{"id":1204,"date":"2024-04-26T05:32:55","date_gmt":"2024-04-26T05:32:55","guid":{"rendered":"https:\/\/www.nicekj.com\/?p=1204"},"modified":"2024-04-26T05:33:16","modified_gmt":"2024-04-26T05:33:16","slug":"openclip","status":"publish","type":"post","link":"https:\/\/www.nicekj.com\/openclip.html","title":{"rendered":"OpenClip"},"content":{"rendered":"

\u8bde\u751f<\/h2>\n

Transformer\u67b6\u6784\u7684\u6a21\u578b\u5728\u8fc7\u53bb\u51e0\u5e74\u91cc\u9010\u6e10\u6210\u4e3a\u4e86\u56fe\u50cf\u9886\u57df\u7684\u4e00\u4e2a\u4e3b\u6d41\u7814\u7a76\u65b9\u5411\u3002\u53d1\u8868\u4e86GPT\u548cWhisper\u7684OpenAI\u4e5f\u4e0d\u7518\u843d\u540e\uff0c\u5c31\u62ff4\u4ebf\u5f20\u4e92\u8054\u7f51\u4e0a\u627e\u5230\u7684\u56fe\u7247\uff0c\u4ee5\u53ca\u56fe\u7247\u5bf9\u5e94\u7684ALT\uff08Alternative\uff09\u6587\u5b57\u8bad\u7ec3\u4e86\u4e00\u4e2a\u53eb\u505aCLIP\uff08Contrastive Language-Image Pretraining\uff09\u7684\u591a\u6a21\u6001\u6a21\u578b\u3002\u8fd9\u4e2a\u6a21\u578b\uff0c\u4e0d\u4ec5\u53ef\u4ee5\u62ff\u6765\u505a\u5e38\u89c1\u7684\u56fe\u7247\u5206\u7c7b\u3001\u76ee\u6807\u68c0\u6d4b\uff0c\u4e5f\u53ef\u80fd\u591f\u7528\u6765\u4f18\u5316\u4e1a\u52a1\u573a\u666f\u91cc\u9762\u7684\u5546\u54c1\u641c\u7d22\u548c\u5185\u5bb9\u63a8\u8350\u3002<\/p>\n

\u591a\u6a21\u6001\u7684CLIP<\/h2>\n

\u6240\u8c13\u201c\u591a\u6a21\u6001\u201d\uff0c\u5c31\u662f\u591a\u79cd\u5a92\u4f53\u5f62\u5f0f\u7684\u5185\u5bb9\u3002\u6211\u4eec\u53ef\u4ee5\u62ffGPT\u6a21\u578b\u6765\u505a\u6570\u5b66\u8bd5\u9898\uff0c\u90a3\u4e48\u5982\u679c\u9047\u5230\u4e00\u4e2a\u5e73\u9762\u51e0\u4f55\u9898\u7684\u8bdd\uff0c\u5149\u6709\u9898\u76ee\u7684\u6587\u5b57\u4fe1\u606f\u662f\u4e0d\u591f\u7684\uff0c\u8fd8\u9700\u8981\u628a\u5bf9\u5e94\u7684\u56fe\u5f62\u4e00\u5e76\u63d0\u4f9b\u7ed9AI\u624d\u53ef\u4ee5\u3002\u8fd9\u4e5f\u662f\u901a\u5f80\u901a\u7528\u4eba\u5de5\u667a\u80fd\u7684\u5fc5\u7ecf\u4e4b\u8def\uff0c\u56e0\u4e3a\u771f\u5b9e\u7684\u4e16\u754c\u5c31\u662f\u4e00\u4e2a\u591a\u6a21\u6001\u7684\u4e16\u754c\u3002<\/p>\n

CLIP\u6a21\u578b\uff0c\u5b83\u4e0d\u4ec5\u80fd\u591f\u5206\u522b\u7406\u89e3\u56fe\u7247\u548c\u6587\u672c\uff0c\u8fd8\u901a\u8fc7\u5bf9\u6bd4\u5b66\u4e60\u5efa\u7acb\u4e86\u56fe\u7247\u548c\u6587\u672c\u4e4b\u95f4\u7684\u5173\u7cfb\u3002\u8fd9\u4e2a\u4e5f\u662f\u672a\u6765\u6211\u4eec\u80fd\u591f\u901a\u8fc7\u5199\u51e0\u4e2a\u63d0\u793a\u8bcd\u5c31\u80fd\u7528AI\u753b\u56fe\u7684\u4e00\u4e2a\u8d77\u70b9\u3002<\/p>\n

\n

\"image.png\"<\/figure>\n<\/p>\n

