Training Report: test_fix

Basic Information

Model Type: wakeword
Version: 1.0.0
Author: Trixy ML Trainer
Created: 2025-10-01T11:11:12.180355+00:00
Last Modified: 2025-10-01T11:11:25.929936+00:00
Description: RepCNN wakeword detection model (standard)

Model Architecture

Architecture: RepCNN
Total Parameters: 2,586,371
Trainable Parameters: 2,586,371
Model Size: 9.88 MB
Layers: 45
Activations: ReLU

Training Results

Total Epochs: 1
Best Epoch: 1
Final Train Loss: 0.336808
Final Train Accuracy: 0.0000
Best Validation Loss: 0.332359
Best Validation Accuracy: 0.6572
Training Duration: 4.99 seconds

Dataset Information

Total Samples: 1,415
Training Samples: 989
Validation Samples: 283
Test Samples: 143
Number of Classes: 0

Hardware Information

Hostname: BPGamer
Platform: Linux 6.6.87.2-microsoft-standard-WSL2
CPU Cores: 32
CUDA Available: True
CUDA Version: 12.8
GPU Count: 1
GPUs: NVIDIA GeForce RTX 5070 Ti
Total Memory: 31.2 GB

Wakeword Information

Wakewords: custom, system_command, negative
Classes:
- 0: custom
- 1: system_command
- 2: negative
Detection Threshold: 0.5

