commit 3547566c4bf5c96a84043aa46327e702083aea80 Author: 2509165046 <2509165046@student.edu.cn> Date: Thu Jul 2 23:58:26 2026 +0800 4-2-image-labeling diff --git a/data/images/标注练习1.jpg b/data/images/标注练习1.jpg new file mode 100644 index 0000000..ce14835 Binary files /dev/null and b/data/images/标注练习1.jpg differ diff --git a/data/reviews.json b/data/reviews.json new file mode 100644 index 0000000..161ff31 --- /dev/null +++ b/data/reviews.json @@ -0,0 +1,7 @@ +[ + {"id": 1, "text": "外卖小哥送得超快，餐盒还是热的，炸鸡酥脆多汁，酸辣粉也很正宗，分量足，五星好评！"}, + {"id": 2, "text": "等了一个半小时才送到，汤全洒了，面坨成一坨，联系客服也不回，太让人失望了。"}, + {"id": 3, "text": "奶茶是用料很扎实的现煮茶，珍珠Q弹有嚼劲，配送员态度也好，下次还会再点。"}, + {"id": 4, "text": "配送速度一般，但披萨味道不错，芝士拉丝效果好，性价比高，值得推荐。"}, + {"id": 5, "text": "点的麻辣烫食材不新鲜，有股怪味，吃完拉肚子，商家推卸责任，再也不点了。"} +] diff --git a/q2_1_crawler/move.html b/q2_1_crawler/move.html new file mode 100644 index 0000000..667e109 --- /dev/null +++ b/q2_1_crawler/move.html @@ -0,0 +1,102 @@ +[ + { + "id": "1", + "title": "泰坦尼克号", + "director": "Frank Darabont", + "year": "2015", + "rating": "6.8", + "duration": "91", + "genre": "科幻", + "actors_count": "3" + }, + { + "id": "2", + "title": "星际穿越", + "director": "陈凯歌", + "year": "2021", + "rating": "6.2", + "duration": "113", + "genre": "科幻", + "actors_count": "2" + }, + { + "id": "3", + "title": "三傻大闹宝莱坞", + "director": "Robert Zemeckis", + "year": "2004", + "rating": "7.4", + "duration": "95", + "genre": "悬疑", + "actors_count": "4" + }, + { + "id": "4", + "title": "阿甘正传", + "director": "James Cameron", + "year": "2013", + "rating": "6.9", + "duration": "93", + "genre": "爱情", + "actors_count": "4" + }, + { + "id": "5", + "title": "放牛班的春天", + "director": "宫崎骏", + "year": "2005", + "rating": "7.1", + "duration": "127", + "genre": "悬疑", + "actors_count": "3" + }, + { + "id": "6", + "title": "千与千寻", + "director": "Christopher Nolan", + "year": "2024", + "rating": "6.4", + "duration": "147", + "genre": "动画", + "actors_count": "3" + }, + { + "id": "7", + "title": "忠犬八公的故事", + "director": "Lasse Hallström", + "year": "2002", + "rating": "6.2", + "duration": "166", + "genre": "剧情", + "actors_count": "4" + }, + { + "id": "8", + "title": "霸王别姬", + "director": "Rajkumar Hirani", + "year": "2005", + "rating": "7.9", + "duration": "149", + "genre": "冒险", + "actors_count": "2" + }, + { + "id": "9", + "title": "肖申克的救赎", + "director": "Christophe Barratier", + "year": "2008", + "rating": "9.3", + "duration": "91", + "genre": "冒险", + "actors_count": "2" + }, + { + "id": "10", + "title": "盗梦空间", + "director": "Christopher Nolan", + "year": "2019", + "rating": "7.1", + "duration": "132", + "genre": "剧情", + "actors_count": "5" + } +] \ No newline at end of file diff --git a/q2_1_crawler/movie.json b/q2_1_crawler/movie.json new file mode 100644 index 0000000..667e109 --- /dev/null +++ b/q2_1_crawler/movie.json @@ -0,0 +1,102 @@ +[ + { + "id": "1", + "title": "泰坦尼克号", + "director": "Frank Darabont", + "year": "2015", + "rating": "6.8", + "duration": "91", + "genre": "科幻", + "actors_count": "3" + }, + { + "id": "2", + "title": "星际穿越", + "director": "陈凯歌", + "year": "2021", + "rating": "6.2", + "duration": "113", + "genre": "科幻", + "actors_count": "2" + }, + { + "id": "3", + "title": "三傻大闹宝莱坞", + "director": "Robert Zemeckis", + "year": "2004", + "rating": "7.4", + "duration": "95", + "genre": "悬疑", + "actors_count": "4" + }, + { + "id": "4", + "title": "阿甘正传", + "director": "James Cameron", + "year": "2013", + "rating": "6.9", + "duration": "93", + "genre": "爱情", + "actors_count": "4" + }, + { + "id": "5", + "title": "放牛班的春天", + "director": "宫崎骏", + "year": "2005", + "rating": "7.1", + "duration": "127", + "genre": "悬疑", + "actors_count": "3" + }, + { + "id": "6", + "title": "千与千寻", + "director": "Christopher Nolan", + "year": "2024", + "rating": "6.4", + "duration": "147", + "genre": "动画", + "actors_count": "3" + }, + { + "id": "7", + "title": "忠犬八公的故事", + "director": "Lasse Hallström", + "year": "2002", + "rating": "6.2", + "duration": "166", + "genre": "剧情", + "actors_count": "4" + }, + { + "id": "8", + "title": "霸王别姬", + "director": "Rajkumar Hirani", + "year": "2005", + "rating": "7.9", + "duration": "149", + "genre": "冒险", + "actors_count": "2" + }, + { + "id": "9", + "title": "肖申克的救赎", + "director": "Christophe Barratier", + "year": "2008", + "rating": "9.3", + "duration": "91", + "genre": "冒险", + "actors_count": "2" + }, + { + "id": "10", + "title": "盗梦空间", + "director": "Christopher Nolan", + "year": "2019", + "rating": "7.1", + "duration": "132", + "genre": "剧情", + "actors_count": "5" + } +] \ No newline at end of file diff --git a/q2_1_crawler/q2_1.py b/q2_1_crawler/q2_1.py new file mode 100644 index 0000000..2d782cf --- /dev/null +++ b/q2_1_crawler/q2_1.py @@ -0,0 +1,65 @@ +import requests +from bs4 import BeautifulSoup as bs +import json + +url = 'https://exam.detr.top/exam-b/movies' +headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/149.0.0.0 Safari/537.36 Edg/149.0.0.0', + 'Referer':'https://exam.detr.top/exam-b/movies'} +req = requests.get(url, headers=headers) +req.encoding="utf-8" + +data=[] + +soup=bs(req.text,"html.parser") +# print(soup) +#id, title, director, year, rating, duration, genre, actors_count + +item=soup.select("table tbody tr" ) +movie_list=[] + +for tr in item: + tds=tr.find_all("td") + tds=list(tds) + # print(tds) + if len(tds)<8: + continue + movie={ + "id":tds[0].get_text(strip=True), + "title":tds[1].get_text(strip=True), + "director":tds[2].get_text(strip=True), + "year":tds[3].get_text(strip=True), + "rating":tds[4].get_text(strip=True), + "duration":tds[5].get_text(strip=True), + "genre":tds[6].get_text(strip=True), + "actors_count":tds[7].get_text(strip=True) + } + movie_list.append(movie) +print(movie_list) + + +with