CLIP\u7684\u601d\u8def\u5e76\u4e0d\u590d\u6742\uff0c\u5c31\u662f\u5229\u7528\u4e92\u8054\u7f51\u4e0a\u5df2\u6709\u7684\u5927\u91cf\u516c\u5f00\u7684\u56fe\u7247\u6570\u636e\u3002\u5176\u4e2d\u6709\u5f88\u591a\u5df2\u7ecf\u901a\u8fc7HTML\u6807\u7b7e\u91cc\u9762\u7684title\u6216\u8005alt\u5b57\u6bb5\uff0c\u63d0\u4f9b\u4e86\u5bf9\u56fe\u7247\u7684\u6587\u672c\u63cf\u8ff0\u3002\u90a3\u6211\u4eec\u53ea\u8981\u8bad\u7ec3\u4e00\u4e2a\u6a21\u578b\uff0c\u5c06\u6587\u672c\u8f6c\u6362\u6210\u4e00\u4e2a\u5411\u91cf\uff0c\u4e5f\u5c06\u56fe\u7247\u8f6c\u6362\u6210\u4e00\u4e2a\u5411\u91cf\u3002\u56fe\u7247\u5411\u91cf\u5e94\u8be5\u548c\u81ea\u5df1\u7684\u6587\u672c\u63cf\u8ff0\u5411\u91cf\u7684\u8ddd\u79bb\u5c3d\u91cf\u8fd1\uff0c\u548c\u5176\u4ed6\u7684\u6587\u672c\u5411\u91cf\u8981\u5c3d\u91cf\u8fdc\u3002\u90a3\u4e48\u8fd9\u4e2a\u6a21\u578b\uff0c\u5c31\u80fd\u591f\u628a\u56fe\u7247\u548c\u6587\u672c\u6620\u5c04\u5230\u540c\u4e00\u4e2a\u7a7a\u95f4\u91cc\u3002\u6211\u4eec\u5c31\u80fd\u591f\u901a\u8fc7\u5411\u91cf\u540c\u65f6\u7406\u89e3\u56fe\u7247\u548c\u6587\u672c\u4e86\u3002<\/p>\n

\u56fe\u7247\u7684\u96f6\u6837\u672c\u5206\u7c7b<\/h2>\n
<\/div>
js<\/span>
\u590d\u5236\u4ee3\u7801<\/div><\/div><\/div><\/span>\nimport<\/span> torch<\/span>\nfrom<\/span> PIL<\/span> import<\/span> Image<\/span><\/span>\nfrom<\/span> IPython<\/span>.display<\/span> import<\/span> display<\/span>\nfrom<\/span> IPython<\/span>.display<\/span> import<\/span> Image<\/span> as<\/span> IPyImage<\/span><\/span>\nfrom<\/span> transformers import<\/span> CLIPProcessor<\/span>, CLIPModel<\/span><\/span>\n<\/span>\nmodel = CLIPModel<\/span>.from_pretrained<\/span>(\"openai\/clip-vit-base-patch32\"<\/span>)<\/span>\nprocessor = CLIPProcessor<\/span>.from_pretrained<\/span>(\"openai\/clip-vit-base-patch32\"<\/span>)<\/span>\n<\/span>\ndef get_image_feature<\/span>(filename<\/span>: str):<\/span>\n    image = Image<\/span>.open<\/span>(filename).convert<\/span>(\"RGB\"<\/span>)<\/span>\n    processed = processor<\/span>(images=image, return_tensors=\"pt\"<\/span>, padding=True<\/span>, truncation=True<\/span>)<\/span>\n    with<\/span> torch.no_grad<\/span>():<\/span>\n        image_features = model.get_image_features<\/span>(pixel_values=processed[\"pixel_values\"<\/span>])<\/span>\n    return<\/span> image_features<\/span>\n<\/span>\ndef get_text_feature<\/span>(text<\/span>: str):<\/span>\n    processed = processor<\/span>(text=text, return_tensors=\"pt\"<\/span>, padding=True<\/span>, truncation=True<\/span>)<\/span>\n    with<\/span> torch.no_grad<\/span>():<\/span>\n        text_features = model.get_text_features<\/span>(processed['input_ids'<\/span>])<\/span>\n    return<\/span> text_features<\/span>\n<\/span>\ndef