Test Results

validation_loss: 0.419020
validation_time: 0.409613
num_samples: 283
metrics: {'accuracy': 0.6113074204946997, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(72), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(38), 'precision_Class_2': np.float64(0.6113074204946997), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7587719298245614), 'support_Class_2': np.int64(173), 'precision_macro': np.float64(0.2037691401648999), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25292397660818716), 'precision_weighted': np.float64(0.37369676235188354), 'recall_weighted': np.float64(0.6113074204946997), 'f1_weighted': np.float64(0.4638429111648379), 'confusion_matrix': [[0, 0, 72], [0, 0, 38], [0, 0, 173]], 'auc_roc_ovr': 0.3426738067277912, 'auc_roc_ovo': np.float64(0.37538380714148883), 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 72.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 38.0}, 'Class_2': {'precision': 0.6113074204946997, 'recall': 1.0, 'f1-score': 0.7587719298245614, 'support': 173.0}, 'accuracy': 0.6113074204946997, 'macro avg': {'precision': 0.2037691401648999, 'recall': 0.3333333333333333, 'f1-score': 0.25292397660818716, 'support': 283.0}, 'weighted avg': {'precision': 0.37369676235188354, 'recall': 0.6113074204946997, 'f1-score': 0.4638429111648379, 'support': 283.0}}, 'intra_class_similarity_mean': np.float32(0.9966352), 'intra_class_similarity_std': np.float32(0.0036274875), 'inter_class_similarity_mean': np.float32(0.99627626), 'inter_class_similarity_std': np.float32(0.0040334975), 'class_separation': np.float32(0.00035893917), 'eer': 1.0, 'eer_threshold': 0.0}
predictions: [2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2]
targets: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2]
probabilities: [[0.3047245442867279, 0.26112738251686096, 0.43414807319641113], [0.3107256293296814, 0.26373130083084106, 0.4255430996417999], [0.30283117294311523, 0.2591616213321686, 0.43800726532936096], [0.30804187059402466, 0.2740178406238556, 0.41794031858444214], [0.3071589171886444, 0.261407732963562, 0.4314333498477936], [0.3042905926704407, 0.26500752568244934, 0.43070188164711], [0.311048299074173, 0.26740482449531555, 0.42154690623283386], [0.3000188171863556, 0.25732743740081787, 0.44265374541282654], [0.3096528649330139, 0.26814961433410645, 0.42219749093055725], [0.3038494288921356, 0.2619953453540802, 0.43415528535842896], [0.3050396740436554, 0.2667369842529297, 0.4282233715057373], [0.3056211471557617, 0.2583440840244293, 0.43603476881980896], [0.3076876103878021, 0.26764312386512756, 0.4246692955493927], [0.3093443810939789, 0.26851367950439453, 0.4221419095993042], [0.3063105344772339, 0.2709181010723114, 0.4227713644504547], [0.3008154332637787, 0.2579774856567383, 0.4412071406841278], [0.30390340089797974, 0.2620357871055603, 0.43406084179878235], [0.3084034323692322, 0.26444393396377563, 0.4271526336669922], [0.30936136841773987, 0.26972755789756775, 0.4209110736846924], [0.307151734828949, 0.26275962591171265, 0.4300886392593384], [0.3052944839000702, 0.25833380222320557, 0.43637165427207947], [0.3029278814792633, 0.26060178875923157, 0.4364703297615051], [0.30678674578666687, 0.2699756324291229, 0.4232375919818878], [0.31087440252304077, 0.2726493179798126, 0.4164762496948242], [0.31243324279785156, 0.27762770652770996, 0.4099389910697937], [0.3020888864994049, 0.2595059275627136, 0.4384051561355591], [0.3144093453884125, 0.2779814898967743, 0.4076091945171356], [0.30793362855911255, 0.26493123173713684, 0.4271351993083954], [0.30788859724998474, 0.2666572034358978, 0.42545419931411743], [0.3009384572505951, 0.2575550079345703, 0.4415065348148346], [0.31380757689476013, 0.2812308371067047, 0.40496161580085754], [0.3030255138874054, 0.2576964497566223, 0.43927812576293945], [0.30721452832221985, 0.27183300256729126, 0.4209524095058441], [0.33389079570770264, 0.2673867344856262, 0.39872246980667114], [0.3066445589065552, 0.26717981696128845, 0.426175594329834], [0.30080413818359375, 0.2601699233055115, 0.43902596831321716], [0.30466076731681824, 0.2596171200275421, 0.4357220530509949], [0.30818596482276917, 0.2663038671016693, 0.4255101680755615], [0.2996703088283539, 0.2570069134235382, 0.4433227479457855], [0.30565252900123596, 0.2614510655403137, 0.43289637565612793], [0.3040764331817627, 0.27134430408477783, 0.42457935214042664], [0.2996430993080139, 0.2578068971633911, 0.44255000352859497], [0.31018564105033875, 0.2754320204257965, 0.41438233852386475], [0.3120267987251282, 0.27765750885009766, 0.41031572222709656], [0.30714473128318787, 0.26837214827537537, 0.42448312044143677], [0.30143409967422485, 0.2578763961791992, 0.44068947434425354], [0.3094130754470825, 0.25829169154167175, 0.43229514360427856], [0.30954259634017944, 0.26739996671676636, 0.423057496547699], [0.3000434339046478, 0.25652214884757996, 0.4434344470500946], [0.3035999536514282, 0.26476046442985535, 0.43163958191871643], [0.30559831857681274, 0.2583524286746979, 0.4360492527484894], [0.30372750759124756, 0.2619587779045105, 0.43431371450424194], [0.31629645824432373, 0.27449703216552734, 0.40920644998550415], [0.3130092918872833, 0.2728237807750702, 0.4141669273376465], [0.30586376786231995, 0.2589760720729828, 0.4351601302623749], [0.3084414303302765, 0.26825225353240967, 0.42330631613731384], [0.3094329237937927, 0.2670883536338806, 0.4234786927700043], [0.30776122212409973, 0.26845574378967285, 0.4237830340862274], [0.30935418605804443, 0.27206164598464966, 0.4185841977596283], [0.30547603964805603, 0.26648011803627014, 0.42804375290870667], [0.3166373670101166, 