open('movie.json', 'w', encoding='utf-8') as f: + json.dump(movie_list, f, ensure_ascii=False, indent=2) + +with open("move.html","w",encoding='utf-8') as f: + json.dump(movie_list, f, ensure_ascii=False, indent=2) + + + + + +# for i in range(len(items)): +# rank=i+1 +# title=items[i].find("span",class_="title").get_text() +# actors=items[i].find("div",class_="bd").get_text().strip() +# try: +# actors=actors.split("主演：")[1].split("\n")[0] +# except: +# actors="无" +# quote=items[i].find("p",class_="quote").get_text().strip() + +# data.append({ +# "rank":rank, +# "title":title, +# "actors":actors, +# "quote":quote +# }) \ No newline at end of file diff --git a/q2_1_crawler/q2_2.py b/q2_1_crawler/q2_2.py new file mode 100644 index 0000000..f30c2f7 --- /dev/null +++ b/q2_1_crawler/q2_2.py @@ -0,0 +1,43 @@ +# ① 找出评分最高和最低的电影，打印电影名 + 评分。 +# ② 统计各类型的电影数量，用字典格式输出。 +# ③ 统计各导演的电影数量，用字典格式输出。 +# ④ 统计 2020 年（含）以后上映的电影数量。 + +import json + + + +with open('movie.json', 'r', encoding='utf-8') as f: + data=json.load(f) + # print(data) + + +sort_movie=sorted(data,key=lambda x:x["rating"]) +min=sort_movie[0] +max=sort_movie[-1] +print("评分最低的电影",min["title"],min["rating"]) +print("评分最高的电影",max["title"],max["rating"]) + +genre_shu={} +for g in data: + ge=g["genre"] + if ge in genre_shu: + genre_shu[ge]+=1 + else: + genre_shu[ge]=1 +print("各类型的电影数量",genre_shu) + +director_shu={} +for d in data: + di=d["director"] + if di in director_shu: + director_shu[di]+=1 + else: + director_shu[di]=1 +print("各导演的电影数量",director_shu) + +a=0 +for y in data: + if int(y["year"]) >= 2020: + a+=1 +print("2020 年（含）以后上映的电影数量",a) \ No newline at end of file diff --git a/q3/q3_1/q3_1_image_label.zip b/q3/q3_1/q3_1_image_label.zip new file mode 100644 index 0000000..5b08d84 Binary files /dev/null and b/q3/q3_1/q3_1_image_label.zip differ diff --git a/q3/q3_2/q3_2_takeout_reviews.json b/q3/q3_2/q3_2_takeout_reviews.json new file mode 100644 index 0000000..94c7925 --- /dev/null +++ b/q3/q3_2/q3_2_takeout_reviews.json @@ -0,0 +1 @@ +[{"id":32,"annotations":[{"id":20,"completed_by":1,"result":[{"value":{"choices":["正面"]},"id":"Onq13IcOtQ","from_name":"sentiment","to_name":"text","type":"choices","origin":"manual"}],"was_cancelled":false,"ground_truth":false,"created_at":"2026-06-25T07:10:52.845728Z","updated_at":"2026-06-25T07:10:52.845818Z","draft_created_at":"2026-06-25T07:10:22.361376Z","lead_time":12.684999999999999,"prediction":{},"result_count":1,"unique_id":"0dd85e72-c68d-4302-ab92-11c8114bd5bb","import_id":null,"last_action":null,"bulk_created":false,"task":32,"project":12,"updated_by":1,"parent_prediction":null,"parent_annotation":null,"last_created_by":null}],"file_upload":"9ce97d9b-reviews.json","drafts":[],"predictions":[],"data":{"id":1,"text":"外卖小哥送得超快，餐盒还是热的，炸鸡酥脆多汁，酸辣粉也很正宗，分量足，五星好评！"