[{"data":1,"prerenderedAt":1352},["ShallowReactive",2],{"content-query-yyPdCEj4Pz":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"body":10,"_type":1346,"_id":1347,"_source":1348,"_file":1349,"_stem":1350,"_extension":1351},"\u002Fdocs\u002Ffeatures\u002Fml-experiments","features",false,"","ML Experiments","Track, compare, and manage machine learning experiments",{"type":11,"children":12,"toc":1302},"root",[13,21,27,34,41,77,83,88,244,250,256,261,305,311,316,379,385,390,483,489,495,500,523,529,534,564,570,575,618,624,630,635,662,668,673,696,702,708,713,746,752,757,835,841,847,904,910,915,946,952,958,963,971,977,982,1035,1041,1046,1064,1070,1076,1117,1123,1146,1152,1175,1181,1187,1205,1211,1229,1235,1253,1259,1296],{"type":14,"tag":15,"props":16,"children":18},"element","h1",{"id":17},"ml-experiments",[19],{"type":20,"value":8},"text",{"type":14,"tag":22,"props":23,"children":24},"p",{},[25],{"type":20,"value":26},"Track all your machine learning experiments in one place. Monitor metrics, compare runs, and manage model versions.",{"type":14,"tag":28,"props":29,"children":31},"h2",{"id":30},"creating-an-experiment",[32],{"type":20,"value":33},"Creating an Experiment",{"type":14,"tag":35,"props":36,"children":38},"h3",{"id":37},"start-tracking",[39],{"type":20,"value":40},"Start Tracking",{"type":14,"tag":42,"props":43,"children":44},"ol",{},[45,58,67,72],{"type":14,"tag":46,"props":47,"children":48},"li",{},[49,51,56],{"type":20,"value":50},"Click ",{"type":14,"tag":52,"props":53,"children":54},"strong",{},[55],{"type":20,"value":8},{"type":20,"value":57}," in the sidebar",{"type":14,"tag":46,"props":59,"children":60},{},[61,62],{"type":20,"value":50},{"type":14,"tag":52,"props":63,"children":64},{},[65],{"type":20,"value":66},"New Experiment",{"type":14,"tag":46,"props":68,"children":69},{},[70],{"type":20,"value":71},"Give your experiment a name (e.g., \"Customer Churn Model\")",{"type":14,"tag":46,"props":73,"children":74},{},[75],{"type":20,"value":76},"Optionally add a description",{"type":14,"tag":35,"props":78,"children":80},{"id":79},"logging-metrics",[81],{"type":20,"value":82},"Logging Metrics",{"type":14,"tag":22,"props":84,"children":85},{},[86],{"type":20,"value":87},"From your notebook, log metrics automatically:",{"type":14,"tag":89,"props":90,"children":94},"pre",{"code":91,"language":92,"meta":7,"className":93,"style":7},"from credvault import experiment\n\n# Start tracking\nexp = experiment.start('Customer Churn Model')\n\n# Log metrics during training\nfor epoch in range(10):\n    loss = model.train_step(data)\n    exp.log_metric('loss', loss, step=epoch)\n    \n    accuracy = model.evaluate(test_data)\n    exp.log_metric('accuracy', accuracy, step=epoch)\n\n# Log final results\nexp.log_metric('final_accuracy', 0.92)\nexp.log_metric('final_loss', 0.18)\n","python","language-python shiki shiki-themes github-dark",[95],{"type":14,"tag":96,"props":97,"children":98},"code",{"__ignoreMap":7},[99,110,120,129,138,146,155,164,173,182,191,200,209,217,226,235],{"type":14,"tag":100,"props":101,"children":104},"span",{"class":102,"line":103},"line",1,[105],{"type":14,"tag":100,"props":106,"children":107},{},[108],{"type":20,"value":109},"from credvault import experiment\n",{"type":14,"tag":100,"props":111,"children":113},{"class":102,"line":112},2,[114],{"type":14,"tag":100,"props":115,"children":117},{"emptyLinePlaceholder":116},true,[118],{"type":20,"value":119},"\n",{"type":14,"tag":100,"props":121,"children":123},{"class":102,"line":122},3,[124],{"type":14,"tag":100,"props":125,"children":126},{},[127],{"type":20,"value":128},"# Start tracking\n",{"type":14,"tag":100,"props":130,"children":132},{"class":102,"line":131},4,[133],{"type":14,"tag":100,"props":134,"children":135},{},[136],{"type":20,"value":137},"exp = experiment.start('Customer Churn Model')\n",{"type":14,"tag":100,"props":139,"children":141},{"class":102,"line":140},5,[142],{"type":14,"tag":100,"props":143,"children":144},{"emptyLinePlaceholder":116},[145],{"type":20,"value":119},{"type":14,"tag":100,"props":147,"children":149},{"class":102,"line":148},6,[150],{"type":14,"tag":100,"props":151,"children":152},{},[153],{"type":20,"value":154},"# Log