cosine_similarity<\/span>(tensor1, tensor2):<\/span>\n    tensor1_normalized = tensor1 \/ tensor1.norm<\/span>(dim=-1<\/span>, keepdim=True<\/span>)<\/span>\n    tensor2_normalized = tensor2 \/ tensor2.norm<\/span>(dim=-1<\/span>, keepdim=True<\/span>)<\/span>\n    return<\/span> (tensor1_normalized * tensor2_normalized).sum<\/span>(dim=-1<\/span>)<\/span>\n<\/span>\nimage_tensor = get_image_feature<\/span>(\".\/data\/cat.jpg\"<\/span>)<\/span>\n<\/span>\ncat_text = \"This is a cat.\"<\/span><\/span>\ncat_text_tensor = get_text_feature<\/span>(cat_text)<\/span>\n<\/span>\ndog_text = \"This is a dog.\"<\/span><\/span>\ndog_text_tensor = get_text_feature<\/span>(dog_text)<\/span>\n<\/span>\ntwo_cats_text = \"There are two cats.\"<\/span><\/span>\ntwo_cats_text_tensor = get_text_feature<\/span>(two_cats_text)<\/span>\n<\/span>\ntruck_text = \"This is a truck.\"<\/span><\/span>\ntruck_text_tensor = get_text_feature<\/span>(truck_text)<\/span>\n<\/span>\ncouch_text = \"This is a couch.\"<\/span><\/span>\ncouch_text_tensor = get_text_feature<\/span>(couch_text)<\/span>\n<\/span>\ndisplay<\/span>(IPyImage<\/span>(filename='.\/cat.jpg'<\/span>))<\/span>\n<\/span>\nprint<\/span>(\"Similarity with cat : \"<\/span>, cosine_similarity<\/span>(image_tensor, cat_text_tensor))<\/span>\nprint<\/span>(\"Similarity with dog : \"<\/span>, cosine_similarity<\/span>(image_tensor, dog_text_tensor))<\/span>\nprint<\/span>(\"Similarity with two cats : \"<\/span>, cosine_similarity<\/span>(image_tensor, two_cats_text_tensor))<\/span>\nprint<\/span>(\"Similarity with truck : \"<\/span>, cosine_similarity<\/span>(image_tensor, truck_text_tensor))<\/span>\n<\/span>\n<\/code><\/pre>\n
\u8f93\u51fa\u7ed3\u679c\uff1a<\/p>\n
\n
\"image.png\"<\/figure>\n<\/p>\n
\u4ee3\u7801\u89e3\u91ca\uff1a<\/p>\n
    \n
  1. \u901a\u8fc7Transformers\u5e93\u7684CLIPModel\u3001CLIPProcessor\uff0c\u52a0\u8f7d\u4e86clip-vit-base-patch32\u8fd9\u4e2a\u6a21\u578b\uff0c\u7528\u6765\u5904\u7406\u6211\u4eec\u7684\u56fe\u7247\u548c\u6587\u672c\u4fe1\u606f\u3002<\/li>\n
  2. get_image_features\u65b9\u6cd5\u91cc\uff0c\u6240\u505a\u4e8b\u60c5\uff1a\n
      \n
    • \u901a\u8fc7CLIPProcessor\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406\uff0c\u53d8\u6210\u4e00\u7cfb\u5217\u7684\u6570\u503c\u7279\u5f81\u8868\u793a\u7684\u5411\u91cf\u3002\u9884\u5904\u7406\u8fc7\u7a0b\uff0c\u5c31\u662f\u628a\u539f\u59cb\u7684\u56fe\u7247\uff0c\u53d8\u6210\u4e00\u4e2a\u4e2a\u50cf\u7d20\u7684RGB\u503c\uff1b\u7136\u540e\u7edf\u4e00\u56fe\u7247\u7684\u5c3a\u5bf8\uff0c\u4ee5\u53ca\u5bf9\u4e8e\u4e0d\u89c4\u5219\u7684\u56fe\u7247\u622a\u53d6\u4e2d\u95f4\u6b63\u65b9\u5f62\u7684\u90e8\u5206\uff0c\u6700\u540e\u8fdb\u884c\u6570\u503c\u7684\u5f52\u4e00\u5316\u3002<\/li>\n