},"meta":{},"created_at":"2026-06-25T07:10:04.754168Z","updated_at":"2026-06-25T07:10:52.903174Z","allow_skip":true,"inner_id":1,"total_annotations":1,"cancelled_annotations":0,"total_predictions":0,"comment_count":0,"unresolved_comment_count":0,"last_comment_updated_at":null,"project":12,"updated_by":1,"comment_authors":[]},{"id":33,"annotations":[{"id":21,"completed_by":1,"result":[{"value":{"choices":["负面"]},"id":"cJxGS89PGV","from_name":"sentiment","to_name":"text","type":"choices","origin":"manual"}],"was_cancelled":false,"ground_truth":false,"created_at":"2026-06-25T07:10:56.177683Z","updated_at":"2026-06-25T07:10:56.177702Z","draft_created_at":"2026-06-25T07:10:40.725815Z","lead_time":4.317,"prediction":{},"result_count":1,"unique_id":"ec19a0ee-0062-4fcb-be7e-3885bbb0e630","import_id":null,"last_action":null,"bulk_created":false,"task":33,"project":12,"updated_by":1,"parent_prediction":null,"parent_annotation":null,"last_created_by":null}],"file_upload":"9ce97d9b-reviews.json","drafts":[],"predictions":[],"data":{"id":2,"text":"等了一个半小时才送到，汤全洒了，面坨成一坨，联系客服也不回，太让人失望了。"},"meta":{},"created_at":"2026-06-25T07:10:04.754256Z","updated_at":"2026-06-25T07:10:56.235324Z","allow_skip":true,"inner_id":2,"total_annotations":1,"cancelled_annotations":0,"total_predictions":0,"comment_count":0,"unresolved_comment_count":0,"last_comment_updated_at":null,"project":12,"updated_by":1,"comment_authors":[]},{"id":34,"annotations":[{"id":22,"completed_by":1,"result":[{"value":{"choices":["正面"]},"id":"TEYHnwsP5p","from_name":"sentiment","to_name":"text","type":"choices","origin":"manual"}],"was_cancelled":false,"ground_truth":false,"created_at":"2026-06-25T07:11:02.709432Z","updated_at":"2026-06-25T07:11:02.709456Z","draft_created_at":"2026-06-25T07:10:44.119328Z","lead_time":5.162,"prediction":{},"result_count":1,"unique_id":"e99cc8d6-b0ec-4ba7-95c9-75e600010972","import_id":null,"last_action":null,"bulk_created":false,"task":34,"project":12,"updated_by":1,"parent_prediction":null,"parent_annotation":null,"last_created_by":null}],"file_upload":"9ce97d9b-reviews.json","drafts":[],"predictions":[],"data":{"id":3,"text":"奶茶是用料很扎实的现煮茶，珍珠Q弹有嚼劲，配送员态度也好，下次还会再点。"},"meta":{},"created_at":"2026-06-25T07:10:04.754319Z","updated_at":"2026-06-25T07:11:02.772529Z","allow_skip":true,"inner_id":3,"total_annotations":1,"cancelled_annotations":0,"total_predictions":0,"comment_count":0,"unresolved_comment_count":0,"last_comment_updated_at":null,"project":12,"updated_by":1,"comment_authors":[]},{"id":35,"annotations":[{"id":23,"completed_by":1,"result":[{"value":{"choices":["正面"]},"id":"WSCVtAKzJa","from_name":"sentiment","to_name":"text","type":"choices","origin":"manual"}],"was_cancelled":false,"ground_truth":false,"created_at":"2026-06-25T07:11:05.544799Z","updated_at":"2026-06-25T07:11:05.544821Z","draft_created_at":"2026-06-25T07:10:46.400039Z","lead_time":3.517,"prediction":{},"result_count":1,"unique_id":"b72bb1ff-9547-471d-8edd-58f5277dca12","import_id":null,"last_action":null,"bulk_created":false,"task":35,"project":12,"updated_by":1,"parent_prediction":null,"parent_annotation":null,"last_created_by":null}],"file_upload":"9ce97d9b-reviews.json","drafts":[],"predictions":[],"data":{"id":4,"text":"配送速度一般，但披萨味道不错，芝士拉丝效果好，性价比高，值得推荐。"