metrics during training\n",{"type":14,"tag":100,"props":156,"children":158},{"class":102,"line":157},7,[159],{"type":14,"tag":100,"props":160,"children":161},{},[162],{"type":20,"value":163},"for epoch in range(10):\n",{"type":14,"tag":100,"props":165,"children":167},{"class":102,"line":166},8,[168],{"type":14,"tag":100,"props":169,"children":170},{},[171],{"type":20,"value":172},"    loss = model.train_step(data)\n",{"type":14,"tag":100,"props":174,"children":176},{"class":102,"line":175},9,[177],{"type":14,"tag":100,"props":178,"children":179},{},[180],{"type":20,"value":181},"    exp.log_metric('loss', loss, step=epoch)\n",{"type":14,"tag":100,"props":183,"children":185},{"class":102,"line":184},10,[186],{"type":14,"tag":100,"props":187,"children":188},{},[189],{"type":20,"value":190},"    \n",{"type":14,"tag":100,"props":192,"children":194},{"class":102,"line":193},11,[195],{"type":14,"tag":100,"props":196,"children":197},{},[198],{"type":20,"value":199},"    accuracy = model.evaluate(test_data)\n",{"type":14,"tag":100,"props":201,"children":203},{"class":102,"line":202},12,[204],{"type":14,"tag":100,"props":205,"children":206},{},[207],{"type":20,"value":208},"    exp.log_metric('accuracy', accuracy, step=epoch)\n",{"type":14,"tag":100,"props":210,"children":212},{"class":102,"line":211},13,[213],{"type":14,"tag":100,"props":214,"children":215},{"emptyLinePlaceholder":116},[216],{"type":20,"value":119},{"type":14,"tag":100,"props":218,"children":220},{"class":102,"line":219},14,[221],{"type":14,"tag":100,"props":222,"children":223},{},[224],{"type":20,"value":225},"# Log final results\n",{"type":14,"tag":100,"props":227,"children":229},{"class":102,"line":228},15,[230],{"type":14,"tag":100,"props":231,"children":232},{},[233],{"type":20,"value":234},"exp.log_metric('final_accuracy', 0.92)\n",{"type":14,"tag":100,"props":236,"children":238},{"class":102,"line":237},16,[239],{"type":14,"tag":100,"props":240,"children":241},{},[242],{"type":20,"value":243},"exp.log_metric('final_loss', 0.18)\n",{"type":14,"tag":28,"props":245,"children":247},{"id":246},"tracking-runs",[248],{"type":20,"value":249},"Tracking Runs",{"type":14,"tag":35,"props":251,"children":253},{"id":252},"what-gets-recorded",[254],{"type":20,"value":255},"What Gets Recorded",{"type":14,"tag":22,"props":257,"children":258},{},[259],{"type":20,"value":260},"Each run captures:",{"type":14,"tag":262,"props":263,"children":264},"ul",{},[265,275,285,295],{"type":14,"tag":46,"props":266,"children":267},{},[268,273],{"type":14,"tag":52,"props":269,"children":270},{},[271],{"type":20,"value":272},"Metrics",{"type":20,"value":274}," - Accuracy, loss, precision, recall, etc.",{"type":14,"tag":46,"props":276,"children":277},{},[278,283],{"type":14,"tag":52,"props":279,"children":280},{},[281],{"type":20,"value":282},"Parameters",{"type":20,"value":284}," - Learning rate, batch size, model architecture",{"type":14,"tag":46,"props":286,"children":287},{},[288,293],{"type":14,"tag":52,"props":289,"children":290},{},[291],{"type":20,"value":292},"Files",{"type":20,"value":294}," - Model weights, artifacts, outputs",{"type":14,"tag":46,"props":296,"children":297},{},[298,303],{"type":14,"tag":52,"props":299,"children":300},{},[301],{"type":20,"value":302},"Metadata",{"type":20,"value":304}," - Start time, duration, status",{"type":14,"tag":35,"props":306,"children":308},{"id":307},"logging-parameters",[309],{"type":20,"value":310},"Logging Parameters",{"type":14,"tag":22,"props":312,"children":313},{},[314],{"type":20,"value":315},"Track the configuration of each run:",{"type":14,"tag":89,"props":317,"children":319},{"code":318,"language":92,"meta":7,"className":93,"style":7},"exp.log_params({\n    'learning_rate': 0.001,\n    'batch_size': 32,\n    'epochs': 100,\n    'optimizer': 'adam',\n    'loss_function': 'crossentropy'\n})\n",[320],{"type":14,"tag":96,"props":321,"children":322},{"__ignoreMap":7},[323,331,339,347,355,363,371],{"type":14,"tag":100,"props":324,"children":325},{"class":102,"line":103},[326],{"type":14,"tag":100,"props":327,"children":328},{},[329],{"type":20,"value":330},"exp.log_params({\n",{"type":14,"tag":100,"props":332,"children":333},{"class":102,"line":112},[334],{"type":14,"tag":100,"props":335,"children":336},{},[337],{"type":20,"value":338},"    