    • \u901a\u8fc7CLIPModel\uff0c\u628a\u4e0a\u9762\u7684\u6570\u503c\u5411\u91cf\uff0c\u63a8\u65ad\u6210\u4e00\u4e2a\u8868\u8fbe\u4e86\u56fe\u7247\u542b\u4e49\u7684\u5f20\u91cf\u3002<\/li>\n<\/ul>\n<\/li>\n
    • get_text_features\u7c7b\u4f3c\uff0c\u5148\u628a\u6587\u672c\u901a\u8fc7CLIPProcessor\u8f6c\u6362\u6210Token\uff0c\u7136\u540e\u518d\u901a\u8fc7\u6a21\u578b\u63a8\u65ad\u51fa\u6587\u672c\u7684\u5f20\u91cf\u3002<\/li>\n
    • \u5b9a\u4e49\u4e86\u4e00\u4e2acosine_similarity\u51fd\u6570\uff0c\u7528\u6765\u8ba1\u7b97\u4e24\u4e2a\u5f20\u91cf\u4e4b\u95f4\u7684\u4f59\u5f26\u76f8\u4f3c\u5ea6\u3002<\/li>\n
    • \u5c31\u662f\u5229\u7528\u4e0a\u9762\u7684\u8fd9\u4e9b\u51fd\u6570\uff0c\u6765\u8ba1\u7b97\u56fe\u7247\u548c\u6587\u672c\u4e4b\u95f4\u7684\u76f8\u4f3c\u5ea6\u3002<\/li>\n<\/ol>\n
      \u4ece\u8f93\u51fa\u7ed3\u679c\u53ef\u4ee5\u770b\u5230\uff0c\u201cThere are two cats.\u201d\u7684\u76f8\u4f3c\u5ea6\u6700\u9ad8\uff0c\u56e0\u4e3a\u56fe\u91cc\u6709\u6c99\u53d1\uff0c\u6240\u4ee5\u201cThis is a couch.\u201d\u7684\u76f8\u4f3c\u5ea6\u4e5f\u8981\u9ad8\u4e8e\u201cThis is a dog.\u201d\u3002\u800cDog\u548cCat\u76f8\u4f3c\uff0c\u6240\u4ee5\u76f8\u4f3c\u5ea6\u4e5f\u6bd4\u5b8c\u5168\u4e0d\u76f8\u5173\u7684Truck\u8981\u9ad8\u3002\u53ef\u4ee5\u5f97\u5230\uff0cCLIP\u6a21\u578b\u5bf9\u56fe\u7247\u548c\u6587\u672c\u7684\u8bed\u4e49\u7406\u89e3\u662f\u975e\u5e38\u51c6\u786e\u7684\u3002<\/p>\n
      \u901a\u8fc7CLIP\u8fdb\u884c\u76ee\u6807\u68c0\u6d4b<\/h2>\n
      \u9664\u4e86\u80fd\u591f\u5b9e\u73b0\u96f6\u6837\u672c\u7684\u56fe\u50cf\u5206\u7c7b\u4e4b\u5916\uff0c\u6211\u4eec\u4e5f\u53ef\u4ee5\u5c06\u5b83\u5e94\u7528\u5230\u96f6\u6837\u672c\u4e0b\u7684\u76ee\u6807\u68c0\u6d4b\u4e2d\u3002\u76ee\u6807\u68c0\u6d4b\u5176\u5b9e\u5c31\u662f\u662f\u5728\u56fe\u50cf\u4e2d\u6846\u51fa\u7279\u5b9a\u533a\u57df\uff0c\u7136\u540e\u5bf9\u8fd9\u4e2a\u533a\u57df\u5185\u7684\u56fe\u50cf\u5185\u5bb9\u8fdb\u884c\u5206\u7c7b\u3002\u56e0\u6b64\uff0c\u6211\u4eec\u540c\u6837\u53ef\u4ee5\u7528 CLIP \u6765\u5b9e\u73b0\u76ee\u6807\u68c0\u6d4b\u4efb\u52a1\u3002<\/p>\n
      <\/div>
      js<\/span>