},"meta":{},"created_at":"2026-06-25T07:10:04.754376Z","updated_at":"2026-06-25T07:11:05.601129Z","allow_skip":true,"inner_id":4,"total_annotations":1,"cancelled_annotations":0,"total_predictions":0,"comment_count":0,"unresolved_comment_count":0,"last_comment_updated_at":null,"project":12,"updated_by":1,"comment_authors":[]},{"id":36,"annotations":[{"id":24,"completed_by":1,"result":[{"value":{"choices":["负面"]},"id":"qLglBmVk44","from_name":"sentiment","to_name":"text","type":"choices","origin":"manual"}],"was_cancelled":false,"ground_truth":false,"created_at":"2026-06-25T07:11:08.643150Z","updated_at":"2026-06-25T07:11:08.643172Z","draft_created_at":"2026-06-25T07:10:48.222193Z","lead_time":2.6100000000000003,"prediction":{},"result_count":1,"unique_id":"7e05411c-9818-4359-872a-ed1e588d0687","import_id":null,"last_action":null,"bulk_created":false,"task":36,"project":12,"updated_by":1,"parent_prediction":null,"parent_annotation":null,"last_created_by":null}],"file_upload":"9ce97d9b-reviews.json","drafts":[],"predictions":[],"data":{"id":5,"text":"点的麻辣烫食材不新鲜，有股怪味，吃完拉肚子，商家推卸责任，再也不点了。"},"meta":{},"created_at":"2026-06-25T07:10:04.754432Z","updated_at":"2026-06-25T07:11:08.699069Z","allow_skip":true,"inner_id":5,"total_annotations":1,"cancelled_annotations":0,"total_predictions":0,"comment_count":0,"unresolved_comment_count":0,"last_comment_updated_at":null,"project":12,"updated_by":1,"comment_authors":[]}] \ No newline at end of file diff --git a/q3/q3_3_质量自评.md b/q3/q3_3_质量自评.md new file mode 100644 index 0000000..f8dde57 --- /dev/null +++ b/q3/q3_3_质量自评.md @@ -0,0 +1,13 @@ +1.标注前准备： +在图片标注中 +（1）边界框必须紧贴目标物轮廓 +（2）标签必须为 cat、dog、car（必须小写英文，不能写成"猫/狗/车"或"Cat/Dog/Car"） +在文本标注中 +（1）标注必须包含每条评论的 id 和 text 字段 +（2）每个标注必须有一个 sentiment 字段，值为 "正面" 或 "负面" +2.标注过程 +在图像的标注中，边界框内的留白较多，解决方法：最大限度地贴近目标轮廓 +在文本的标注中，可能会遇到情感模糊的问题，解决办法：抓住关键词进行标注。 +3.标注后检查 +在图像标注中，再次检查边界框与目标轮廓是否贴合，检查导出的 yolo 文件内容是否完整 +在文本标注中，对照关键词检查情感是否判断正确，检查导出的 json文件内容是否完整 \ No newline at end of file diff --git a/q4/q4_1/q4_1.py b/q4/q4_1/q4_1.py new file mode 100644 index 0000000..b7526d6 --- /dev/null +++ b/q4/q4_1/q4_1.py @@ -0,0 +1,33 @@ +import matplotlib.pyplot as plt +import json + + +with open('movie.json', 'r', encoding='utf-8') as f: + data=json.load(f) + # print(data) + +genre_shu={} +for g in data: + ge=g["genre"] + if ge in genre_shu: + genre_shu[ge]+=1 + else: + genre_shu[ge]=1 +# print("各类型的电影数量",genre_shu) + +genre_lei=list(genre_shu.keys()) +genre_liang=list(genre_shu.values()) + +print(genre_liang) + +plt.figure(figsize=(14, 