'learning_rate': 0.001,\n",{"type":14,"tag":100,"props":340,"children":341},{"class":102,"line":122},[342],{"type":14,"tag":100,"props":343,"children":344},{},[345],{"type":20,"value":346},"    'batch_size': 32,\n",{"type":14,"tag":100,"props":348,"children":349},{"class":102,"line":131},[350],{"type":14,"tag":100,"props":351,"children":352},{},[353],{"type":20,"value":354},"    'epochs': 100,\n",{"type":14,"tag":100,"props":356,"children":357},{"class":102,"line":140},[358],{"type":14,"tag":100,"props":359,"children":360},{},[361],{"type":20,"value":362},"    'optimizer': 'adam',\n",{"type":14,"tag":100,"props":364,"children":365},{"class":102,"line":148},[366],{"type":14,"tag":100,"props":367,"children":368},{},[369],{"type":20,"value":370},"    'loss_function': 'crossentropy'\n",{"type":14,"tag":100,"props":372,"children":373},{"class":102,"line":157},[374],{"type":14,"tag":100,"props":375,"children":376},{},[377],{"type":20,"value":378},"})\n",{"type":14,"tag":35,"props":380,"children":382},{"id":381},"saving-artifacts",[383],{"type":20,"value":384},"Saving Artifacts",{"type":14,"tag":22,"props":386,"children":387},{},[388],{"type":20,"value":389},"Save model files and results:",{"type":14,"tag":89,"props":391,"children":393},{"code":392,"language":92,"meta":7,"className":93,"style":7},"# Save trained model\nmodel.save('model.pkl')\nexp.log_artifact('model.pkl')\n\n# Save predictions\npredictions.to_csv('predictions.csv')\nexp.log_artifact('predictions.csv')\n\n# Save visualizations\nplt.savefig('confusion_matrix.png')\nexp.log_artifact('confusion_matrix.png')\n",[394],{"type":14,"tag":96,"props":395,"children":396},{"__ignoreMap":7},[397,405,413,421,428,436,444,452,459,467,475],{"type":14,"tag":100,"props":398,"children":399},{"class":102,"line":103},[400],{"type":14,"tag":100,"props":401,"children":402},{},[403],{"type":20,"value":404},"# Save trained model\n",{"type":14,"tag":100,"props":406,"children":407},{"class":102,"line":112},[408],{"type":14,"tag":100,"props":409,"children":410},{},[411],{"type":20,"value":412},"model.save('model.pkl')\n",{"type":14,"tag":100,"props":414,"children":415},{"class":102,"line":122},[416],{"type":14,"tag":100,"props":417,"children":418},{},[419],{"type":20,"value":420},"exp.log_artifact('model.pkl')\n",{"type":14,"tag":100,"props":422,"children":423},{"class":102,"line":131},[424],{"type":14,"tag":100,"props":425,"children":426},{"emptyLinePlaceholder":116},[427],{"type":20,"value":119},{"type":14,"tag":100,"props":429,"children":430},{"class":102,"line":140},[431],{"type":14,"tag":100,"props":432,"children":433},{},[434],{"type":20,"value":435},"# Save predictions\n",{"type":14,"tag":100,"props":437,"children":438},{"class":102,"line":148},[439],{"type":14,"tag":100,"props":440,"children":441},{},[442],{"type":20,"value":443},"predictions.to_csv('predictions.csv')\n",{"type":14,"tag":100,"props":445,"children":446},{"class":102,"line":157},[447],{"type":14,"tag":100,"props":448,"children":449},{},[450],{"type":20,"value":451},"exp.log_artifact('predictions.csv')\n",{"type":14,"tag":100,"props":453,"children":454},{"class":102,"line":166},[455],{"type":14,"tag":100,"props":456,"children":457},{"emptyLinePlaceholder":116},[458],{"type":20,"value":119},{"type":14,"tag":100,"props":460,"children":461},{"class":102,"line":175},[462],{"type":14,"tag":100,"props":463,"children":464},{},[465],{"type":20,"value":466},"# Save visualizations\n",{"type":14,"tag":100,"props":468,"children":469},{"class":102,"line":184},[470],{"type":14,"tag":100,"props":471,"children":472},{},[473],{"type":20,"value":474},"plt.savefig('confusion_matrix.png')\n",{"type":14,"tag":100,"props":476,"children":477},{"class":102,"line":193},[478],{"type":14,"tag":100,"props":479,"children":480},{},[481],{"type":20,"value":482},"exp.log_artifact('confusion_matrix.png')\n",{"type":14,"tag":28,"props":484,"children":486},{"id":485},"comparing-experiments",[487],{"type":20,"value":488},"Comparing