      \u590d\u5236\u4ee3\u7801<\/div><\/div><\/div>from<\/span> transformers import<\/span> pipeline<\/span>\nimport<\/span> cv2<\/span>\nfrom<\/span> matplotlib import<\/span> pyplot as<\/span> plt<\/span>\n<\/span>\ndetector = pipeline<\/span>(model=\"google\/owlvit-base-patch32\"<\/span>, task=\"zero-shot-object-detection\"<\/span>)<\/span>\ndetected = detector<\/span>(<\/span>\n    \".\/cat.jpg\"<\/span>,<\/span>\n    candidate_labels=[\"cat\"<\/span>, \"dog\"<\/span>, \"truck\"<\/span>, \"couch\"<\/span>, \"remote\"<\/span>],<\/span>\n)<\/span>\n<\/span>\n# Read<\/span> the image<\/span>\nimage_path = \".\/cat.jpg\"<\/span><\/span>\nimage = cv2.imread<\/span>(image_path)<\/span>\n<\/span>\n# Convert<\/span> the image from<\/span> BGR<\/span> to RGB<\/span> format<\/span>\nimage = cv2.cvtColor<\/span>(image, cv2.COLOR_BGR2RGB<\/span>)<\/span>\n<\/span>\n# Draw<\/span> the bounding box and label for<\/span> each detected object<\/span>\nfor<\/span> detection in<\/span> detected<\/span>:<\/span>\n    box = detection['box'<\/span>]<\/span>\n    label = detection['label'<\/span>]<\/span>\n    score = detection['score'<\/span>]<\/span>\n    <\/span>\n    # Draw<\/span> the bounding box and label on the image<\/span>\n    xmin, ymin, xmax, ymax = box['xmin'<\/span>], box['ymin'<\/span>], box['xmax'<\/span>], box['ymax'<\/span>]<\/span>\n    cv2.rectangle<\/span>(image, (xmin, ymin), (xmax, ymax), (0<\/span>, 255<\/span>, 0<\/span>), 2<\/span>)<\/span>\n    cv2.putText<\/span>(image, f\"{label}: {score:.2f}\"<\/span>, (xmin, ymin - 10<\/span>), cv2.FONT_HERSHEY_SIMPLEX<\/span>, 0.5<\/span>, (0<\/span>, 255<\/span>, 0<\/span>), 2<\/span>)<\/span>\n<\/span>\n# Display<\/span> the image in<\/span> Jupyter<\/span> Notebook<\/span><\/span>\nplt.imshow<\/span>(image)<\/span>\nplt.axis<\/span>('off'<\/span>)<\/span>\nplt.show<\/span>()<\/span>\n<\/code><\/pre>\n
      \u8f93\u51fa\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n
      \n
      \"image.png\"<\/figure>\n<\/p>\n
      \u4ece\u6700\u540e\u7684\u8f93\u51fa\u7ed3\u679c\u6765\u770b\uff0c\u65e0\u8bba\u662f\u732b\u54aa\u3001\u9065\u63a7\u5668\u8fd8\u662f\u6c99\u53d1\uff0c\u90fd\u88ab\u51c6\u786e\u5730\u6846\u9009\u51fa\u6765\u4e86\u3002<\/p>\n
      \u603b\u7ed3<\/h2>\n
      \u901a\u8fc7 CLIP \u6a21\u578b\uff0c\u6211\u4eec\u53ef\u4ee5\u5bf9\u4efb\u610f\u7269\u54c1\u540d\u79f0\u8fdb\u884c\u96f6\u6837\u672c\u5206\u7c7b\u3002\u8fdb\u4e00\u6b65\u5730\uff0c\u6211\u4eec\u8fd8\u80fd\u8fdb\u884c\u96f6\u6837\u672c\u7684\u76ee\u6807\u68c0\u6d4b\u3002\u800c\u6587\u672c\u548c\u56fe\u7247\u5728\u540c\u4e00\u4e2a\u5411\u91cf\u7a7a\u95f4\u7684\u8fd9\u4e2a\u7279\u6027\uff0c\u4e5f\u80fd\u591f\u8ba9\u6211\u4eec\u76f4\u63a5\u5229\u7528\u8fd9\u4e2a\u6a21\u578b\u8fdb\u4e00\u6b65\u4f18\u5316\u6211\u4eec\u7684\u5546\u54c1\u641c\u7d22\u529f\u80fd\u3002\u6211\u4eec\u53ef\u4ee5\u62ff\u6587\u672c\u7684\u5411\u91cf\uff0c\u901a\u8fc7\u627e\u5230\u4f59\u5f26\u8ddd\u79bb\u6700\u8fd1\u7684\u5546\u54c1\u56fe\u7247\u6765\u4f18\u5316\u641c\u7d22\u7684\u53ec\u56de\u8fc7\u7a0b\u3002\u6211\u4eec\u4e5f\u80fd\u76f4\u63a5\u62ff\u56fe\u7247\u5411\u91cf\uff0c\u5b9e\u73b0\u4ee5\u56fe\u641c\u56fe\u8fd9\u6837\u7684\u529f\u80fd\u3002<\/p>\n
