时间戳

首页 > 解决方案 > 如何将数据框的字符串值映射到某个数字以绘制集群？

python - 如何将数据框的字符串值映射到某个数字以绘制集群？

问题描述

我想通过我知道的唯一方法将字符串映射到某个整数值来绘制我的数据集的集群。像

data_mapped=data.copy()
data_mapped['Language']=data_mapped['Language'].map({'English':0,'French':1,'German':2})
data_mapped

但在这个例子中，我只有 3 个可以通过这种方法映射的唯一语言值

现在我不知道如何将多个唯一字符串值转换为整数值并绘制集群？我想按 (Color,fabric,dress_type) 之类的列进行聚类，并且我想对整个数据进行聚类？

现在我的数据集是：

file_list=glob.glob('json_file/[!Merg_all]*json')
merg_all_list=[]
for file in file_list:
    print(file)
    raw_data=pd.read_json(str(file))
    raw_data.head()
    for i in raw_data['product']:
        merg_all_list.append(i)

.json 文件：

[{"product": {"brand_name": "So Kamal", "designer": "So Kamal", "title": "So Kamal Women Summer Collection Mustard Lawn 1PC -Unstitched Shirt DPL19 49 LA00964-Std-MST", "description": "description   specifications of so kamal women summer collection mustard lawn 1pc  unstitched shirt dpl19 49 la00964 std mst brand so kamal sku 105972128_pk 1253666066 features 1pc  unstitched main material lawn season summer material family lawn what's in the box 1x 1pc unstitched suit", "dress_type": "shirt", "where_to_wear": "", "color": "mustard", "stitched": false, "season": "summer", "price": 1120, "currency": "Rs", "product_id": "So Kamal Women Summer Collection Mustard Lawn 1PC -Unstitched Shirt DPL19 49 LA00964-Std-MST", "collection_url": "https://lawncollection.pk/brands/", "source": "https://lawncollection.pk/so-kamal-women-summer-collection-mustard-lawn-1pc-unstitched-shirt-dpl19-49-la00964-std-mst.html", "fabric": "lawn", "gender": "women", "frontpic": "https://lawncollection.pk/public/images/products//2019/04/so-kamal-women-summer-collection-mustard-lawn-1pc-unstitched-shirt-dpl19-49-la00964-std-mst-image1.jpeg", "backpic": "https://lawncollection.pk/public/images/products//2019/04/so-kamal-women-summer-collection-mustard-lawn-1pc-unstitched-shirt-dpl19-49-la00964-std-mst-image2.jpeg", "otherpics": ["https://lawncollection.pk/public/images/products//2019/04/so-kamal-women-summer-collection-mustard-lawn-1pc-unstitched-shirt-dpl19-49-la00964-std-mst-image1.jpeg", "https://lawncollection.pk/public/images/products//2019/04/so-kamal-women-summer-collection-mustard-lawn-1pc-unstitched-shirt-dpl19-49-la00964-std-mst-image2.jpeg"], "sku": "SKU: 105972128_PK-1253666066", "details": "https://lawncollection.pk/so-kamal-women-summer-collection-mustard-lawn-1pc-unstitched-shirt-dpl19-49-la00964-std-mst.html https:  lawncollection.pk so kamal women summer collection mustard lawn 1pc unstitched shirt dpl19 49 la00964 std mst.html so kamal so kamal women summer collection mustard lawn 1pc -unstitched shirt dpl19 49 la00964-std-mst description   specifications of so kamal women summer collection mustard lawn 1pc  unstitched shirt dpl19 49 la00964 std mst brand so kamal sku 105972128_pk 1253666066 features 1pc  unstitched main material lawn season summer material family lawn what's in the box 1x 1pc unstitched suit", "Category1_list": "unstitched", "size": {"xs": false, "s": false, "m": false, "xl": false, "xxl": false}}}]

数据框

    brand_name  designer    title   description dress_type  where_to_wear   color   stitched    season  price   ... source  fabric  gender  frontpic    backpic otherpics   details Category1_list  size    sku
0   Polo Ralph Lauren   Polo Ralph Lauren   Long Sleeve Knit Magic Fleece Sweatshirt    - Casual graphic print sweatshirt- Crew neckli...   sweatshirt      black   True        8544    ... https://www.zalora.com.ph/polo-ralph-lauren-lo...   cotton  man static.ph.zalora.net/p/polo-ralph-lauren-3175-...   static.ph.zalora.net/p/polo-ralph-lauren-3175-...   [static.ph.zalora.net/p/polo-ralph-lauren-3175...   https://www.zalora.com.ph/polo-ralph-lauren-lo...       {'xs': False, 's': False, 'm': False, 'xl': Fa...   NaN
1   Polo Ralph Lauren   Polo Ralph Lauren   Basic Mesh Polo Shirt   - Colour block polo shirt with brand print- Un...   shirt       red True        9265    ... https://www.zalora.com.ph/polo-ralph-lauren-ba...   cotton  man static.ph.zalora.net/p/polo-ralph-lauren-7554-...   static.ph.zalora.net/p/polo-ralph-lauren-7555-...   [static.ph.zalora.net/p/polo-ralph-lauren-7554...   https://www.zalora.com.ph/polo-ralph-lauren-ba...       {'xs': False, 's': False, 'm': False, 'xl': Fa...   NaN
2   MANGO Man   MANGO Man   Faux Shearling Denim Jacket - Denim jacket with wash detail- Collar neckli...   jacket      blue    True        4995    ... https://www.zalora.com.ph/mango-man-faux-shear...   denim   man static.ph.zalora.net/p/mango-man-9782-7201341-...   static.ph.zalora.net/p/mango-man-9783-7201341-...   [static.ph.zalora.net/p/mango-man-9782-7201341...   https://www.zalora.com.ph/mango-man-faux-shear...       {'xs': False, 's': False, 'm': False, 'xl': Fa...   NaN
3   Polo Ralph Lauren   Polo Ralph Lauren   Knit Magic Fleece Hoodie    - Embroidered front hoodie- Unlined- Hooded ne...               True        10598   ... https://www.zalora.com.ph/polo-ralph-lauren-kn...   cotton  man static.ph.zalora.net/p/polo-ralph-lauren-2320-...   static.ph.zalora.net/p/polo-ralph-lauren-2320-...   [static.ph.zalora.net/p/polo-ralph-lauren-2320...   https://www.zalora.com.ph/polo-ralph-lauren-kn...       {'xs': False, 's': True, 'm': True, 'xl': True...   NaN
4   MANGO Man   MANGO Man   Turtleneck Flecked Sweater  - Solid hue speckle-knit sweatshirt- High neck...   sweatshirt      brown   True        2995    ... https://www.zalora.com.ph/mango-man-turtleneck...   cotton  man static.ph.zalora.net/p/mango-man-1900-5990341-...   static.ph.zalora.net/p/mango-man-1900-5990341-...   [static.ph.zalora.net/p/mango-man-1900-5990341...   https://www.zalora.com.ph/mango-man-turtleneck...       {'xs': False, 's': False, 'm': False, 'xl': Fa...   NaN

标签： pythonpandasdataframecluster-analysisk-means

解决方案

为您的数据选择适当的可视化技术。

在分类数据上，条形图比散点图更合适，因为您不需要 x 轴是数字。

还要选择适当的算法... K-means 仅对连续变量有意义。将类别编码为 k-means 的整数是错误的。在您的情况下，k-means 会假设英语和德语的平均值恰好是法语。

推荐阅读