Experiments",{"type":14,"tag":35,"props":490,"children":492},{"id":491},"view-all-runs",[493],{"type":20,"value":494},"View All Runs",{"type":14,"tag":22,"props":496,"children":497},{},[498],{"type":20,"value":499},"See all runs in an experiment with a table:",{"type":14,"tag":262,"props":501,"children":502},{},[503,508,513,518],{"type":14,"tag":46,"props":504,"children":505},{},[506],{"type":20,"value":507},"Metrics columns",{"type":14,"tag":46,"props":509,"children":510},{},[511],{"type":20,"value":512},"Parameters columns",{"type":14,"tag":46,"props":514,"children":515},{},[516],{"type":20,"value":517},"Status and duration",{"type":14,"tag":46,"props":519,"children":520},{},[521],{"type":20,"value":522},"Click any run for details",{"type":14,"tag":35,"props":524,"children":526},{"id":525},"side-by-side-comparison",[527],{"type":20,"value":528},"Side-by-Side Comparison",{"type":14,"tag":22,"props":530,"children":531},{},[532],{"type":20,"value":533},"Compare two runs:",{"type":14,"tag":42,"props":535,"children":536},{},[537,549,554,559],{"type":14,"tag":46,"props":538,"children":539},{},[540,542,547],{"type":20,"value":541},"Select ",{"type":14,"tag":52,"props":543,"children":544},{},[545],{"type":20,"value":546},"Compare",{"type":20,"value":548}," from the toolbar",{"type":14,"tag":46,"props":550,"children":551},{},[552],{"type":20,"value":553},"Choose two runs to compare",{"type":14,"tag":46,"props":555,"children":556},{},[557],{"type":20,"value":558},"See metric differences highlighted",{"type":14,"tag":46,"props":560,"children":561},{},[562],{"type":20,"value":563},"View parameter differences",{"type":14,"tag":35,"props":565,"children":567},{"id":566},"finding-best-run",[568],{"type":20,"value":569},"Finding Best Run",{"type":14,"tag":22,"props":571,"children":572},{},[573],{"type":20,"value":574},"Filter runs by metrics:",{"type":14,"tag":262,"props":576,"children":577},{},[578,588,598,608],{"type":14,"tag":46,"props":579,"children":580},{},[581,586],{"type":14,"tag":52,"props":582,"children":583},{},[584],{"type":20,"value":585},"Best Accuracy",{"type":20,"value":587}," - Highest accuracy value",{"type":14,"tag":46,"props":589,"children":590},{},[591,596],{"type":14,"tag":52,"props":592,"children":593},{},[594],{"type":20,"value":595},"Lowest Loss",{"type":20,"value":597}," - Minimum loss value",{"type":14,"tag":46,"props":599,"children":600},{},[601,606],{"type":14,"tag":52,"props":602,"children":603},{},[604],{"type":20,"value":605},"Fastest",{"type":20,"value":607}," - Shortest training time",{"type":14,"tag":46,"props":609,"children":610},{},[611,616],{"type":14,"tag":52,"props":612,"children":613},{},[614],{"type":20,"value":615},"Custom",{"type":20,"value":617}," - Define your own criteria",{"type":14,"tag":28,"props":619,"children":621},{"id":620},"managing-models",[622],{"type":20,"value":623},"Managing Models",{"type":14,"tag":35,"props":625,"children":627},{"id":626},"version-your-models",[628],{"type":20,"value":629},"Version Your Models",{"type":14,"tag":22,"props":631,"children":632},{},[633],{"type":20,"value":634},"Tag important runs as model versions:",{"type":14,"tag":42,"props":636,"children":637},{},[638,643,652,657],{"type":14,"tag":46,"props":639,"children":640},{},[641],{"type":20,"value":642},"Find a successful run",{"type":14,"tag":46,"props":644,"children":645},{},[646,647],{"type":20,"value":50},{"type":14,"tag":52,"props":648,"children":649},{},[650],{"type":20,"value":651},"Save as Model",{"type":14,"tag":46,"props":653,"children":654},{},[655],{"type":20,"value":656},"Give it a version name (e.g., \"v1.0-production\")",{"type":14,"tag":46,"props":658,"children":659},{},[660],{"type":20,"value":661},"Add description",{"type":14,"tag":35,"props":663,"children":665},{"id":664},"model-registry",[666],{"type":20,"value":667},"Model Registry",{"type":14,"tag":22,"props":669,"children":670},{},[671],{"type":20,"value":672},"Your saved models appear in the Model Registry:",{"type":14,"tag":262,"props":674,"children":675},{},[676,681,686,691],{"type":14,"tag":46,"props":677,"children":678},{},[679],{"type":20,"value":680},"View all versions",{"type":14,"tag":46,"props":682,"children":683},{},[684],{"type":20,"value":685},"See