import os import numpy as np import torch import torch.nn as nn import gradio as gr from torchvision.models import efficientnet_v2_m, EfficientNet_V2_M_Weights import torch.nn.functional as F from torchvision import transforms from PIL import Image from data_manager import get_dog_description from urllib.parse import quote os.system('pip install ultralytics') from ultralytics import YOLO from PIL import ImageDraw # 下載YOLOv8預訓練模型 model_yolo = YOLO('yolov8n.pt') # 使用 YOLOv8 預訓練模型 dog_breeds = ["Afghan_Hound", "African_Hunting_Dog", "Airedale", "American_Staffordshire_Terrier", "Appenzeller", "Australian_Terrier", "Bedlington_Terrier", "Bernese_Mountain_Dog", "Blenheim_Spaniel", "Border_Collie", "Border_Terrier", "Boston_Bull", "Bouvier_Des_Flandres", "Brabancon_Griffon", "Brittany_Spaniel", "Cardigan", "Chesapeake_Bay_Retriever", "Chihuahua", "Dandie_Dinmont", "Doberman", "English_Foxhound", "English_Setter", "English_Springer", "EntleBucher", "Eskimo_Dog", "French_Bulldog", "German_Shepherd", "German_Short-Haired_Pointer", "Gordon_Setter", "Great_Dane", "Great_Pyrenees", "Greater_Swiss_Mountain_Dog", "Ibizan_Hound", "Irish_Setter", "Irish_Terrier", "Irish_Water_Spaniel", "Irish_Wolfhound", "Italian_Greyhound", "Japanese_Spaniel", "Kerry_Blue_Terrier", "Labrador_Retriever", "Lakeland_Terrier", "Leonberg", "Lhasa", "Maltese_Dog", "Mexican_Hairless", "Newfoundland", "Norfolk_Terrier", "Norwegian_Elkhound", "Norwich_Terrier", "Old_English_Sheepdog", "Pekinese", "Pembroke", "Pomeranian", "Rhodesian_Ridgeback", "Rottweiler", "Saint_Bernard", "Saluki", "Samoyed", "Scotch_Terrier", "Scottish_Deerhound", "Sealyham_Terrier", "Shetland_Sheepdog", "Shih-Tzu", "Siberian_Husky", "Staffordshire_Bullterrier", "Sussex_Spaniel", "Tibetan_Mastiff", "Tibetan_Terrier", "Walker_Hound", "Weimaraner", "Welsh_Springer_Spaniel", "West_Highland_White_Terrier", "Yorkshire_Terrier", "Affenpinscher", "Basenji", "Basset", "Beagle", "Black-and-Tan_Coonhound", "Bloodhound", "Bluetick", "Borzoi", "Boxer", "Briard", "Bull_Mastiff", "Cairn", "Chow", "Clumber", "Cocker_Spaniel", "Collie", "Curly-Coated_Retriever", "Dhole", "Dingo", "Flat-Coated_Retriever", "Giant_Schnauzer", "Golden_Retriever", "Groenendael", "Keeshond", "Kelpie", "Komondor", "Kuvasz", "Malamute", "Malinois", "Miniature_Pinscher", "Miniature_Poodle", "Miniature_Schnauzer", "Otterhound", "Papillon", "Pug", "Redbone", "Schipperke", "Silky_Terrier", "Soft-Coated_Wheaten_Terrier", "Standard_Poodle", "Standard_Schnauzer", "Toy_Poodle", "Toy_Terrier", "Vizsla", "Whippet", "Wire-Haired_Fox_Terrier"] class MultiHeadAttention(nn.Module): def __init__(self, in_dim, num_heads=8): super().__init__() self.num_heads = num_heads self.head_dim = max(1, in_dim // num_heads) self.scaled_dim = self.head_dim * num_heads self.fc_in = nn.Linear(in_dim, self.scaled_dim) self.query = nn.Linear(self.scaled_dim, self.scaled_dim) self.key = nn.Linear(self.scaled_dim, self.scaled_dim) self.value = nn.Linear(self.scaled_dim, self.scaled_dim) self.fc_out = nn.Linear(self.scaled_dim, in_dim) def forward(self, x): N = x.shape[0] x = self.fc_in(x) q = self.query(x).view(N, self.num_heads, self.head_dim) k = self.key(x).view(N, self.num_heads, self.head_dim) v = self.value(x).view(N, self.num_heads, self.head_dim) energy = torch.einsum("nqd,nkd->nqk", [q, k]) attention = F.softmax(energy / (self.head_dim ** 0.5), dim=2) out = torch.einsum("nqk,nvd->nqd", [attention, v]) out = out.reshape(N, self.scaled_dim) out = self.fc_out(out) return out class BaseModel(nn.Module): def __init__(self, num_classes, device='cuda' if torch.cuda.is_available() else 'cpu'): super().__init__() self.device = device self.backbone = efficientnet_v2_m(weights=EfficientNet_V2_M_Weights.IMAGENET1K_V1) self.feature_dim = self.backbone.classifier[1].in_features self.backbone.classifier = nn.Identity() self.num_heads = max(1, min(8, self.feature_dim // 64)) self.attention = MultiHeadAttention(self.feature_dim, num_heads=self.num_heads) self.classifier = nn.Sequential( nn.LayerNorm(self.feature_dim), nn.Dropout(0.3), nn.Linear(self.feature_dim, num_classes) ) self.to(device) def forward(self, x): x = x.to(self.device) features = self.backbone(x) attended_features = self.attention(features) logits = self.classifier(attended_features) return