0.2673494517803192, 0.4160131812095642], [0.30343228578567505, 0.2623022496700287, 0.43426552414894104], [0.30195605754852295, 0.2582823634147644, 0.43976157903671265], [0.3257077634334564, 0.2796502411365509, 0.39464202523231506], [0.30185359716415405, 0.2588660717010498, 0.43928030133247375], [0.30393776297569275, 0.25885942578315735, 0.43720272183418274], [0.3246135711669922, 0.26291128993034363, 0.4124751687049866], [0.31989583373069763, 0.2797277867794037, 0.40037640929222107], [0.31644389033317566, 0.26575127243995667, 0.4178048372268677], [0.29990220069885254, 0.2567955255508423, 0.4433022141456604], [0.30618157982826233, 0.2644655406475067, 0.42935290932655334], [0.340969055891037, 0.26598668098449707, 0.3930443227291107], [0.3096659481525421, 0.27248579263687134, 0.41784825921058655], [0.3112565875053406, 0.272981733083725, 0.41576170921325684], [0.30995073914527893, 0.2690833508968353, 0.42096588015556335], [0.299455851316452, 0.25807780027389526, 0.44246625900268555], [0.3146135210990906, 0.2765198051929474, 0.40886667370796204], [0.30717554688453674, 0.26836514472961426, 0.4244593381881714], [0.3076019287109375, 0.2758239805698395, 0.4165741205215454], [0.31834906339645386, 0.2889685332775116, 0.39268240332603455], [0.2996683716773987, 0.25705716013908386, 0.44327449798583984], [0.30271053314208984, 0.2563190162181854, 0.44097045063972473], [0.3083384037017822, 0.26651501655578613, 0.42514654994010925], [0.307513952255249, 0.26438257098197937, 0.428103506565094], [0.3033374845981598, 0.2592509686946869, 0.43741151690483093], [0.3187502324581146, 0.26809626817703247, 0.41315343976020813], [0.3100959360599518, 0.2658807039260864, 0.424023300409317], [0.3001222014427185, 0.2571133077144623, 0.4427645206451416], [0.31058406829833984, 0.2676277756690979, 0.42178818583488464], [0.30451396107673645, 0.270537406206131, 0.42494866251945496], [0.30533090233802795, 0.26968181133270264, 0.4249872863292694], [0.30559784173965454, 0.26200374960899353, 0.43239837884902954], [0.30346107482910156, 0.258309930562973, 0.438228964805603], [0.3153536021709442, 0.2629656195640564, 0.421680748462677], [0.31310322880744934, 0.2705264389514923, 0.41637033224105835], [0.30561190843582153, 0.268335223197937, 0.42605283856391907], [0.31430521607398987, 0.2783298194408417, 0.40736502408981323], [0.307454377412796, 0.26184728741645813, 0.43069836497306824], [0.3110523223876953, 0.275924950838089, 0.4130226969718933], [0.30274584889411926, 0.2580240070819855, 0.43923020362854004], [0.30142804980278015, 0.25920864939689636, 0.43936336040496826], [0.305351585149765, 0.25897151231765747, 0.4356768727302551], [0.30801352858543396, 0.26980578899383545, 0.4221806824207306], [0.3109542429447174, 0.2809315323829651, 0.4081141948699951], [0.30786070227622986, 0.26352810859680176, 0.42861124873161316], [0.3084365427494049, 0.26345428824424744, 0.42810919880867004], [0.3081333041191101, 0.2601141631603241, 0.431752473115921], [0.3061281740665436, 0.2662574350833893, 0.4276144206523895], [0.30510973930358887, 0.2591572105884552, 0.4357330799102783], [0.30337074398994446, 0.26704198122024536, 0.42958730459213257], [0.3074008822441101, 0.2649642527103424, 0.4276348352432251], [0.3208828270435333, 0.26952266693115234, 0.40959447622299194], [0.3198268413543701, 0.26822924613952637, 0.4119439423084259], [0.32492929697036743, 0.26740679144859314, 0.40766391158103943], [0.3175393342971802, 0.2642245888710022, 0.41823598742485046], [0.30133113265037537, 0.26314324140548706, 0.43552568554878235], [0.3131498098373413, 0.2733291685581207, 0.41352102160453796], [0.3051753044128418, 0.26650726795196533, 0.42831742763519287], [0.3060658276081085, 0.26498571038246155, 0.42894846200942993], [0.30555811524391174, 0.26506316661834717, 0.4293786883354187], [0.32373350858688354, 0.2664750814437866, 0.4097914397716522], [0.3079107105731964, 0.27789580821990967, 0.4141935110092163], [0.33040598034858704, 0.2725030481815338, 0.39709097146987915], [0.31000542640686035, 0.26513737440109253, 0.4248572289943695], [0.3193601071834564, 0.27759313583374023, 0.40304678678512573], [0.30346301198005676, 0.26198577880859375, 0.4345512092113495], [0.3182131052017212, 0.27128663659095764, 0.4105002284049988], [0.33416879177093506, 0.26681774854660034, 0.399013489484787], [0.3068411350250244, 0.26709726452827454, 0.4260616600513458], [0.33031827211380005, 0.26937243342399597, 0.400309294462204], [0.31086236238479614, 0.26886463165283203, 0.42027297616004944], [0.32209450006484985, 0.2680931091308594, 0.4098123610019684], [0.30548107624053955, 0.26380670070648193, 0.4307122230529785], [0.3084432780742645, 0.2677218019962311, 0.423834890127182], [0.3010537922382355, 0.2612248957157135, 0.43772128224372864], [0.30713579058647156, 0.26537492871284485, 0.4274892508983612], [0.3042013645172119, 0.26403331756591797, 0.4317653179168701], [0.3198586106300354, 0.26626652479171753, 0.41387486457824707], [0.3121587634086609, 0.2680797278881073, 0.4197614789009094], [0.3177639842033386, 0.26640090346336365, 0.4158351719379425], [0.3151736557483673, 0.27653029561042786, 0.40829604864120483], [0.3068219721317291, 0.2701321244239807, 0.4230458736419678], [0.29978591203689575, 0.25761035084724426, 0.44260379672050476], [0.3168798089027405, 0.26532384753227234, 0.4177963137626648], [0.3228106200695038, 0.2688470184803009, 0.4083423912525177], [0.3155139982700348, 0.269830584526062, 0.4146554470062256], [0.3345581591129303, 0.2702496349811554, 0.3951922059059143], [0.30997517704963684, 0.26620715856552124, 0.42381763458251953], [0.31997931003570557, 0.26554247736930847, 0.41447821259498596], [0.30899104475975037, 