metrics for each version",{"type":14,"tag":46,"props":687,"children":688},{},[689],{"type":20,"value":690},"Download model files",{"type":14,"tag":46,"props":692,"children":693},{},[694],{"type":20,"value":695},"Promote to production",{"type":14,"tag":28,"props":697,"children":699},{"id":698},"visualizations",[700],{"type":20,"value":701},"Visualizations",{"type":14,"tag":35,"props":703,"children":705},{"id":704},"metric-graphs",[706],{"type":20,"value":707},"Metric Graphs",{"type":14,"tag":22,"props":709,"children":710},{},[711],{"type":20,"value":712},"Automatically see charts:",{"type":14,"tag":262,"props":714,"children":715},{},[716,726,736],{"type":14,"tag":46,"props":717,"children":718},{},[719,724],{"type":14,"tag":52,"props":720,"children":721},{},[722],{"type":20,"value":723},"Line graphs",{"type":20,"value":725}," - Metric trends over time",{"type":14,"tag":46,"props":727,"children":728},{},[729,734],{"type":14,"tag":52,"props":730,"children":731},{},[732],{"type":20,"value":733},"Scatter plots",{"type":20,"value":735}," - Relationship between metrics",{"type":14,"tag":46,"props":737,"children":738},{},[739,744],{"type":14,"tag":52,"props":740,"children":741},{},[742],{"type":20,"value":743},"Parallel coordinates",{"type":20,"value":745}," - Compare multiple runs",{"type":14,"tag":35,"props":747,"children":749},{"id":748},"custom-analysis",[750],{"type":20,"value":751},"Custom Analysis",{"type":14,"tag":22,"props":753,"children":754},{},[755],{"type":20,"value":756},"Export data for custom analysis:",{"type":14,"tag":89,"props":758,"children":760},{"code":759,"language":92,"meta":7,"className":93,"style":7},"# Download experiment data\ndata = exp.get_runs()\n\n# Plot custom comparison\nimport matplotlib.pyplot as plt\nplt.scatter(data['learning_rate'], data['accuracy'])\nplt.xlabel('Learning Rate')\nplt.ylabel('Accuracy')\nplt.show()\n",[761],{"type":14,"tag":96,"props":762,"children":763},{"__ignoreMap":7},[764,772,780,787,795,803,811,819,827],{"type":14,"tag":100,"props":765,"children":766},{"class":102,"line":103},[767],{"type":14,"tag":100,"props":768,"children":769},{},[770],{"type":20,"value":771},"# Download experiment data\n",{"type":14,"tag":100,"props":773,"children":774},{"class":102,"line":112},[775],{"type":14,"tag":100,"props":776,"children":777},{},[778],{"type":20,"value":779},"data = exp.get_runs()\n",{"type":14,"tag":100,"props":781,"children":782},{"class":102,"line":122},[783],{"type":14,"tag":100,"props":784,"children":785},{"emptyLinePlaceholder":116},[786],{"type":20,"value":119},{"type":14,"tag":100,"props":788,"children":789},{"class":102,"line":131},[790],{"type":14,"tag":100,"props":791,"children":792},{},[793],{"type":20,"value":794},"# Plot custom comparison\n",{"type":14,"tag":100,"props":796,"children":797},{"class":102,"line":140},[798],{"type":14,"tag":100,"props":799,"children":800},{},[801],{"type":20,"value":802},"import matplotlib.pyplot as plt\n",{"type":14,"tag":100,"props":804,"children":805},{"class":102,"line":148},[806],{"type":14,"tag":100,"props":807,"children":808},{},[809],{"type":20,"value":810},"plt.scatter(data['learning_rate'], data['accuracy'])\n",{"type":14,"tag":100,"props":812,"children":813},{"class":102,"line":157},[814],{"type":14,"tag":100,"props":815,"children":816},{},[817],{"type":20,"value":818},"plt.xlabel('Learning Rate')\n",{"type":14,"tag":100,"props":820,"children":821},{"class":102,"line":166},[822],{"type":14,"tag":100,"props":823,"children":824},{},[825],{"type":20,"value":826},"plt.ylabel('Accuracy')\n",{"type":14,"tag":100,"props":828,"children":829},{"class":102,"line":175},[830],{"type":14,"tag":100,"props":831,"children":832},{},[833],{"type":20,"value":834},"plt.show()\n",{"type":14,"tag":28,"props":836,"children":838},{"id":837},"team-collaboration",[839],{"type":20,"value":840},"Team Collaboration",{"type":14,"tag":35,"props":842,"children":844},{"id":843},"sharing-experiments",[845],{"type":20,"value":846},"Sharing Experiments",{"type":14,"tag":42,"props":848,"children":849},{},[850,861,866],{"type":14,"tag":46,"props":851,"children":852},{},[853,854,859],{"type":20,"value":50},{"type":14,"tag":52,"props":855,"children":856},{},[857],{"type":20,"value":858},"Share",{"type":20,"value":860}," on