| combination,fold,method,lyrics_encoder,original_val_f1,original_test_f1,seed,val_f1,test_f1,test_accuracy,epochs,n_train,n_val,n_test,modalities,themes,note |
| instrument,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.4700842550250773,0.4461180729062705,42.0,0.47008425502507734,0.4461180729062705,0.4538152610441767,11.0,2488.0,498.0,747.0,,, |
| instrument,1,xattn,"theme-tuned BhashaEmbed, fold 1",0.47277156125221,0.4685535812765652,42.0,0.47277156125221004,0.4685535812765652,0.4706666666666667,14.0,2487.0,496.0,750.0,,, |
| instrument,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.4805518276050191,0.4641476984583772,42.0,0.48055182760501913,0.4641476984583772,0.46900269541778977,13.0,2495.0,496.0,742.0,,, |
| instrument,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.4653513903366366,0.4520269853603187,42.0,0.46535139033663664,0.4520269853603187,0.46112600536193027,13.0,2489.0,498.0,746.0,,, |
| instrument,4,xattn,"theme-tuned BhashaEmbed, fold 4",0.4577521222537613,0.4550624525684154,42.0,0.4577521222537613,0.4550624525684154,0.45588235294117646,14.0,2484.0,501.0,748.0,,, |
| instrument+mix,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.4996775428397753,0.486071056447438,42.0,0.4996775428397753,0.486071056447438,0.48728246318607765,11.0,2488.0,498.0,747.0,,, |
| instrument+mix,1,mlp,"theme-tuned BhashaEmbed, fold 1",0.5371986177894004,0.5155519377315119,42.0,0.5371986177894004,0.5155519377315119,0.516,26.0,2487.0,496.0,750.0,,, |
| instrument+mix,2,gmu,"theme-tuned BhashaEmbed, fold 2",0.5408606402458862,0.52716790735528,42.0,0.5408606402458862,0.52716790735528,0.5296495956873315,14.0,2495.0,496.0,742.0,,, |
| instrument+mix,3,gmu,"theme-tuned BhashaEmbed, fold 3",0.5383814495247243,0.5038815804801962,42.0,0.5383814495247243,0.5038815804801962,0.5040214477211796,18.0,2489.0,498.0,746.0,,, |
| instrument+mix,4,xattn,"theme-tuned BhashaEmbed, fold 4",0.5200603047992153,0.5020303428921921,42.0,0.5200603047992153,0.5020303428921921,0.5026737967914439,14.0,2484.0,501.0,748.0,,, |
| lyrics,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.8122301852036925,0.6987697790590008,42.0,0.8122301852036925,0.6987697790590008,0.6987951807228916,36.0,2488.0,498.0,747.0,,, |
| lyrics,1,mlp,"theme-tuned BhashaEmbed, fold 1",0.7314652528224305,0.691567274676409,42.0,0.7314652528224305,0.691567274676409,0.692,32.0,2487.0,496.0,750.0,,, |
| lyrics,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.7573462483663526,0.6566252351630265,42.0,0.7573462483663526,0.6566252351630265,0.6563342318059299,21.0,2495.0,496.0,742.0,,, |
| lyrics,3,mlp,"theme-tuned BhashaEmbed, fold 3",0.8185636161707073,0.6655610718949654,42.0,0.8185636161707073,0.6655610718949654,0.6662198391420912,24.0,2489.0,498.0,746.0,,, |
| lyrics,4,mlp,"theme-tuned BhashaEmbed, fold 4",0.8056909760892167,0.6997170669752799,42.0,0.8056909760892167,0.6997170669752799,0.6991978609625669,18.0,2484.0,501.0,748.0,,, |
| lyrics+instrument,0,gmu,"theme-tuned BhashaEmbed, fold 0",0.8230695586562757,0.7186210370810162,42.0,0.8230695586562757,0.7186210370810162,0.7188755020080321,27.0,2488.0,498.0,747.0,,, |
| lyrics+instrument,1,gmu,"theme-tuned BhashaEmbed, fold 1",0.7433057553450131,0.7037071610944744,42.0,0.7433057553450131,0.7037071610944744,0.7053333333333334,32.0,2487.0,496.0,750.0,,, |
| lyrics+instrument,2,xattn,"theme-tuned BhashaEmbed, fold 2",0.777968622507567,0.6643347271058822,42.0,0.777968622507567,0.6643347271058822,0.6644204851752021,31.0,2495.0,496.0,742.0,,, |