0.26682543754577637, 0.42418354749679565], [0.3189978301525116, 0.2695659101009369, 0.4114362895488739], [0.3089173436164856, 0.26408037543296814, 0.4270022511482239], [0.32114797830581665, 0.26511695981025696, 0.4137350916862488], [0.3100159764289856, 0.26602494716644287, 0.42395907640457153], [0.3138783574104309, 0.26606836915016174, 0.4200532138347626], [0.303737074136734, 0.2616591453552246, 0.434603750705719], [0.3019508123397827, 0.25929689407348633, 0.43875229358673096], [0.31134533882141113, 0.26977774500846863, 0.41887688636779785], [0.30687305331230164, 0.2707437574863434, 0.422383189201355], [0.34270501136779785, 0.2665007412433624, 0.3907942473888397], [0.3153343200683594, 0.27163225412368774, 0.41303345561027527], [0.31463271379470825, 0.2722056806087494, 0.41316157579421997], [0.32545626163482666, 0.26811477541923523, 0.4064289331436157], [0.32116496562957764, 0.26902467012405396, 0.409810334444046], [0.30862554907798767, 0.2761959135532379, 0.41517847776412964], [0.3131198585033417, 0.2681114077568054, 0.4187687337398529], [0.3090015947818756, 0.26585593819618225, 0.4251424968242645], [0.30941733717918396, 0.2722790539264679, 0.41830360889434814], [0.32524800300598145, 0.31025683879852295, 0.3644952178001404], [0.3204421401023865, 0.2621908485889435, 0.41736701130867004], [0.31723934412002563, 0.2660518288612366, 0.4167087972164154], [0.3137023150920868, 0.2669464647769928, 0.419351190328598], [0.3085237443447113, 0.2706618309020996, 0.4208144247531891], [0.3306116461753845, 0.2674080729484558, 0.4019802510738373], [0.30769285559654236, 0.27050745487213135, 0.4217996895313263], [0.3192599415779114, 0.26768434047698975, 0.4130557179450989], [0.3279280364513397, 0.2898523509502411, 0.3822196424007416], [0.31427469849586487, 0.26752525568008423, 0.4182000458240509], [0.3049515187740326, 0.26824402809143066, 0.42680448293685913], [0.3182542026042938, 0.2647026479244232, 0.41704311966896057], [0.320796400308609, 0.26930439472198486, 0.40989917516708374], [0.31003907322883606, 0.2689981460571289, 0.42096278071403503], [0.30053064227104187, 0.2574267089366913, 0.44204264879226685], [0.309773325920105, 0.2624090313911438, 0.4278176426887512], [0.3051762282848358, 0.2613767981529236, 0.4334470331668854], [0.30893170833587646, 0.27276545763015747, 0.41830286383628845], [0.3151955008506775, 0.2706984281539917, 0.4141061007976532], [0.30675384402275085, 0.265058308839798, 0.42818787693977356], [0.3159369230270386, 0.2697056531906128, 0.41435742378234863], [0.32119297981262207, 0.2688104212284088, 0.40999656915664673], [0.30834221839904785, 0.2635388672351837, 0.42811891436576843], [0.31426554918289185, 0.2688785791397095, 0.41685590147972107], [0.3072235882282257, 0.26227009296417236, 0.4305063784122467], [0.3114877939224243, 0.268349826335907, 0.4201623797416687], [0.3083108365535736, 0.2696421146392822, 0.4220470190048218], [0.3107966184616089, 0.2716081440448761, 0.4175952672958374], [0.3076229393482208, 0.2625507116317749, 0.4298263192176819], [0.3206953704357147, 0.2852208912372589, 0.39408376812934875], [0.3122718036174774, 0.26909151673316956, 0.4186367392539978], [0.3280956745147705, 0.26465827226638794, 0.4072459936141968], [0.3329694867134094, 0.27028143405914307, 0.3967491388320923], [0.3090248107910156, 0.2741168439388275, 0.41685837507247925], [0.3270832300186157, 0.26880744099617004, 0.40410932898521423], [0.3064338266849518, 0.2635648250579834, 0.43000128865242004], [0.3313497006893158, 0.26440903544425964, 0.4042412340641022], [0.32109248638153076, 0.2668223977088928, 0.4120851159095764], [0.31685301661491394, 0.2697656452655792, 0.4133813977241516], [0.3034021556377411, 0.2651430070400238, 0.4314548373222351], [0.3222673237323761, 0.2688958942890167, 0.40883687138557434], [0.3060990273952484, 0.26927563548088074, 0.42462533712387085], [0.3063269853591919, 0.2604864835739136, 0.43318650126457214], [0.31763726472854614, 0.26761576533317566, 0.4147469997406006], [0.3176720142364502, 0.26601091027259827, 0.41631701588630676], [0.3179763853549957, 0.27258238196372986, 0.4094412624835968], [0.3210701048374176, 0.27120861411094666, 0.40772125124931335], [0.3281856179237366, 0.2884422242641449, 0.3833721876144409], [0.31036993861198425, 0.2710619866847992, 0.41856807470321655], [0.3187180161476135, 0.26549261808395386, 0.4157893657684326], [0.33028680086135864, 0.2708090543746948, 0.39890414476394653], [0.30719152092933655, 0.25997796654701233, 0.43283048272132874], [0.3126133382320404, 0.2687758207321167, 0.4186108410358429], [0.3068144619464874, 0.2696222960948944, 0.4235631823539734], [0.3108985722064972, 0.27128705382347107, 0.4178144335746765], [0.29997938871383667, 0.25656256079673767, 0.44345808029174805], [0.3003670871257782, 0.2582642436027527, 0.44136863946914673], [0.3108135461807251, 0.27191656827926636, 0.41726991534233093], [0.30872297286987305, 0.2794252038002014, 0.41185182332992554], [0.3066796660423279, 0.2676856517791748, 0.4256347119808197], [0.3073292374610901, 0.2729242444038391, 0.4197465479373932], [0.30846819281578064, 0.2697969377040863, 0.42173483967781067], [0.31137174367904663, 0.26923510432243347, 0.4193931519985199], [0.30530741810798645, 0.26730382442474365, 0.4273887276649475], [0.3005823493003845, 0.25841978192329407, 0.4409978985786438], [0.30856990814208984, 0.27213215827941895, 0.4192979335784912], [0.30514633655548096, 0.26236042380332947, 0.4324932098388672], [0.31306883692741394, 0.2669556438922882, 0.41997548937797546], [0.31673455238342285, 0.26417380571365356, 0.41909170150756836], [0.3211798667907715, 0.2616962492465973, 0.4171238839626312], [0.31505870819091797, 