experiment",{"type":14,"tag":46,"props":862,"children":863},{},[864],{"type":20,"value":865},"Select team members",{"type":14,"tag":46,"props":867,"children":868},{},[869,871],{"type":20,"value":870},"Choose permission level:\n",{"type":14,"tag":262,"props":872,"children":873},{},[874,884,894],{"type":14,"tag":46,"props":875,"children":876},{},[877,882],{"type":14,"tag":52,"props":878,"children":879},{},[880],{"type":20,"value":881},"View",{"type":20,"value":883}," - See results only",{"type":14,"tag":46,"props":885,"children":886},{},[887,892],{"type":14,"tag":52,"props":888,"children":889},{},[890],{"type":20,"value":891},"Comment",{"type":20,"value":893}," - Add notes and insights",{"type":14,"tag":46,"props":895,"children":896},{},[897,902],{"type":14,"tag":52,"props":898,"children":899},{},[900],{"type":20,"value":901},"Edit",{"type":20,"value":903}," - Create new runs",{"type":14,"tag":35,"props":905,"children":907},{"id":906},"notebooks-and-experiments",[908],{"type":20,"value":909},"Notebooks and Experiments",{"type":14,"tag":22,"props":911,"children":912},{},[913],{"type":20,"value":914},"Link experiments to your notebooks:",{"type":14,"tag":89,"props":916,"children":918},{"code":917,"language":92,"meta":7,"className":93,"style":7},"# Run notebook, automatically log to experiment\n# Share notebook with team\n# Team sees both code and results together\n",[919],{"type":14,"tag":96,"props":920,"children":921},{"__ignoreMap":7},[922,930,938],{"type":14,"tag":100,"props":923,"children":924},{"class":102,"line":103},[925],{"type":14,"tag":100,"props":926,"children":927},{},[928],{"type":20,"value":929},"# Run notebook, automatically log to experiment\n",{"type":14,"tag":100,"props":931,"children":932},{"class":102,"line":112},[933],{"type":14,"tag":100,"props":934,"children":935},{},[936],{"type":20,"value":937},"# Share notebook with team\n",{"type":14,"tag":100,"props":939,"children":940},{"class":102,"line":122},[941],{"type":14,"tag":100,"props":942,"children":943},{},[944],{"type":20,"value":945},"# Team sees both code and results together\n",{"type":14,"tag":28,"props":947,"children":949},{"id":948},"best-practices",[950],{"type":20,"value":951},"Best Practices",{"type":14,"tag":35,"props":953,"children":955},{"id":954},"naming-convention",[956],{"type":20,"value":957},"Naming Convention",{"type":14,"tag":22,"props":959,"children":960},{},[961],{"type":20,"value":962},"Use clear, descriptive names:",{"type":14,"tag":89,"props":964,"children":966},{"code":965},"Good:\n- \"Customer_Churn_v2_GridSearch\"\n- \"NLP_Sentiment_Adam_LR0.001\"\n\nAvoid:\n- \"test1\", \"experiment2\", \"model_final\"\n",[967],{"type":14,"tag":96,"props":968,"children":969},{"__ignoreMap":7},[970],{"type":20,"value":965},{"type":14,"tag":35,"props":972,"children":974},{"id":973},"log-everything",[975],{"type":20,"value":976},"Log Everything",{"type":14,"tag":22,"props":978,"children":979},{},[980],{"type":20,"value":981},"Log parameters that might matter later:",{"type":14,"tag":89,"props":983,"children":985},{"code":984,"language":92,"meta":7,"className":93,"style":7},"exp.log_params({\n    'data_version': '2024_Q1',\n    'train_test_split': 0.8,\n    'random_seed': 42,\n    'data_augmentation': True\n})\n",[986],{"type":14,"tag":96,"props":987,"children":988},{"__ignoreMap":7},[989,996,1004,1012,1020,1028],{"type":14,"tag":100,"props":990,"children":991},{"class":102,"line":103},[992],{"type":14,"tag":100,"props":993,"children":994},{},[995],{"type":20,"value":330},{"type":14,"tag":100,"props":997,"children":998},{"class":102,"line":112},[999],{"type":14,"tag":100,"props":1000,"children":1001},{},[1002],{"type":20,"value":1003},"    'data_version': '2024_Q1',\n",{"type":14,"tag":100,"props":1005,"children":1006},{"class":102,"line":122},[1007],{"type":14,"tag":100,"props":1008,"children":1009},{},[1010],{"type":20,"value":1011},"    'train_test_split': 0.8,\n",{"type":14,"tag":100,"props":1013,"children":1014},{"class":102,"line":131},[1015],{"type":14,"tag":100,"props":1016,"children":1017},{},[1018],{"type":20,"value":1019},"    