logits, attended_features num_classes = 120 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') model = BaseModel(num_classes=num_classes, device=device) checkpoint = torch.load('best_model_81_dog.pth', map_location=torch.device('cpu')) model.load_state_dict(checkpoint['model_state_dict']) # evaluation mode model.eval() # Image preprocessing function def preprocess_image(image): # If the image is numpy.ndarray turn into PIL.Image if isinstance(image, np.ndarray): image = Image.fromarray(image) # Use torchvision.transforms to process images transform = transforms.Compose([ transforms.Resize((224, 224)), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]), ]) return transform(image).unsqueeze(0) def get_akc_breeds_link(): return "https://www.akc.org/dog-breeds/" # def predict(image): # if image is None: # return "Please upload an image to get started.", gr.update(visible=False), gr.update(visible=False), gr.update(visible=False) # try: # image_tensor = preprocess_image(image) # with torch.no_grad(): # output = model(image_tensor) # logits = output[0] if isinstance(output, tuple) else output # probabilities = F.softmax(logits, dim=1) # topk_probs, topk_indices = torch.topk(probabilities, k=3) # top1_prob = topk_probs[0][0].item() # topk_breeds = [dog_breeds[idx.item()] for idx in topk_indices[0]] # topk_probs_percent = [f"{prob.item() * 100:.2f}%" for prob in topk_probs[0]] # if top1_prob >= 0.5: # breed = topk_breeds[0] # description = get_dog_description(breed) # return format_description(description, breed), gr.update(visible=False), gr.update(visible=False), gr.update(visible=False) # elif top1_prob < 0.2: # return ("The image is too unclear or the dog breed is not in the dataset. Please upload a clearer image of the dog.", # gr.update(visible=False), gr.update(visible=False), gr.update(visible=False)) # else: # explanation = ( # f"The model couldn't confidently identify the breed. Here are the top 3 possible breeds:\n\n" # f"1. **{topk_breeds[0]}** ({topk_probs_percent[0]} confidence)\n" # f"2. **{topk_breeds[1]}** ({topk_probs_percent[1]} confidence)\n" # f"3. **{topk_breeds[2]}** ({topk_probs_percent[2]} confidence)\n\n" # "Click on a button to view more information about the breed." # ) # return explanation, gr.update(visible=True, value=f"More about {topk_breeds[0]}"), gr.update(visible=True, value=f"More about {topk_breeds[1]}"), gr.update(visible=True, value=f"More about {topk_breeds[2]}") # except Exception as e: # return f"An error occurred: {e}", gr.update(visible=False), gr.update(visible=False), gr.update(visible=False) # def format_description(description, breed): # if isinstance(description, dict): # formatted_description = "\n\n".join([f"**{key}**: {value}" for key, value in description.items()]) # else: # formatted_description = description # akc_link = get_akc_breeds_link() # formatted_description += f"\n\n**Want to learn more about dog breeds?** [Visit the AKC dog breeds page]({akc_link}) and search for {breed} to find detailed information." # disclaimer = ("\n\n*Disclaimer: The external link provided leads to the American Kennel Club (AKC) dog breeds page. " # "You may need to search for the specific breed on that page. " # "I am not responsible for the content on external sites. " # "Please refer to the AKC's terms of use and privacy policy.*") # formatted_description += disclaimer # return formatted_description # def show_details(breed): # breed_name = breed.split("More about ")[-1] # description = get_dog_description(breed_name) # return format_description(description, breed_name) # with gr.Blocks(css=""" # .container { # max-width: 900px; # margin: 0 auto; # padding: 20px; # background-color: rgba(255, 255, 255, 0.9); # border-radius: 15px; # box-shadow: 0 0 20px rgba(0, 0, 0, 0.1); # } # .gr-form { display: flex; flex-direction: column; align-items: center; } # .gr-box { width: 100%; max-width: 500px; } # .output-markdown, .output-image { # margin-top: 20px; # padding: 15px; # background-color: #f5f5f5; # border-radius: 10px; # } # .examples { # display: flex; # justify-content: center; # flex-wrap: wrap; # gap: 10px; # margin-top: 20px; # } # .examples img { # width: 100px; # height: 100px; # object-fit: cover; # } # """) as iface: # gr.HTML("