0.26620668172836304, 0.4187346398830414], [0.31334519386291504, 0.2728043794631958, 0.4138503670692444], [0.3225318491458893, 0.2735983729362488, 0.40386977791786194], [0.3151249289512634, 0.2667115330696106, 0.4181634783744812], [0.33289623260498047, 0.2659495174884796, 0.4011542797088623], [0.31651806831359863, 0.2641047239303589, 0.41937729716300964], [0.30999335646629333, 0.2655988037586212, 0.42440786957740784], [0.31355947256088257, 0.27676287293434143, 0.409677654504776], [0.31289350986480713, 0.2650105953216553, 0.4220958948135376], [0.3177407681941986, 0.2690183222293854, 0.41324087977409363], [0.30274391174316406, 0.25632941722869873, 0.440926730632782], [0.31470656394958496, 0.2671387195587158, 0.41815468668937683], [0.3222196698188782, 0.2718784511089325, 0.40590187907218933], [0.3300307095050812, 0.26775312423706055, 0.40221619606018066], [0.31588220596313477, 0.2652113139629364, 0.41890648007392883], [0.32235217094421387, 0.26617223024368286, 0.41147559881210327], [0.3106209933757782, 0.278135746717453, 0.4112432897090912], [0.30084657669067383, 0.2608465254306793, 0.43830692768096924], [0.32217973470687866, 0.2682374119758606, 0.40958288311958313], [0.3147556781768799, 0.2773243188858032, 0.4079200029373169], [0.3046095669269562, 0.27008068561553955, 0.4253097176551819], [0.33431512117385864, 0.2693495750427246, 0.396335244178772], [0.32260066270828247, 0.272585928440094, 0.4048134684562683], [0.3124454915523529, 0.26861754059791565, 0.41893696784973145], [0.31366994976997375, 0.2716602683067322, 0.41466978192329407], [0.33406293392181396, 0.2688366770744324, 0.39710041880607605], [0.3136119544506073, 0.26512211561203003, 0.42126592993736267], [0.30579474568367004, 0.2713998854160309, 0.42280545830726624], [0.3306726813316345, 0.26835232973098755, 0.4009750187397003], [0.31333234906196594, 0.26682212948799133, 0.41984546184539795], [0.3009106516838074, 0.2606675326824188, 0.4384218156337738], [0.31622207164764404, 0.271098792552948, 0.41267910599708557], [0.3255620002746582, 0.2617549002170563, 0.41268303990364075], [0.3201005458831787, 0.267785906791687, 0.41211357712745667], [0.31202685832977295, 0.26540428400039673, 0.4225688576698303], [0.31264597177505493, 0.2694295346736908, 0.41792452335357666], [0.31757882237434387, 0.27221763134002686, 0.41020357608795166], [0.30967462062835693, 0.2705460786819458, 0.4197792708873749], [0.31621405482292175, 0.26630789041519165, 0.4174780547618866], [0.32906559109687805, 0.2676960527896881, 0.40323835611343384], [0.3363478481769562, 0.26117804646492004, 0.40247413516044617], [0.3045779764652252, 0.26520171761512756, 0.4302203357219696], [0.3146213889122009, 0.267109215259552, 0.4182693660259247], [0.3178431987762451, 0.26277104020118713, 0.41938579082489014], [0.3336517810821533, 0.26526448130607605, 0.40108373761177063]]
standard_test: {'validation_loss': 0.42087624139255947, 'validation_time': 0.3021385669708252, 'num_samples': 143, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'auc_roc_ovr': 0.3056155404961261, 'auc_roc_ovo': np.float64(0.33144351247799525), 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}, 'intra_class_similarity_mean': np.float32(0.9967397), 'intra_class_similarity_std': np.float32(0.0035206827), 'inter_class_similarity_mean': np.float32(0.996594), 'inter_class_similarity_std': np.float32(0.0038162724), 'class_separation': np.float32(0.00014567375), 'eer': 1.0, 'eer_threshold': 0.0}, 'predictions': [2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2], 'targets': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2], 'probabilities': [[0.3017352521419525, 0.25861093401908875, 0.43965378403663635], [0.3071253299713135, 0.26695889234542847, 0.42591577768325806], [0.30088117718696594, 0.258255273103714, 0.4408635199069977], [0.30493786931037903, 0.265408992767334, 0.429653137922287], [0.30595049262046814, 0.26388823986053467, 0.4301612973213196], [0.30005258321762085, 0.25928592681884766, 0.4406615197658539], [0.3074952960014343, 0.2583010196685791, 0.4342036545276642], [0.30468040704727173, 0.2692243456840515, 0.42609524726867676], [0.3106515109539032, 0.2732471227645874, 0.4161013662815094], [0.30596408247947693, 0.27061939239501953, 0.42341652512550354], [0.3081496059894562, 0.27141958475112915, 0.4204307794570923], [0.309489905834198, 0.2727978825569153, 0.4177122116088867], [0.30941447615623474, 0.2655227780342102, 0.4250626862049103], [0.299723356962204, 0.26192235946655273, 0.4383542835712433], [0.3126049339771271, 0.26947644352912903, 0.4179186522960663], [0.30847713351249695, 0.2836199402809143, 0.40790289640426636], [0.30621084570884705, 0.2666028141975403, 0.4271863102912903], [0.3341320753097534, 0.26722270250320435, 0.39864519238471985], [0.3058941662311554, 0.2701980769634247, 0.4239077866077423], [0.31238895654678345, 0.2695004343986511, 0.41811057925224304], [0.3075244128704071, 0.26631149649620056, 0.42616409063339233], [0.31097495555877686, 0.2725098431110382, 0.41651520133018494], [0.3047417998313904, 0.2613122761249542, 0.4339459240436554], [0.30475643277168274, 0.2661256790161133, 0.4291178286075592], [0.31143718957901, 0.2728448212146759, 0.4157179594039917], [0.303751140832901, 0.26473817229270935, 0.4315106272697449], [0.305176705121994, 0.2664410471916199, 0.4283822178840637], [0.3051299750804901, 0.2580701410770416, 0.43679991364479065], [0.3008837401866913, 0.2566692531108856, 0.4424469769001007], [0.2989761233329773, 0.25658467411994934, 0.44443920254707336], [0.3014707863330841, 0.25677812099456787, 0.4417511224746704], [0.3064368665218353, 0.26711416244506836, 0.4264489710330963], [0.3003121614456177, 