[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KFWRVTS22CZPYSS0PP2RKC":3,"skill-same:01M6KFWRVTS22CZPYSS0PP2RKC":57},{"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":30,"linkOnly":31,"markdownContent":32,"markdownBody":33,"frontMatter":34,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":40,"ownerHtmlUrl":41,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":42,"install":43,"resources":56},"01M6KFWRVTS22CZPYSS0PP2RKC","run2_glm-rmse-metrics",null,"Exact RMSE metric computation for GLM lake temperature evaluation - verified matching procedure with correct threshold values.","cxcscmu\u002FSkillLearnBench","cxcscmu","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F142838272?v=4","01M6KCQDFQWPDSN0MKZY96MSQP","01M6KCQDM7G70TKJ28NTSQC0RZ",92,"MIT","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fb2-self-feedback-claude-sonnet-4-6\u002Ftemperature-simulation\u002Frun2_glm-rmse-metrics\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29],"metrics","glm","run2 glm rmse metrics","Evaluation","correct","matching","with","temperature","values","verified","for","metric","2026-11-10 10:12:30",false,"---\nname: run2_glm-rmse-metrics\ndescription: Exact RMSE metric computation for GLM lake temperature evaluation - verified matching procedure with correct threshold values.\n---\n\n# GLM RMSE Metrics Skill (Improved)\n\n## Metric Definitions\n- **overall_rmse**: RMSE over ALL matched pairs\n- **annual_deep_rmse**: RMSE for `rounded_depth >= 13`, all months\n- **summer_deep_rmse**: RMSE for `rounded_depth >= 13` AND `month in [6,7,8,9]`\n\n## RMSE Thresholds (Lake Mendota)\n- `overall_rmse \u003C 1.60`\n- `annual_deep_rmse \u003C 1.55`\n- `summer_deep_rmse \u003C 1.70`\n\n## Complete Evaluation Script\n\n```python\nimport netCDF4 as nc\nimport numpy as np\nimport pandas as pd\nimport json\n\ndef evaluate_and_save(nc_path='\u002Froot\u002Foutput\u002Foutput.nc',\n                       obs_path='\u002Froot\u002Ffield_temp_oxy.csv',\n                       metrics_path='\u002Froot\u002Fmetrics.json'):\n    # Build matched pairs\n    obs = pd.read_csv(obs_path, parse_dates=['datetime'])\n    obs['rounded_depth'] = obs['depth'].round().astype(int)\n\n    ds = nc.Dataset(nc_path)\n    time_var = ds.variables['time']\n    times = nc.num2date(time_var[:], time_var.units)\n    sim_times = pd.to_datetime([t.strftime('%Y-%m-%d %H:%M:%S') for t in times])\n\n    z = ds.variables['z'][:]\n    temp = ds.variables['temp'][:]\n    NS = ds.variables['NS'][:]\n\n    records = []\n    for i in range(len(sim_times)):\n        n = int(NS[i])\n        z_surf = float(z[i, n-1, 0, 0])\n        for j in range(n):\n            z_layer = float(z[i, j, 0, 0])\n            rdepth = int(round(z_surf - z_layer))\n            records.append({'datetime': sim_times[i], 'rounded_depth': rdepth,\n                             'sim_temp': float(temp[i, j, 0, 0])})\n\n    sim_df = pd.DataFrame(records)\n    sim_df = sim_df.groupby(['datetime', 'rounded_depth'])['sim_temp'].mean().reset_index()\n    merged = obs.merge(sim_df, on=['datetime', 'rounded_depth'])\n    ds.close()\n\n    # Compute metrics\n    residuals = merged['sim_temp'] - merged['temp']\n    overall_rmse = float(np.sqrt(np.mean(residuals**2)))\n\n    deep = merged[merged['rounded_depth'] >= 13]\n    annual_deep_rmse = float(np.sqrt(np.mean((deep['sim_temp'] - deep['temp'])**2)))\n\n    summer_deep = deep[deep['datetime'].dt.month.isin([6, 7, 8, 9])]\n    summer_deep_rmse = float(np.sqrt(np.mean((summer_deep['sim_temp'] - summer_deep['temp'])**2)))\n\n    