🐶 Dog Breed Classifier 🔍

") # gr.HTML("

Upload a picture of a dog, and the model will predict its breed, provide detailed information, and include an extra information link!

") # with gr.Row(): # input_image = gr.Image(label="Upload a dog image", type="numpy") # output = gr.Markdown(label="Prediction Results") # with gr.Row(): # btn1 = gr.Button("View More 1", visible=False) # btn2 = gr.Button("View More 2", visible=False) # btn3 = gr.Button("View More 3", visible=False) # input_image.change(predict, inputs=input_image, outputs=[output, btn1, btn2, btn3]) # btn1.click(show_details, inputs=btn1, outputs=output) # btn2.click(show_details, inputs=btn2, outputs=output) # btn3.click(show_details, inputs=btn3, outputs=output) # gr.Examples( # examples=['Border_Collie.jpg', 'Golden_Retriever.jpeg', 'Saint_Bernard.jpeg', 'French_Bulldog.jpeg', 'Samoyed.jpg'], # inputs=input_image # ) # gr.HTML('For more details on this project and other work, feel free to visit my GitHub Dog Breed Classifier') # # launch the program # if __name__ == "__main__": # iface.launch() # 使用 YOLOv8 進行狗偵測 def detect_dogs(image): results = yolo_model.predict(image) dogs = [] for result in results: for box in result.boxes: if box.cls == 16: # COCO 資料集中的狗類別是16 xyxy = box.xyxy confidence = box.conf cropped_image = image.crop((xyxy[0], xyxy[1], xyxy[2], xyxy[3])) dogs.append((cropped_image, confidence, xyxy)) return dogs def predict(image): if image is None: return "Please upload an image to get started.", gr.update(visible=False), gr.update(visible=False), gr.update(visible=False) try: # 確保圖片轉換為 PIL.Image 格式 if isinstance(image, np.ndarray): image = Image.fromarray(image) # 使用 YOLO 偵測狗 dogs = detect_dogs(image) if len(dogs) == 0: return "No dogs detected or the image is too unclear. Please upload a clearer image.", gr.update(visible=False), gr.update(visible=False), gr.update(visible=False) # 開始處理每一隻狗 explanations = [] visible_buttons = [] annotated_image = image.copy() for i, (cropped_image, confidence, box) in enumerate(dogs): image_tensor = preprocess_image(cropped_image) with torch.no_grad(): output = model(image_tensor) logits = output[0] if isinstance(output, tuple) else output probabilities = F.softmax(logits, dim=1) topk_probs, topk_indices = torch.topk(probabilities, k=3) top1_prob = topk_probs[0][0].item() topk_breeds = [dog_breeds[idx.item()] for idx in topk_indices[0]] topk_probs_percent = [f"{prob.item() * 100:.2f}%" for prob in topk_probs[0]] # 標註狗的邊界框 draw = ImageDraw.Draw(annotated_image) draw.rectangle(box.tolist(), outline="red", width=3) draw.text((box[0], box[1]), f"Dog {i+1}", fill="red") # 信心度大於 50%，顯示詳細品種資訊 if top1_prob >= 0.5: breed = topk_breeds[0] description = get_dog_description(breed) explanations.append(f"Dog {i+1}: **{breed}**\n{format_description(description, breed)}") # 信心度 20%-49%，顯示 Top 3 品種 elif 0.2 <= top1_prob < 0.5: explanation = ( f"Dog {i+1}: Detected with moderate confidence. Here are the top 3 possible breeds:\n" f"1. **{topk_breeds[0]}** ({topk_probs_percent[0]})\n" f"2. **{topk_breeds[1]}** ({topk_probs_percent[1]})\n" f"3. **{topk_breeds[2]}** ({topk_probs_percent[2]})\n" ) explanations.append(explanation) visible_buttons.extend([f"More about {topk_breeds[0]}", f"More about {topk_breeds[1]}", f"More about {topk_breeds[2]}"]) else: explanations.append(f"Dog {i+1}: The image is unclear or the breed is not in the dataset.") final_explanation = "\n\n".join(explanations) return annotated_image, final_explanation, gr.update(visible=len(visible_buttons) >= 1), gr.update(visible=len(visible_buttons) >= 2), gr.update(visible=len(visible_buttons) >= 3) except Exception as e: return f"An error occurred: {e}", gr.update(visible=False), gr.update(visible=False), gr.update(visible=False) def format_description(description, breed): if isinstance(description, dict): formatted_description = "\n\n".join([f"**{key}**: {value}" for key, value in description.items()]) else: formatted_description = description akc_link = get_akc_breeds_link() formatted_description += f"\n\n**Want to learn more about dog breeds?** [Visit the AKC dog breeds page]({akc_link}) and search for {breed} to find detailed information." disclaimer = ("\n\n*Disclaimer: The external link provided leads to the American Kennel Club (AKC) dog breeds page. " "You may need to search for the specific breed on that page. " "I am not responsible for the content on external sites. " "Please refer to the AKC's terms of use and privacy policy.*") formatted_description += disclaimer return formatted_description def show_details(breed): breed_name = breed.split("More about ")[-1] description = get_dog_description(breed_name) return format_description(description, breed_name) with gr.Blocks(css=""" .container { max-width: 900px; margin: 0 auto; padding: 20px; background-color: rgba(255, 255, 255, 0.9); border-radius: 15px; box-shadow: 0 0 20px rgba(0, 0, 0, 0.1); } .gr-form { display: flex; flex-direction: column; align-items: center; } .gr-box { width: 100%; max-width: 500px; } .output-markdown, .output-image { margin-top: 20px; padding: 15px; background-color: #f5f5f5; border-radius: 10px; } .examples { display: flex; justify-content: center; flex-wrap: wrap; gap: 10px; margin-top: 20px; } .examples img { width: 100px; height: 100px; object-fit: cover; } """) as iface: gr.HTML("

🐶 Dog Breed Classifier 🔍

") gr.HTML("

Upload a picture of a dog, and the model will predict its breed, provide detailed information, and include an extra information link!

") with gr.Row(): input_image = gr.Image(label="Upload a dog image", type="numpy") output_image = gr.Image(label="Annotated Image") output = gr.Markdown(label="Prediction Results") with gr.Row(): btn1 = gr.Button("View More 1", visible=False) btn2 = gr.Button("View More 2", visible=False) btn3 = gr.Button("View More 3", visible=False) input_image.change(predict, inputs=input_image, outputs=[output_image, output, btn1, btn2, btn3]) btn1.click(show_details, inputs=btn1, outputs=output) btn2.click(show_details, inputs=btn2, outputs=output) btn3.click(show_details, inputs=btn3, outputs=output) gr.Examples( examples=['Border_Collie.jpg', 'Golden_Retriever.jpeg', 'Saint_Bernard.jpeg', 'French_Bulldog.jpeg', 'Samoyed.jpg'], inputs=input_image ) gr.HTML('For more details on this project and other work, feel free to visit my GitHub Dog Breed Classifier') # launch the program if __name__ == "__main__": iface.launch()