0.259091317653656, 0.4405965209007263], [0.31423866748809814, 0.2704278826713562, 0.41533342003822327], [0.30345165729522705, 0.2641698718070984, 0.4323784410953522], [0.3055073618888855, 0.2791942059993744, 0.4152984321117401], [0.3035268783569336, 0.2599391043186188, 0.43653401732444763], [0.3064519464969635, 0.2568172216415405, 0.4367308020591736], [0.3024507761001587, 0.260928213596344, 0.4366210103034973], [0.3087906837463379, 0.2686595618724823, 0.4225497841835022], [0.29877930879592896, 0.25861743092536926, 0.4426032602787018], [0.30679619312286377, 0.2640116512775421, 0.42919212579727173], [0.3145713210105896, 0.2704143524169922, 0.4150143265724182], [0.30401745438575745, 0.2598167061805725, 0.43616586923599243], [0.3116820752620697, 0.26032379269599915, 0.42799413204193115], [0.3141288459300995, 0.2860267162322998, 0.3998444080352783], [0.3090631067752838, 0.2697979211807251, 0.4211390018463135], [0.3107915222644806, 0.26479706168174744, 0.424411416053772], [0.31453830003738403, 0.2799464762210846, 0.405515193939209], [0.3107013404369354, 0.26218560338020325, 0.42711305618286133], [0.30510690808296204, 0.2595248818397522, 0.4353681802749634], [0.3059536814689636, 0.26340988278388977, 0.4306364357471466], [0.3063610792160034, 0.2675130367279053, 0.4261259138584137], [0.30417686700820923, 0.2625185251235962, 0.4333045482635498], [0.3083677291870117, 0.267241507768631, 0.4243907928466797], [0.29894357919692993, 0.25787675380706787, 0.4431796371936798], [0.30890771746635437, 0.2678249180316925, 0.4232673645019531], [0.31565263867378235, 0.26902711391448975, 0.4153202176094055], [0.3022858798503876, 0.26295027136802673, 0.4347638189792633], [0.32190582156181335, 0.2673308849334717, 0.4107632637023926], [0.3114017844200134, 0.2662045359611511, 0.42239370942115784], [0.29948338866233826, 0.25834915041923523, 0.4421674311161041], [0.3224128484725952, 0.2716115415096283, 0.4059756100177765], [0.31155240535736084, 0.2761816084384918, 0.4122660458087921], [0.30895671248435974, 0.27538400888442993, 0.41565924882888794], [0.31960567831993103, 0.2635710835456848, 0.41682320833206177], [0.3316591680049896, 0.2704908549785614, 0.3978499174118042], [0.3100110590457916, 0.2691488265991211, 0.42084014415740967], [0.32667988538742065, 0.26980021595954895, 0.4035199284553528], [0.29896917939186096, 0.25767672061920166, 0.44335412979125977], [0.3041950464248657, 0.2614487409591675, 0.4343562126159668], [0.3173302710056305, 0.26905569434165955, 0.4136140048503876], [0.3045911192893982, 0.263355553150177, 0.4320533275604248], [0.3229025900363922, 0.2686153054237366, 0.40848204493522644], [0.3160848319530487, 0.2666145861148834, 0.4173005521297455], [0.31643083691596985, 0.26921766996383667, 0.41435152292251587], [0.31025364995002747, 0.2656334638595581, 0.42411285638809204], [0.3109579086303711, 0.2686743438243866, 0.4203677773475647], [0.3213699162006378, 0.26790374517440796, 0.41072627902030945], [0.30661746859550476, 0.26197585463523865, 0.43140673637390137], [0.31542712450027466, 0.276903361082077, 0.40766945481300354], [0.32134464383125305, 0.2693459391593933, 0.409309446811676], [0.30215755105018616, 0.25799718499183655, 0.4398452937602997], [0.3104988932609558, 0.2708398699760437, 0.4186612665653229], [0.33934712409973145, 0.26337605714797974, 0.3972768485546112], [0.3079881966114044, 0.2685640752315521, 0.42344778776168823], [0.3197750151157379, 0.2626304626464844, 0.4175945818424225], [0.30151882767677307, 0.259518563747406, 0.43896257877349854], [0.31453943252563477, 0.2695581912994385, 0.41590234637260437], [0.31866052746772766, 0.2653796970844269, 0.41595980525016785], [0.3186214864253998, 0.27180975675582886, 0.409568727016449], [0.32090482115745544, 0.26685458421707153, 0.4122406244277954], [0.3108522891998291, 0.2705916166305542, 0.4185561239719391], [0.3081262409687042, 0.26444652676582336, 0.4274272620677948], [0.3080379366874695, 0.2639397084712982, 0.42802226543426514], [0.30787572264671326, 0.2648699879646301, 0.42725419998168945], [0.30737248063087463, 0.26898443698883057, 0.4236431419849396], [0.3038056194782257, 0.26216933131217957, 0.43402501940727234], [0.3060416281223297, 0.26243627071380615, 0.43152207136154175], [0.30612242221832275, 0.2614396810531616, 0.43243786692619324], [0.3029974400997162, 0.2671559453010559, 0.4298466145992279], [0.32055526971817017, 0.2658596634864807, 0.4135851263999939], [0.3099210262298584, 0.27424436807632446, 0.4158346652984619], [0.33067068457603455, 0.2669003903865814, 0.40242883563041687], [0.32063478231430054, 0.2720324397087097, 0.40733274817466736], [0.33818840980529785, 0.26688531041145325, 0.3949262797832489], [0.31758084893226624, 0.26856473088264465, 0.4138544499874115], [0.31539425253868103, 0.2661346197128296, 0.418471097946167], [0.309383362531662, 0.27098485827445984, 0.4196317791938782], [0.33442530035972595, 0.2678622603416443, 0.39771243929862976], [0.3186868131160736, 0.2704772353172302, 0.41083601117134094], [0.3097934126853943, 0.26353275775909424, 0.4266738295555115], [0.3062141537666321, 0.26958924531936646, 0.42419660091400146], [0.3242741823196411, 0.26146888732910156, 0.41425690054893494], [0.31226709485054016, 0.2678442895412445, 0.41988861560821533], [0.3202114403247833, 0.2677757441997528, 0.41201281547546387], [0.3116990923881531, 0.2722617983818054, 0.4160391688346863], [0.3181087374687195, 0.2726636230945587, 0.4092276096343994], [0.31220734119415283, 0.26952192187309265, 0.4182707965373993], [0.3075433373451233, 0.2566990554332733, 0.4357575476169586], [0.2997368276119232, 0.2582499086856842, 0.4420132637023926], [0.30358558893203735, 