[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KFPVWRPSF3X6JN5DYVTYPG":3,"skill-same:01M6KFPVWRPSF3X6JN5DYVTYPG":53},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":26,"linkOnly":27,"markdownContent":28,"markdownBody":29,"frontMatter":30,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":34,"repositoryHtmlUrl":35,"repositoryHomepageUrl":36,"ownerHtmlUrl":37,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":38,"install":39,"resources":52},"01M6KFPVWRPSF3X6JN5DYVTYPG","run2_dbscan-clustering",null,"Efficient DBSCAN clustering with a custom parameterized distance metric using precomputed distance matrices for speed, plus centroid computation.","cxcscmu\u002FSkillLearnBench","cxcscmu","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F142838272?v=4","01M6KCQDFQWPDSN0MKZY96MSQP","01M6KCQDM7G70TKJ28NTSQC0RZ",93,"MIT","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fb2-self-feedback-claude-sonnet-4-6\u002Fdbscan-parameter-tuning\u002Frun2_dbscan-clustering\u002FSKILL.md",[],[18,19,20,21,22,23,24,25],"clustering","run2 dbscan clustering","speed","matrices","with","for","metric","custom","2026-11-10 10:12:30",false,"---\nname: run2_dbscan-clustering\ndescription: Efficient DBSCAN clustering with a custom parameterized distance metric using precomputed distance matrices for speed, plus centroid computation.\n---\n\n# DBSCAN with Custom Distance Metric (Efficient)\n\n## Setup\n\n```python\nfrom sklearn.cluster import DBSCAN\nimport numpy as np\nfrom scipy.spatial.distance import pdist, squareform\n```\n\n## Custom Weighted Distance Metric\n\nFor Mars cloud clustering:\n```\nd(a, b) = sqrt((w * Δx)² + ((2 - w) * Δy)²)\n```\n\nWhen w=1, equals standard Euclidean. w>1 attenuates y-distances; w\u003C1 attenuates x-distances.\n\n## Performance: Precomputed Distance Matrix\n\nUsing a Python callable as metric in DBSCAN is **very slow** (O(N²) Python function calls).\nInstead, precompute the full distance matrix and pass with `metric='precomputed'`:\n\n```python\ndef compute_distance_matrix(points, shape_weight):\n    \"\"\"Compute the pairwise custom distance matrix for a set of points.\"\"\"\n    w = shape_weight\n    # Use pdist with a vectorized lambda for efficiency\n    def custom_metric(a, b):\n        dx = a[0] - b[0]\n        dy = a[1] - b[1]\n        return np.sqrt((w * dx)**2 + ((2 - w) * dy)**2)\n\n    pairwise = pdist(points, metric=custom_metric)\n    return squareform(pairwise)\n\n\ndef cluster_points(points, epsilon, min_samples, shape_weight):\n    \"\"\"\n    Run DBSCAN with precomputed custom distance matrix.\n    Returns: array of cluster centroids (shape M x 2), or empty if no clusters.\n    \"\"\"\n    if len(points) \u003C min_samples:\n        return np.array([]).reshape(0, 2)\n\n    dist_matrix = compute_distance_matrix(points, shape_weight)\n    db = DBSCAN(eps=epsilon, min_samples=min_samples, metric='precomputed')\n    labels = db.fit_predict(dist_matrix)\n\n    centroids = []\n    for label in set(labels):\n        if label == -1:  # Skip noise\n            continue\n        centroid = points[labels == label].mean(axis=0)\n        centroids.append(centroid)\n\n    return np.array(centroids) if centroids else np.array([]).reshape(0, 2)\n```\n\n## Notes on Behavior\n\n- Points with fewer than `min_samples` neighbors (within `epsilon`) are labeled as noise (-1)\n- Noise points are NOT included in any cluster centroid\n- If ALL points are noise → returns empty array (treat as no clusters found → F1=0, delta=NaN)\n- Cluster centroid = arithmetic mean of all points in that cluster\n\n## Edge Cases\n\n- `len(points) \u003C min_samples`: return empty (DBSCAN requires at least min_samples points to form a core)\n- Single image with 0 citizen science points: return 0.0 F1 and NaN delta immediately\n- Image in expert_df but not in citsci_df: same as 0 citizen science points\n\n## Hyperparameter Grid\n\n```python\nimport