12)) +plt.bar(genre_lei, genre_liang, # 类别, 数值 + width=0.6) # 柱子宽度（0~1之间） + +# 标题和标签 +plt.title('类型电影数量分布', fontsize=14) +plt.xlabel('类型名称', fontsize=12) +plt.ylabel('电影数量', fontsize=12) + +plt.show() + diff --git a/q4/q4_1/q4_1_bar.png b/q4/q4_1/q4_1_bar.png new file mode 100644 index 0000000..0ed7e05 Binary files /dev/null and b/q4/q4_1/q4_1_bar.png differ diff --git a/q4/q4_2/q4_2.py b/q4/q4_2/q4_2.py new file mode 100644 index 0000000..a3e4b3d --- /dev/null +++ b/q4/q4_2/q4_2.py @@ -0,0 +1,24 @@ +import matplotlib.pyplot as plt +import json + +rating=[] +duration=[] + +with open('movie.json', 'r', encoding='utf-8') as f: + data=json.load(f) + # print(data) + for i in data: + rating.append(i["rating"]) + duration.append(i["duration"]) +plt.figure(figsize=(12, 8)) +plt.scatter(duration, rating, + c='red', + s=80, # 点的大小 + alpha=0.6, # 透明度 + edgecolors='white') # 点的边框 +plt.title('时长与评分关系散点图', fontsize=14) +plt.xlabel('时长', fontsize=12) +plt.ylabel('评分', fontsize=12) +plt.grid(True, linestyle='--', alpha=0.5) +plt.show() + diff --git a/q4/q4_2/q4_2_scatter.png b/q4/q4_2/q4_2_scatter.png new file mode 100644 index 0000000..5dbac63 Binary files /dev/null and b/q4/q4_2/q4_2_scatter.png differ diff --git a/q4/q4_3a/q4_3a.py b/q4/q4_3a/q4_3a.py new file mode 100644 index 0000000..bc84d8b --- /dev/null +++ b/q4/q4_3a/q4_3a.py @@ -0,0 +1,19 @@ +import matplotlib.pyplot as plt +import json +rating=[] + +with open('movie.json', 'r', encoding='utf-8') as f: + data=json.load(f) + # print(data) + for i in data: + rating.append(i["rating"]) + +plt.figure(figsize=(12,8)) +plt.hist(rating, # 数据 + bins=3, # 分成几个柱子 + color='#3498DB', # 颜色 + edgecolor='white') # 柱子边框颜色 +plt.title('评分分布', fontsize=14) +plt.xlabel('评分', fontsize=13) +plt.grid(True, linestyle='--', alpha=0.5, axis='y') +plt.show() \ No newline at end of file diff --git a/q4/q4_3a/q4_3a_hist.png b/q4/q4_3a/q4_3a_hist.png new file mode 100644 index 0000000..6c68736 Binary files /dev/null and b/q4/q4_3a/q4_3a_hist.png differ diff --git a/q4/q4_3b/q4_3b.py b/q4/q4_3b/q4_3b.py new file mode 100644 index 0000000..38865a2 --- /dev/null +++ b/q4/q4_3b/q4_3b.py @@ -0,0 +1,19 @@ +import matplotlib.pyplot as plt +import json +duration=[] + +with open('movie.json', 'r', encoding='utf-8') as f: + data=json.load(f) + # print(data) + for i in data: + duration.append(i["duration"]) + +plt.figure(figsize=(12,8)) +plt.hist( duration, # 数据 + bins=3, # 分成几个柱子 + color='#3498DB', # 颜色 + edgecolor='white') # 柱子边框颜色 +plt.title('时长分布', fontsize=14) +plt.xlabel('时长（分钟）', fontsize=13) +plt.grid(True, linestyle='--', alpha=0.5, axis='y') +plt.show() \ No newline at end of file diff --git a/q4/q4_3b/q4_3b_hist.png b/q4/q4_3b/q4_3b_hist.png new file mode 100644 index 0000000..b142813 Binary files /dev/null and b/q4/q4_3b/q4_3b_hist.png differ