0.26294657588005066, 0.43346792459487915], [0.3100058138370514, 0.26688897609710693, 0.42310523986816406], [0.31653860211372375, 0.2672443687915802, 0.41621702909469604], [0.3073798418045044, 0.26265496015548706, 0.42996513843536377], [0.31256118416786194, 0.26791274547576904, 0.41952604055404663], [0.31627511978149414, 0.2693970501422882, 0.41432780027389526], [0.3170633614063263, 0.26907283067703247, 0.41386377811431885], [0.3047417998313904, 0.26213112473487854, 0.43312710523605347], [0.30497539043426514, 0.2651849091053009, 0.42983970046043396], [0.3111489713191986, 0.2696484327316284, 0.4192025661468506], [0.317131370306015, 0.2656865119934082, 0.4171820878982544], [0.30900484323501587, 0.26348960399627686, 0.4275055527687073], [0.3067573308944702, 0.2664094865322113, 0.4268331825733185], [0.32191216945648193, 0.2692750096321106, 0.40881288051605225], [0.3204025328159332, 0.26899418234825134, 0.41060328483581543], [0.32075029611587524, 0.2661472260951996, 0.41310250759124756], [0.3109661638736725, 0.26691409945487976, 0.42211973667144775], [0.32554829120635986, 0.2670576870441437, 0.4073939919471741], [0.3064644932746887, 0.26974165439605713, 0.42379385232925415], [0.3289620280265808, 0.3048240840435028, 0.3662138879299164], [0.3118077516555786, 0.2755418121814728, 0.4126504361629486], [0.31023815274238586, 0.2705705761909485, 0.41919130086898804]], 'test_type': 'standard'}
augmentation_robustness: {'noise_light': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'noise_moderate': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'noise_heavy': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'speed_light': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'speed_moderate': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'volume_light': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'volume_moderate': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}, 'combined_light': {'loss': 0.42087624139255947, 'accuracy': 0.6083916083916084, 'metrics': {'accuracy': 0.6083916083916084, 'precision_Class_0': np.float64(0.0), 'recall_Class_0': np.float64(0.0), 'f1_Class_0': np.float64(0.0), 'support_Class_0': np.int64(37), 'precision_Class_1': np.float64(0.0), 'recall_Class_1': np.float64(0.0), 'f1_Class_1': np.float64(0.0), 'support_Class_1': np.int64(19), 'precision_Class_2': np.float64(0.6083916083916084), 'recall_Class_2': np.float64(1.0), 'f1_Class_2': np.float64(0.7565217391304347), 'support_Class_2': np.int64(87), 'precision_macro': np.float64(0.2027972027972028), 'recall_macro': np.float64(0.3333333333333333), 'f1_macro': np.float64(0.25217391304347825), 'precision_weighted': np.float64(0.3701403491613282), 'recall_weighted': np.float64(0.6083916083916084), 'f1_weighted': np.float64(0.46026147765278197), 'confusion_matrix': [[0, 0, 37], [0, 0, 19], [0, 0, 87]], 'classification_report': {'Class_0': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 37.0}, 'Class_1': {'precision': 0.0, 'recall': 0.0, 'f1-score': 0.0, 'support': 19.0}, 'Class_2': {'precision': 0.6083916083916084, 'recall': 1.0, 'f1-score': 0.7565217391304347, 'support': 87.0}, 'accuracy': 0.6083916083916084, 'macro avg': {'precision': 0.2027972027972028, 'recall': 0.3333333333333333, 'f1-score': 0.25217391304347825, 'support': 143.0}, 'weighted avg': {'precision': 0.3701403491613282, 'recall': 0.6083916083916084, 'f1-score': 0.46026147765278197, 'support': 143.0}}}}}
performance_profile: {'inference_performance': {'mean_inference_time': np.float64(0.0007604619610356167), 'std_inference_time': np.float64(0.0002851131165727356), 'min_inference_time': np.float64(0.0005847840220667422), 'max_inference_time': np.float64(0.0015751940081827343), 'median_inference_time': np.float64(0.0006416804972104728), 'throughput_samples_per_second': np.float64(1314.9901655017356)}, 'memory_usage': {'initial_memory_mb': 66.53759765625, 'peak_memory_mb': 74.88720703125, 'final_memory_mb': 66.5380859375, 'memory_increase_mb': 8.349609375}, 'model_parameters': {'total_parameters': 2586371, 'trainable_parameters': 2586371, 'non_trainable_parameters': 0}, 'model_size': {'total_size_bytes': 10360412, 'total_size_mb': 9.88045883178711, 'parameter_size_mb': 9.866222381591797, 'buffer_size_mb': 0.0142364501953125}, 'batch_performance': {'batch_1': {'inference_time': np.float64(0.0006412367452867329), 'throughput': np.float64(1559.4864258018213)}, 'batch_8': {'inference_time': np.float64(0.0015252916957251728), 'throughput': np.float64(5244.898416755978)}, 'batch_16': {'inference_time': np.float64(0.007292331804637797), 'throughput': np.float64(2194.085572165583)}}}
robustness_tests: {'confidence_thresholds': {}, 'input_corruptions': {'zero_out_10': {'loss': 0.4230751593907674, 'accuracy': np.float64(0.6083916083916084)}, 'zero_out_25': {'loss': 0.42847098575698006, 'accuracy': np.float64(0.6083916083916084)}, 'gaussian_noise_01': {'loss': 0.4207393493917253, 'accuracy': np.float64(0.6083916083916084)}, 'gaussian_noise_02': {'loss': 0.4201605220635732, 'accuracy': np.float64(0.6083916083916084)}, 'dropout_10': {'loss': 0.4190783103307088, 'accuracy': np.float64(0.6083916083916084)}, 'dropout_25': {'loss': 0.4176873068014781, 'accuracy': np.float64(0.6083916083916084)}}, 'adversarial_robustness': {'epsilon_0.01': {'loss': 0.43043965101242065, 'accuracy': np.float64(0.6083916083916084)}, 'epsilon_0.05': {'loss': 0.4658129612604777, 'accuracy': np.float64(0.6083916083916084)}, 'epsilon_0.1': {'loss': 0.504215892818239, 'accuracy': np.float64(0.6083916083916084)}, 'epsilon_0.2': {'loss': 0.5654362936814626, 'accuracy': np.float64(0.43356643356643354)}}}

