create_model(num_classes5).to(device)# load model weightsmodel_weight_path ./weights/model-86.pthmodel.load_state_dict(torch.load(model_weight_path, map_locationdevice))model.eval()with torch.no_grad():# predict classoutput torch.squeeze(model(img.to(device))).cpu()predict torch.softmax(output, dim0)predict_cla torch.argmax(predict).numpy()# print_res class: {} prob: {:.3}.format(class_indict[str(predict_cla)],# predict[predict_cla].numpy())# plt.title(print_res)for i in range(len(predict)):print(class: {:10} prob: {:.3}.format(class_indict[str(i)],predict[i].numpy()))# plt.show()res class_indict[str(list(predict.numpy()).index(max(predict.numpy())))]num %.2f % (max(predict.numpy()) * 100) %print(res,num)return res,max(predict.numpy())# print(class_indict[str(list(predict.numpy()).index(max(predict.numpy())))])def main():device torch.device(cuda:0 if torch.cuda.is_available() else cpu)img_size 224data_transform transforms.Compose([transforms.Resize(int(img_size * 1.143)),transforms.CenterCrop(img_size),transforms.ToTensor(),transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])# read class_indictjson_path ./class_indices.jsonassert os.path.exists(json_path), file: {} dose not exist..format(json_path)json_file open(json_path, r)class_indict json.load(json_file)# create modelmodel create_model(num_classes5).to(device)# load model weightsmodel_weight_path ./weights/model-86.pthmodel.load_state_dict(torch.load(model_weight_path, map_locationdevice))model.eval()# load imagedata_root os.path.abspath(os.path.join(os.getcwd(), ../)) # get data root pathall_dir os.path.join(data_root, data_set) # flower data set path# img_path_list [../tulip.jpg, ../rose.jpg]img_list []test_dir os.path.join(all_dir, jpg) # testtest_datasets datasets.ImageFolder(test_dir, transformdata_transform)for img_path, idx in test_datasets.imgs:assert os.path.exists(img_path), file: {} dose not exist..format(img_path)# img_path ./tulip.jpgassert os.path.exists(img_path), file: {} dose not exist..format(img_path)img Image.open(img_path)plt.imshow(img)# [N, C, H, W]img data_transform(img)# expand batch dimensionimg torch.unsqueeze(img, dim0)with torch.no_grad():# predict classoutput torch.squeeze(model(img.to(device))).cpu()predict torch.softmax(output, dim0)predict_cla torch.argmax(predict).numpy()print_res image: {} class: {} prob: {:.3}.format(img_path, class_indict[str(predict_cla)],predict[predict_cla].numpy())print(print_res)测试结果 四、PyQt界面实现 当整个项目构建完成后使用PyQt5编写可视化界面可以支持花卉图像的检测。运行主界面.py然后点击文件夹图片传入待检测的花卉图像即可。经过花卉识别系统识别后会输出相应的类别和置信度。
参考资料 论文https://arxiv.org/pdf/2103.14030.pdf代码https://github.com/microsoft/Swin-Transformertimmhttps://hub.fastgit.org/rwightman/pytorch-image-models/blob/master/timm/models/swin_transformer.pySwin_Transformer网络模型详解资料详解Swin_Transformer (SwinT)