metrics = {\n        'overall_rmse': overall_rmse,\n        'annual_deep_rmse': annual_deep_rmse,\n        'summer_deep_rmse': summer_deep_rmse,\n        'overall_n_pairs': int(len(merged)),\n        'annual_deep_n_pairs': int(len(deep)),\n        'summer_deep_n_pairs': int(len(summer_deep))\n    }\n\n    with open(metrics_path, 'w') as f:\n        json.dump(metrics, f, indent=2)\n\n    return metrics\n```\n\n## Verified Results (Lake Mendota, best parameters)\n```json\n{\n  \"overall_rmse\": 1.3662,\n  \"annual_deep_rmse\": 1.3468,\n  \"summer_deep_rmse\": 1.4713,\n  \"overall_n_pairs\": 2819,\n  \"annual_deep_n_pairs\": 1327,\n  \"summer_deep_n_pairs\": 678\n}\n```\n\n## Common Pitfalls\n- Do NOT use nearest-time matching or interpolation\n- Do NOT use alternative depth binning (only round to nearest integer)\n- Merge must be exact on both `datetime` AND `rounded_depth`\n- Use `.groupby().mean()` on sim_df before merging to handle duplicate rounded depths\n","# GLM RMSE Metrics Skill (Improved)\n\n## Metric Definitions\n- **overall_rmse**: RMSE over ALL matched pairs\n- **annual_deep_rmse**: RMSE for `rounded_depth >= 13`, all months\n- **summer_deep_rmse**: RMSE for `rounded_depth >= 13` AND `month in [6,7,8,9]`\n\n## RMSE Thresholds (Lake Mendota)\n- `overall_rmse \u003C 1.60`\n- `annual_deep_rmse \u003C 1.55`\n- `summer_deep_rmse \u003C 1.70`\n\n## Complete Evaluation Script\n\n```python\nimport netCDF4 as nc\nimport numpy as np\nimport pandas as pd\nimport json\n\ndef evaluate_and_save(nc_path='\u002Froot\u002Foutput\u002Foutput.nc',\n                       obs_path='\u002Froot\u002Ffield_temp_oxy.csv',\n                       metrics_path='\u002Froot\u002Fmetrics.json'):\n    # Build matched pairs\n    obs = pd.read_csv(obs_path, parse_dates=['datetime'])\n    obs['rounded_depth'] = obs['depth'].round().astype(int)\n\n    ds = nc.Dataset(nc_path)\n    time_var = ds.variables['time']\n    times = nc.num2date(time_var[:], time_var.units)\n    sim_times = pd.to_datetime([t.strftime('%Y-%m-%d %H:%M:%S') for t in times])\n\n    z = ds.variables['z'][:]\n    temp = ds.variables['temp'][:]\n    NS = ds.variables['NS'][:]\n\n    records = []\n    for i in range(len(sim_times)):\n        n = int(NS[i])\n        z_surf = float(z[i, n-1, 0, 0])\n        for j in range(n):\n            z_layer = float(z[i, j, 0, 0])\n            rdepth = int(round(z_surf - z_layer))\n            records.append({'datetime': sim_times[i], 'rounded_depth': rdepth,\n                             'sim_temp': float(temp[i, j, 0, 0])})\n\n    sim_df = pd.DataFrame(records)\n    sim_df = sim_df.groupby(['datetime', 'rounded_depth'])['sim_temp'].mean().reset_index()\n    merged = obs.merge(sim_df, on=['datetime', 'rounded_depth'])\n    ds.close()\n\n    # Compute metrics\n    residuals = merged['sim_temp'] - merged['temp']\n    overall_rmse = float(np.sqrt(np.mean(residuals**2)))\n\n    deep = merged[merged['rounded_depth'] >= 13]\n    annual_deep_rmse = float(np.sqrt(np.mean((deep['sim_temp'] - deep['temp'])**2)))\n\n    summer_deep = deep[deep['datetime'].dt.month.isin([6, 7, 8, 9])]\n    summer_deep_rmse = float(np.sqrt(np.mean((summer_deep['sim_temp'] - summer_deep['temp'])**2)))\n\n    metrics = {\n        'overall_rmse': overall_rmse,\n        'annual_deep_rmse': annual_deep_rmse,\n        'summer_deep_rmse': summer_deep_rmse,\n        'overall_n_pairs': int(len(merged)),\n        'annual_deep_n_pairs': int(len(deep)),\n        'summer_deep_n_pairs': int(len(summer_deep))\n    }\n\n    with open(metrics_path, 'w') as f:\n        json.dump(metrics, f, indent=2)\n\n    return metrics\n```\n\n## Verified Results (Lake Mendota, best parameters)\n```json\n{\n  \"overall_rmse\": 1.3662,\n  \"annual_deep_rmse\": 1.3468,\n  \"summer_deep_rmse\": 1.4713,\n  \"overall_n_pairs\": 2819,\n  \"annual_deep_n_pairs\": 1327,\n  \"summer_deep_n_pairs\": 678\n}\n```\n\n## Common Pitfalls\n- Do NOT use nearest-time matching or interpolation\n- Do NOT use alternative depth binning (only round to nearest integer)\n- Merge must be exact on both `datetime` AND `rounded_depth`\n- Use `.groupby().mean()` on sim_df before merging to handle duplicate rounded depths\n",{"name":5,"description":7},"main","726e1f79b96b0cb3d729d0e55a1c847e0b5353ae","skills\u002Fb2-self-feedback-claude-sonnet-4-6\u002Ftemperature-simulation\u002Frun2_glm-rmse-metrics",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":44,"prompt":54,"zipAvailable":55},[45,48,51],{"tool":46,"command":47},"npx","npx skills add cxcscmu\u002FSkillLearnBench --skill run2_glm-rmse-metrics",{"tool":49,"command":50},"bunx","bunx skills add cxcscmu\u002FSkillLearnBench --skill run2_glm-rmse-metrics",{"tool":52,"command":53},"pnpm","pnpm dlx skills add cxcscmu\u002FSkillLearnBench --skill run2_glm-rmse-metrics","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fb2-self-feedback-claude-sonnet-4-6\u002Ftemperature-simulation\u002Frun2_glm-rmse-metrics\u002FSKILL.md",true,[],[58,76,100,118,135,146],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":69,"publishedAt":30},"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",[64],{"categoryNo":65,"categoryName":66,"taxonomyType":67,"taxonomyTypeDesc":68},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[70,71,72,73,74,75],"object counter","object","vision","computer","Computer Vision","image",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":85,"publishedAt":30},"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",[82],{"categoryNo":83,"categoryName":84,"taxonomyType":67,"taxonomyTypeDesc":68},"01JBSEEDCAT000000000000023","Skill与智能体开发",[86,87,88,89,90,91,92,93,94,28,95,96,97,98,99],"workflow","planning","planning workflow","travel","travel planning","travel planning workflow","structured","performing","plan","user","it","requirements","essential","itineraries",{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":30},"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",[],[107,108,109,110,111,112,113,89,114,115,90,116,28,117,87],"search","tool","dataset","bundled dataset search tool","restaurants","Tools","up","look","flights","you","database",{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":30},"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",[],[125,126,19,28,127,18,128,129,130,23,131,132,116,133,134],"glm output","output","tasks","json","basics","guidance","verifier","processing","root","calibration",{"skillNo":136,"name":137,"nameZh":6,"description":138,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":139,"categories":140,"keywords":142,"publishedAt":30},"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",[141],{"categoryNo":83,"categoryName":84,"taxonomyType":67,"taxonomyTypeDesc":68},[143,134,19,126,130,129,127,144,145,28,29],"glm calibration","setup","companion",{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":30},"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",[],[153,19,129,154,143,134,28,126,155,156,157,158,159,21,127],"glm basics","inspect","relevant","files","good","use","how"]