Wakeword Detection Specific Results

Classification Performance

Confusion Matrix

Actual\Predicted    custom system_c negative 
      custom         0        0       37 
system_comma         0        0       19 
    negative         0        0       87

Robustness Tests

Input Corruption Robustness

zero_out_10: 0.6084 accuracy
zero_out_25: 0.6084 accuracy
gaussian_noise_01: 0.6084 accuracy
gaussian_noise_02: 0.6084 accuracy
dropout_10: 0.6084 accuracy
dropout_25: 0.6084 accuracy

Performance Profile

Inference time: 0.76 ms
Throughput: 1315.0 samples/sec
Model size: 9.88 MB

Deployment Recommendations

Detection Threshold

Recommended threshold: 0.5
Monitor false positive/negative rates in production
Consider threshold adjustment based on use case requirements

Hardware Requirements

Suitable for real-time processing on most devices

Model Optimization

Consider using lightweight variant for edge deployment
Model has been reparameterized for efficient inference

training_report.md 49 KB History Raw

Training Report: test_fix

Basic Information

Model Architecture

Training Results

Dataset Information

Hardware Information

Wakeword Information

Test Results

Wakeword Detection Specific Results

Classification Performance

Confusion Matrix

Robustness Tests

Input Corruption Robustness

Performance Profile

Deployment Recommendations

Detection Threshold

Hardware Requirements

Model Optimization

training_report.md 49 KB

History Raw