'random_seed': 42,\n",{"type":14,"tag":100,"props":1021,"children":1022},{"class":102,"line":140},[1023],{"type":14,"tag":100,"props":1024,"children":1025},{},[1026],{"type":20,"value":1027},"    'data_augmentation': True\n",{"type":14,"tag":100,"props":1029,"children":1030},{"class":102,"line":148},[1031],{"type":14,"tag":100,"props":1032,"children":1033},{},[1034],{"type":20,"value":378},{"type":14,"tag":35,"props":1036,"children":1038},{"id":1037},"organize-experiments",[1039],{"type":20,"value":1040},"Organize Experiments",{"type":14,"tag":22,"props":1042,"children":1043},{},[1044],{"type":20,"value":1045},"Group related experiments:",{"type":14,"tag":262,"props":1047,"children":1048},{},[1049,1054,1059],{"type":14,"tag":46,"props":1050,"children":1051},{},[1052],{"type":20,"value":1053},"Use experiment folders",{"type":14,"tag":46,"props":1055,"children":1056},{},[1057],{"type":20,"value":1058},"Add tags: #urgent, #production, #testing",{"type":14,"tag":46,"props":1060,"children":1061},{},[1062],{"type":20,"value":1063},"Write clear descriptions",{"type":14,"tag":28,"props":1065,"children":1067},{"id":1066},"common-workflows",[1068],{"type":20,"value":1069},"Common Workflows",{"type":14,"tag":35,"props":1071,"children":1073},{"id":1072},"hyperparameter-tuning",[1074],{"type":20,"value":1075},"Hyperparameter Tuning",{"type":14,"tag":42,"props":1077,"children":1078},{},[1079,1084,1107,1112],{"type":14,"tag":46,"props":1080,"children":1081},{},[1082],{"type":20,"value":1083},"Create experiment \"HPO_Search\"",{"type":14,"tag":46,"props":1085,"children":1086},{},[1087,1089],{"type":20,"value":1088},"For each parameter combination:\n",{"type":14,"tag":262,"props":1090,"children":1091},{},[1092,1097,1102],{"type":14,"tag":46,"props":1093,"children":1094},{},[1095],{"type":20,"value":1096},"Log parameters",{"type":14,"tag":46,"props":1098,"children":1099},{},[1100],{"type":20,"value":1101},"Train model",{"type":14,"tag":46,"props":1103,"children":1104},{},[1105],{"type":20,"value":1106},"Log metrics",{"type":14,"tag":46,"props":1108,"children":1109},{},[1110],{"type":20,"value":1111},"Compare all runs",{"type":14,"tag":46,"props":1113,"children":1114},{},[1115],{"type":20,"value":1116},"Select best parameters",{"type":14,"tag":35,"props":1118,"children":1120},{"id":1119},"model-comparison",[1121],{"type":20,"value":1122},"Model Comparison",{"type":14,"tag":42,"props":1124,"children":1125},{},[1126,1131,1136,1141],{"type":14,"tag":46,"props":1127,"children":1128},{},[1129],{"type":20,"value":1130},"Create separate experiments for each approach",{"type":14,"tag":46,"props":1132,"children":1133},{},[1134],{"type":20,"value":1135},"Train multiple models",{"type":14,"tag":46,"props":1137,"children":1138},{},[1139],{"type":20,"value":1140},"Compare side-by-side",{"type":14,"tag":46,"props":1142,"children":1143},{},[1144],{"type":20,"value":1145},"Choose best one",{"type":14,"tag":35,"props":1147,"children":1149},{"id":1148},"production-baseline",[1150],{"type":20,"value":1151},"Production Baseline",{"type":14,"tag":42,"props":1153,"children":1154},{},[1155,1160,1165,1170],{"type":14,"tag":46,"props":1156,"children":1157},{},[1158],{"type":20,"value":1159},"Create \"Production_Baseline\" experiment",{"type":14,"tag":46,"props":1161,"children":1162},{},[1163],{"type":20,"value":1164},"Log current production model metrics",{"type":14,"tag":46,"props":1166,"children":1167},{},[1168],{"type":20,"value":1169},"Compare new experiments against baseline",{"type":14,"tag":46,"props":1171,"children":1172},{},[1173],{"type":20,"value":1174},"Only promote if improvement is significant",{"type":14,"tag":28,"props":1176,"children":1178},{"id":1177},"troubleshooting",[1179],{"type":20,"value":1180},"Troubleshooting",{"type":14,"tag":35,"props":1182,"children":1184},{"id":1183},"metrics-not-appearing",[1185],{"type":20,"value":1186},"Metrics not appearing",{"type":14,"tag":262,"props":1188,"children":1189},{},[1190,1195,1200],{"type":14,"tag":46,"props":1191,"children":1192},{},[1193],{"type":20,"value":1194},"Ensure you're inside an active