numpy as np\nmin_samples_range = list(range(3, 10))              # 3,4,5,6,7,8,9\nepsilon_range = list(range(4, 25, 2))               # 4,6,8,...,24\nshape_weight_range = [round(0.9 + 0.1*i, 1) for i in range(11)]  # 0.9..1.9\n```\n","# DBSCAN with Custom Distance Metric (Efficient)\n\n## Setup\n\n```python\nfrom sklearn.cluster import DBSCAN\nimport numpy as np\nfrom scipy.spatial.distance import pdist, squareform\n```\n\n## Custom Weighted Distance Metric\n\nFor Mars cloud clustering:\n```\nd(a, b) = sqrt((w * Δx)² + ((2 - w) * Δy)²)\n```\n\nWhen w=1, equals standard Euclidean. w>1 attenuates y-distances; w\u003C1 attenuates x-distances.\n\n## Performance: Precomputed Distance Matrix\n\nUsing a Python callable as metric in DBSCAN is **very slow** (O(N²) Python function calls).\nInstead, precompute the full distance matrix and pass with `metric='precomputed'`:\n\n```python\ndef compute_distance_matrix(points, shape_weight):\n    \"\"\"Compute the pairwise custom distance matrix for a set of points.\"\"\"\n    w = shape_weight\n    # Use pdist with a vectorized lambda for efficiency\n    def custom_metric(a, b):\n        dx = a[0] - b[0]\n        dy = a[1] - b[1]\n        return np.sqrt((w * dx)**2 + ((2 - w) * dy)**2)\n\n    pairwise = pdist(points, metric=custom_metric)\n    return squareform(pairwise)\n\n\ndef cluster_points(points, epsilon, min_samples, shape_weight):\n    \"\"\"\n    Run DBSCAN with precomputed custom distance matrix.\n    Returns: array of cluster centroids (shape M x 2), or empty if no clusters.\n    \"\"\"\n    if len(points) \u003C min_samples:\n        return np.array([]).reshape(0, 2)\n\n    dist_matrix = compute_distance_matrix(points, shape_weight)\n    db = DBSCAN(eps=epsilon, min_samples=min_samples, metric='precomputed')\n    labels = db.fit_predict(dist_matrix)\n\n    centroids = []\n    for label in set(labels):\n        if label == -1:  # Skip noise\n            continue\n        centroid = points[labels == label].mean(axis=0)\n        centroids.append(centroid)\n\n    return np.array(centroids) if centroids else np.array([]).reshape(0, 2)\n```\n\n## Notes on Behavior\n\n- Points with fewer than `min_samples` neighbors (within `epsilon`) are labeled as noise (-1)\n- Noise points are NOT included in any cluster centroid\n- If ALL points are noise → returns empty array (treat as no clusters found → F1=0, delta=NaN)\n- Cluster centroid = arithmetic mean of all points in that cluster\n\n## Edge Cases\n\n- `len(points) \u003C min_samples`: return empty (DBSCAN requires at least min_samples points to form a core)\n- Single image with 0 citizen science points: return 0.0 F1 and NaN delta immediately\n- Image in expert_df but not in citsci_df: same as 0 citizen science points\n\n## Hyperparameter Grid\n\n```python\nimport numpy as np\nmin_samples_range = list(range(3, 10))              # 3,4,5,6,7,8,9\nepsilon_range = list(range(4, 25, 2))               # 4,6,8,...,24\nshape_weight_range = [round(0.9 + 0.1*i, 1) for i in range(11)]  # 0.9..1.9\n```\n",{"name":5,"description":7},"main","726e1f79b96b0cb3d729d0e55a1c847e0b5353ae","skills\u002Fb2-self-feedback-claude-sonnet-4-6\u002Fdbscan-parameter-tuning\u002Frun2_dbscan-clustering",7,"https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench","https:\u002F\u002Fcxcscmu.github.io\u002FSkillLearnBench","https:\u002F\u002Fgithub.com\u002Fcxcscmu","2026-10-02 22:55:36",{"source":8,"skillName":5,"commands":40,"prompt":50,"zipAvailable":51},[41,44,47],{"tool":42,"command":43},"npx","npx skills add cxcscmu\u002FSkillLearnBench --skill run2_dbscan-clustering",{"tool":45,"command":46},"bunx","bunx skills add cxcscmu\u002FSkillLearnBench --skill run2_dbscan-clustering",{"tool":48,"command":49},"pnpm","pnpm dlx skills add cxcscmu\u002FSkillLearnBench --skill run2_dbscan-clustering","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fb2-self-feedback-claude-sonnet-4-6\u002Fdbscan-parameter-tuning\u002Frun2_dbscan-clustering\u002FSKILL.md",true,[],[54,72,96,114,134,145],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":65,"publishedAt":26},"01M6KM5KH175P5ABP5JMX2SBDQ","object_counter","Count occurrences of an object in the image using computer vision algorithm.