| lyrics+instrument,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.8230801529653682,0.6744343789818227,42.0,0.8230801529653682,0.6744343789818227,0.6742627345844504,43.0,2489.0,498.0,746.0,,, |
| lyrics+instrument,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.8195269220864971,0.709970101992085,42.0,0.8195269220864971,0.709970101992085,0.7098930481283422,20.0,2484.0,501.0,748.0,,, |
| lyrics+instrument+mix,0,xattn,"theme-tuned BhashaEmbed, fold 0",0.823127595788049,0.7106551109201749,42.0,0.823127595788049,0.7106551109201749,0.7108433734939759,35.0,2488.0,498.0,747.0,,, |
| lyrics+instrument+mix,1,xattn,"theme-tuned BhashaEmbed, fold 1",0.7454969848726747,0.708421945122392,42.0,0.7454969848726747,0.708421945122392,0.7093333333333334,43.0,2487.0,496.0,750.0,,, |
| lyrics+instrument+mix,2,gmu,"theme-tuned BhashaEmbed, fold 2",0.7900611810273998,0.6912683876969591,42.0,0.7900611810273998,0.6912683876969591,0.6913746630727763,26.0,2495.0,496.0,742.0,,, |
| lyrics+instrument+mix,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.8235073558697601,0.679956186589753,42.0,0.8235073558697601,0.679956186589753,0.6809651474530831,32.0,2489.0,498.0,746.0,,, |
| lyrics+instrument+mix,4,mlp,"theme-tuned BhashaEmbed, fold 4",0.8153424778150776,0.705684590593435,42.0,0.8153424778150776,0.705684590593435,0.7058823529411765,16.0,2484.0,501.0,748.0,,, |
| lyrics+mix,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.821077103866755,0.7102991502138896,42.0,0.821077103866755,0.7102991502138896,0.7108433734939759,25.0,2488.0,498.0,747.0,,, |
| lyrics+mix,1,gmu,"theme-tuned BhashaEmbed, fold 1",0.7470817039811359,0.7005134786789823,42.0,0.7470817039811359,0.7005134786789823,0.7013333333333334,25.0,2487.0,496.0,750.0,,, |
| lyrics+mix,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.7937392762573197,0.6799918100047826,42.0,0.7937392762573197,0.6799918100047826,0.6792452830188679,33.0,2495.0,496.0,742.0,,, |
| lyrics+mix,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.8173415367536406,0.6475315724477511,42.0,0.8173415367536406,0.6475315724477511,0.6474530831099196,14.0,2489.0,498.0,746.0,,, |
| lyrics+mix,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.8203539939849792,0.714573140481808,42.0,0.8203539939849792,0.714573140481808,0.7152406417112299,26.0,2484.0,501.0,748.0,,, |
| lyrics+vocal,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.8208360718851488,0.6920532562528109,42.0,0.8208360718851488,0.6920532562528109,0.6921017402945113,17.0,2488.0,498.0,747.0,,, |
| lyrics+vocal,1,mlp,"theme-tuned BhashaEmbed, fold 1",0.756195902950362,0.6937554421806452,42.0,0.756195902950362,0.6937554421806452,0.6946666666666667,39.0,2487.0,496.0,750.0,,, |
| lyrics+vocal,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.7856273628332451,0.6826255682512298,42.0,0.7856273628332451,0.6826255682512298,0.683288409703504,22.0,2495.0,496.0,742.0,,, |
| lyrics+vocal,3,gmu,"theme-tuned BhashaEmbed, fold 3",0.8271394422310757,0.6967737065173767,42.0,0.8271394422310757,0.6967737065173767,0.6970509383378016,29.0,2489.0,498.0,746.0,,, |
| lyrics+vocal,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.8260352791073062,0.7000597205032677,42.0,0.8260352791073062,0.7000597205032677,0.6991978609625669,17.0,2484.0,501.0,748.0,,, |
| lyrics+vocal+instrument,0,xattn,"theme-tuned BhashaEmbed, fold 0",0.8214253989618472,0.7056197618000604,42.0,0.8214253989618472,0.7056197618000604,0.7068273092369478,32.0,2488.0,498.0,747.0,,, |
| lyrics+vocal+instrument,1,xattn,"theme-tuned BhashaEmbed, fold 1",0.7458329781793099,0.7026125696187692,42.0,0.7458329781793099,0.7026125696187692,0.704,45.0,2487.0,496.0,750.0,,, |