experiment",{"type":14,"tag":46,"props":1196,"children":1197},{},[1198],{"type":20,"value":1199},"Check metric names are spelled correctly",{"type":14,"tag":46,"props":1201,"children":1202},{},[1203],{"type":20,"value":1204},"Verify metric values are numbers",{"type":14,"tag":35,"props":1206,"children":1208},{"id":1207},"cant-find-my-run",[1209],{"type":20,"value":1210},"Can't find my run",{"type":14,"tag":262,"props":1212,"children":1213},{},[1214,1219,1224],{"type":14,"tag":46,"props":1215,"children":1216},{},[1217],{"type":20,"value":1218},"Try filtering by date range",{"type":14,"tag":46,"props":1220,"children":1221},{},[1222],{"type":20,"value":1223},"Search by parameter values",{"type":14,"tag":46,"props":1225,"children":1226},{},[1227],{"type":20,"value":1228},"Check if in a different experiment",{"type":14,"tag":35,"props":1230,"children":1232},{"id":1231},"export-data",[1233],{"type":20,"value":1234},"Export data",{"type":14,"tag":262,"props":1236,"children":1237},{},[1238,1243,1248],{"type":14,"tag":46,"props":1239,"children":1240},{},[1241],{"type":20,"value":1242},"Download as CSV for external analysis",{"type":14,"tag":46,"props":1244,"children":1245},{},[1246],{"type":20,"value":1247},"Export model files",{"type":14,"tag":46,"props":1249,"children":1250},{},[1251],{"type":20,"value":1252},"Generate reports",{"type":14,"tag":28,"props":1254,"children":1256},{"id":1255},"related-topics",[1257],{"type":20,"value":1258},"Related Topics",{"type":14,"tag":262,"props":1260,"children":1261},{},[1262,1274,1285],{"type":14,"tag":46,"props":1263,"children":1264},{},[1265,1272],{"type":14,"tag":1266,"props":1267,"children":1269},"a",{"href":1268},"\u002Fdocs\u002Ffeatures\u002Fnotebook",[1270],{"type":20,"value":1271},"Notebook",{"type":20,"value":1273}," - Run experiments here",{"type":14,"tag":46,"props":1275,"children":1276},{},[1277,1283],{"type":14,"tag":1266,"props":1278,"children":1280},{"href":1279},"\u002Fdocs\u002Ffeatures\u002Fai-assistant",[1281],{"type":20,"value":1282},"AI Assistant",{"type":20,"value":1284}," - Get model suggestions",{"type":14,"tag":46,"props":1286,"children":1287},{},[1288,1294],{"type":14,"tag":1266,"props":1289,"children":1291},{"href":1290},"\u002Fdocs\u002Ffeatures\u002Fapi-keys",[1292],{"type":20,"value":1293},"API Keys",{"type":20,"value":1295}," - Integrate with external ML tools",{"type":14,"tag":1297,"props":1298,"children":1299},"style",{},[1300],{"type":20,"value":1301},"html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}",{"title":7,"searchDepth":112,"depth":112,"links":1303},[1304,1308,1313,1318,1322,1326,1330,1335,1340,1345],{"id":30,"depth":112,"text":33,"children":1305},[1306,1307],{"id":37,"depth":122,"text":40},{"id":79,"depth":122,"text":82},{"id":246,"depth":112,"text":249,"children":1309},[1310,1311,1312],{"id":252,"depth":122,"text":255},{"id":307,"depth":122,"text":310},{"id":381,"depth":122,"text":384},{"id":485,"depth":112,"text":488,"children":1314},[1315,1316,1317],{"id":491,"depth":122,"text":494},{"id":525,"depth":122,"text":528},{"id":566,"depth":122,"text":569},{"id":620,"depth":112,"text":623,"children":1319},[1320,1321],{"id":626,"depth":122,"text":629},{"id":664,"depth":122,"text":667},{"id":698,"depth":112,"text":701,"children":1323},[1324,1325],{"id":704,"depth":122,"text":707},{"id":748,"depth":122,"text":751},{"id":837,"depth":112,"text":840,"children":1327},[1328,1329],{"id":843,"depth":122,"text":846},{"id":906,"depth":122,"text":909},{"id":948,"depth":112,"text":951,"children":1331},[1332,1333,1334],{"id":954,"depth":122,"text":957},{"id":973,"depth":122,"text":976},{"id":1037,"depth":122,"text":1040},{"id":1066,"depth":112,"text":1069,"children":1336},[1337,1338,1339],{"id":1072,"depth":122,"text":1075},{"id":1119,"depth":122,"text":1122},{"id":1148,"depth":122,"text":1151},{"id":1177,"depth":112,"text":1180,"children":1341},[1342,1343,1344],{"id":1183,"depth":122,"text":1186},{"id":1207,"depth":122,"text":1210},{"id":1231,"depth":122,"text":1234},{"id":1255,"depth":112,"text":1258},"markdown","content:docs:features:ml-experiments.md","content","docs\u002Ffeatures\u002Fml-experiments.md","docs\u002Ffeatures\u002Fml-experiments","md",1782233761656]