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Fvideo-object-counting\u002Fobject_counter\u002FSKILL.md",[60],{"categoryNo":61,"categoryName":62,"taxonomyType":63,"taxonomyTypeDesc":64},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[66,67,68,69,70,71],"object counter","object","vision","computer","Computer Vision","image",{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":81,"publishedAt":26},"01M6KM51TMVXG79MMPCS6Y8G86","travel-planning-workflow","Invoke this skill when performing travel planning. It defines a structured workflow for generating itineraries. Following this workflow is essential to ensure the final plan satisfies user requirements.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftravel-planning\u002Ftravel-planning-workflow\u002FSKILL.md",[78],{"categoryNo":79,"categoryName":80,"taxonomyType":63,"taxonomyTypeDesc":64},"01JBSEEDCAT000000000000023","Skill与智能体开发",[82,83,84,85,86,87,88,89,90,23,91,92,93,94,95],"workflow","planning","planning workflow","travel","travel planning","travel planning workflow","structured","performing","plan","user","it","requirements","essential","itineraries",{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":26},"01M6KM4DY820XHS9YTSK5QP5HD","bundled-dataset-search-tool","invoke this skill when you need to perform database search for travel planning. This skill provides some useful pre-packaged tools to look up accommodations, attractions, cities, driving distance, flights, and restaurants from the bundled dataset.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftravel-planning\u002Fbundled-dataset-search-tool\u002FSKILL.md",[],[103,104,105,106,107,108,109,85,110,111,86,112,23,113,83],"search","tool","dataset","bundled dataset search tool","restaurants","Tools","up","look","flights","you","database",{"skillNo":115,"name":116,"nameZh":6,"description":117,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":118,"categories":119,"keywords":120,"publishedAt":26},"01M6KM465BB9PQ0B548HQSVAWZ","glm-output","Output-processing guidance for GLM tasks. Especially useful after glm-basics and glm-calibration when you need verifier-matching metrics from output.nc and a final \u002Froot\u002Fmetrics.json.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftemperature-simulation\u002Fglm-output\u002FSKILL.md",[],[121,122,123,23,124,125,126,127,128,129,130,131,112,132,133],"glm output","output","glm","tasks","metrics","json","basics","guidance","matching","verifier","processing","root","calibration",{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":141,"publishedAt":26},"01M6KM40KY3QWDW5W97S39R2BC","glm-calibration","Calibration guidance for GLM tasks. Often most effective after glm-basics has clarified the setup; glm-output is the companion skill for exact final metric computation.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftemperature-simulation\u002Fglm-calibration\u002FSKILL.md",[140],{"categoryNo":79,"categoryName":80,"taxonomyType":63,"taxonomyTypeDesc":64},[142,133,123,122,128,127,124,143,144,23,24],"glm calibration","setup","companion",{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":151,"publishedAt":26},"01M6KM3VHHVSGMGHJQB3PPT6YG","glm-basics","A good starting point for GLM calibration tasks. Use to inspect glm3.nml, confirm how GLM runs, and identify the relevant files before moving on to calibration and output evaluation.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftemperature-simulation\u002Fglm-basics\u002FSKILL.md",[],[152,123,127,153,142,133,23,122,154,155,156,157,158,159,124],"glm basics","inspect","relevant","files","good","use","how","Evaluation"]