| lyrics+vocal+instrument,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.7799265613775672,0.6459621478965833,42.0,0.7799265613775672,0.6459621478965833,0.6455525606469003,16.0,2495.0,496.0,742.0,,, |
| lyrics+vocal+instrument,3,mlp,"theme-tuned BhashaEmbed, fold 3",0.8246992150320152,0.7029135572420189,42.0,0.8246992150320152,0.7029135572420189,0.7037533512064343,36.0,2489.0,498.0,746.0,,, |
| lyrics+vocal+instrument,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.821048796474493,0.7046448631327835,42.0,0.821048796474493,0.7046448631327835,0.7045454545454546,12.0,2484.0,501.0,748.0,,, |
| lyrics+vocal+instrument+mix,0,xattn,"theme-tuned BhashaEmbed, fold 0",0.820782888332559,0.698717420060317,42.0,0.820782888332559,0.698717420060317,0.6987951807228916,36.0,2488.0,498.0,747.0,,, |
| lyrics+vocal+instrument+mix,1,mlp,"theme-tuned BhashaEmbed, fold 1",0.7467325471842294,0.6958091914509947,42.0,0.7467325471842294,0.6958091914509947,0.6973333333333334,28.0,2487.0,496.0,750.0,,, |
| lyrics+vocal+instrument+mix,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.7902728819508027,0.6752143834920327,42.0,0.7902728819508027,0.6752143834920327,0.6752021563342318,24.0,2495.0,496.0,742.0,,, |
| lyrics+vocal+instrument+mix,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.8149934721967993,0.6672320952924936,42.0,0.8149934721967993,0.6672320952924936,0.6675603217158177,24.0,2489.0,498.0,746.0,,, |
| lyrics+vocal+instrument+mix,4,mlp,"theme-tuned BhashaEmbed, fold 4",0.8234838519037591,0.7142637356502748,42.0,0.8234838519037591,0.7142637356502748,0.713903743315508,19.0,2484.0,501.0,748.0,,, |
| lyrics+vocal+mix,0,xattn,"theme-tuned BhashaEmbed, fold 0",0.8281003367456656,0.6958941773792675,42.0,0.8281003367456656,0.6958941773792675,0.6974564926372155,35.0,2488.0,498.0,747.0,,, |
| lyrics+vocal+mix,1,gmu,"theme-tuned BhashaEmbed, fold 1",0.7577065355600524,0.6852597482153101,42.0,0.7577065355600524,0.6852597482153101,0.6866666666666666,15.0,2487.0,496.0,750.0,,, |
| lyrics+vocal+mix,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.7877603503410434,0.6681715498827105,42.0,0.7877603503410434,0.6681715498827105,0.6684636118598383,22.0,2495.0,496.0,742.0,,, |
| lyrics+vocal+mix,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.8193490404193923,0.6828040075995647,42.0,0.8193490404193923,0.6828040075995647,0.6836461126005362,34.0,2489.0,498.0,746.0,,, |
| lyrics+vocal+mix,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.8215977790675333,0.7194492213802988,42.0,0.8215977790675333,0.7194492213802988,0.7192513368983957,17.0,2484.0,501.0,748.0,,, |
| mix,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.5003362503962236,0.4885087358877681,42.0,0.5003362503962236,0.48850873588776816,0.4899598393574297,17.0,2488.0,498.0,747.0,,, |
| mix,1,xattn,"theme-tuned BhashaEmbed, fold 1",0.5235147697005602,0.4919678525423758,42.0,0.5235147697005602,0.49196785254237585,0.49333333333333335,27.0,2487.0,496.0,750.0,,, |
| mix,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.5361717574319459,0.5053672772796988,42.0,0.5361717574319459,0.5053672772796988,0.5067385444743935,14.0,2495.0,496.0,742.0,,, |
| mix,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.5138865336394306,0.5126116581302644,42.0,0.5138865336394306,0.5126116581302644,0.517426273458445,15.0,2489.0,498.0,746.0,,, |
| mix,4,mlp,"theme-tuned BhashaEmbed, fold 4",0.515068615442378,0.5023883427794152,42.0,0.515068615442378,0.5023883427794152,0.5053475935828877,11.0,2484.0,501.0,748.0,,, |
| vocal,0,xattn,"theme-tuned BhashaEmbed, fold 0",0.5263340359513751,0.4964111295899753,42.0,0.5263340359513751,0.49641112958997535,0.4979919678714859,25.0,2488.0,498.0,747.0,,, |