      CLIP \u8fd9\u6837\u7684\u591a\u6a21\u6001\u6a21\u578b\uff0c\u8fdb\u4e00\u6b65\u62d3\u5c55\u4e86\u6211\u4eec AI \u7684\u80fd\u529b\u3002\u6211\u4eec\u73b0\u5728\u5199\u51e0\u4e2a\u63d0\u793a\u8bed\uff0c\u5c31\u80fd\u8ba9 AI \u62e5\u6709\u7ed8\u753b\u7684\u80fd\u529b\uff0c\u8fd9\u4e00\u70b9\u4e5f\u53ef\u4ee5\u8ba4\u4e3a\u662f\u53d1\u7aef\u4e8e\u6b64\u7684\u3002<\/p>","protected":false},"excerpt":{"rendered":"
      \u8bde\u751f Transformer\u67b6\u6784\u7684\u6a21\u578b\u5728\u8fc7\u53bb\u51e0\u5e74\u91cc\u9010\u6e10\u6210\u4e3a\u4e86\u56fe\u50cf\u9886\u57df\u7684\u4e00\u4e2a\u4e3b\u6d41\u7814\u7a76\u65b9\u5411\u3002\u53d1\u8868\u4e86GPT\u548cWhisper\u7684OpenAI\u4e5f\u4e0d\u7518\u843d\u540e\uff0c\u5c31\u62ff4\u4ebf\u5f20\u4e92\u8054\u7f51\u4e0a\u627e\u5230\u7684\u56fe\u7247\uff0c\u4ee5\u53ca\u56fe\u7247\u5bf9\u5e94\u7684ALT\uff08Alt<\/p>\n","protected":false},"author":1,"featured_media":2424,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"rank_math_title":"\u56fe\u7247 | \u5411\u91cf | CLIP - \u5948\u65af\u79d1\u6280\u793e\u533a","rank_math_description":"\u5948\u65af\u79d1\u6280\u793e\u533a\u4e13\u6ce8\u4e8e\u56fe\u7247\u3001\u5411\u91cf\u3001CLIP\u3001import\u3001\u6a21\u578b\uff0c\u63d0\u4f9b\u4e13\u4e1a\u7684\u5206\u4eab\u548c\u8ba8\u8bba\u5e73\u53f0\u3002","rank_math_focus_keyword":"\u56fe\u7247,\u5411\u91cf,CLIP,import,\u6a21\u578b","views":"45","footnotes":""},"categories":[3],"tags":[128,129,136,126,127],"collection":[],"class_list":["post-1204","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-fenlei2","tag-128","tag-129","tag-136","tag-gpt","tag-ai"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/posts\/1204","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/comments?post=1204"}],"version-history":[{"count":0,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/posts\/1204\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/media\/2424"}],"wp:attachment":[{"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/media?parent=1204"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/categories?post=1204"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/tags?post=1204"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.nicekj.com\/nicekj2024\/wp\/v2\/collection?post=1204"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}