| vocal,1,xattn,"theme-tuned BhashaEmbed, fold 1",0.5229548252155484,0.5017997599915476,42.0,0.5229548252155484,0.5017997599915476,0.5026666666666667,15.0,2487.0,496.0,750.0,,, |
| vocal,2,xattn,"theme-tuned BhashaEmbed, fold 2",0.5624030004067485,0.4942822466593294,42.0,0.5624030004067485,0.49428224665932946,0.4973045822102426,23.0,2495.0,496.0,742.0,,, |
| vocal,3,xattn,"theme-tuned BhashaEmbed, fold 3",0.525623663578047,0.4884536649417602,42.0,0.525623663578047,0.48845366494176023,0.4932975871313673,22.0,2489.0,498.0,746.0,,, |
| vocal,4,mlp,"theme-tuned BhashaEmbed, fold 4",0.5244867777732465,0.4862258537127303,42.0,0.5244867777732465,0.48622585371273036,0.48663101604278075,22.0,2484.0,501.0,748.0,,, |
| vocal+instrument,0,mlp,"theme-tuned BhashaEmbed, fold 0",0.5432375318656966,0.502058848497705,42.0,0.5432375318656966,0.502058848497705,0.5020080321285141,19.0,2488.0,498.0,747.0,,, |
| vocal+instrument,1,mlp,"theme-tuned BhashaEmbed, fold 1",0.5371541233901065,0.509502695131556,42.0,0.5371541233901065,0.509502695131556,0.512,36.0,2487.0,496.0,750.0,,, |
| vocal+instrument,2,late,"theme-tuned BhashaEmbed, fold 2",0.5624612068337824,0.5176264536144836,,,,,,,,,"['vocal', 'instrument']","['romance', 'sadness', 'celebration', 'motivational']","late fusion: average the class probabilities of the single-modality models vocal/fold_2, instrument/fold_2" |
| vocal+instrument,3,mlp,"theme-tuned BhashaEmbed, fold 3",0.5343063319590855,0.515475799036193,42.0,0.5343063319590855,0.515475799036193,0.5187667560321716,18.0,2489.0,498.0,746.0,,, |
| vocal+instrument,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.5281688393262916,0.5060683231585856,42.0,0.5281688393262916,0.5060683231585856,0.5053475935828877,25.0,2484.0,501.0,748.0,,, |
| vocal+instrument+mix,0,gmu,"theme-tuned BhashaEmbed, fold 0",0.5405991707444369,0.5059345545834361,42.0,0.5405991707444369,0.5059345545834361,0.5113788487282463,13.0,2488.0,498.0,747.0,,, |
| vocal+instrument+mix,1,xattn,"theme-tuned BhashaEmbed, fold 1",0.5563348184927218,0.5140207789911702,42.0,0.5563348184927218,0.5140207789911702,0.516,28.0,2487.0,496.0,750.0,,, |
| vocal+instrument+mix,2,mlp,"theme-tuned BhashaEmbed, fold 2",0.5706230063570612,0.5065702363571579,42.0,0.5706230063570612,0.5065702363571579,0.5080862533692723,12.0,2495.0,496.0,742.0,,, |
| vocal+instrument+mix,3,gmu,"theme-tuned BhashaEmbed, fold 3",0.5367278816532842,0.5067658175309705,42.0,0.5367278816532842,0.5067658175309705,0.5080428954423593,15.0,2489.0,498.0,746.0,,, |
| vocal+instrument+mix,4,mlp,"theme-tuned BhashaEmbed, fold 4",0.5357516898543613,0.5305088714195546,42.0,0.5357516898543613,0.5305088714195546,0.5334224598930482,18.0,2484.0,501.0,748.0,,, |
| vocal+mix,0,gmu,"theme-tuned BhashaEmbed, fold 0",0.5423120972967386,0.529783755119147,42.0,0.5423120972967386,0.529783755119147,0.5287817938420348,16.0,2488.0,498.0,747.0,,, |
| vocal+mix,1,mlp,"theme-tuned BhashaEmbed, fold 1",0.5541633312090593,0.5017758568103949,42.0,0.5541633312090593,0.5017758568103949,0.504,21.0,2487.0,496.0,750.0,,, |
| vocal+mix,2,late,"theme-tuned BhashaEmbed, fold 2",0.5665349237740762,0.5230647907698776,,,,,,,,,"['vocal', 'mix']","['romance', 'sadness', 'celebration', 'motivational']","late fusion: average the class probabilities of the single-modality models vocal/fold_2, mix/fold_2" |
| vocal+mix,3,gmu,"theme-tuned BhashaEmbed, fold 3",0.5587289059511282,0.5137783826601575,42.0,0.5587289059511282,0.5137783826601575,0.514745308310992,26.0,2489.0,498.0,746.0,,, |
| vocal+mix,4,gmu,"theme-tuned BhashaEmbed, fold 4",0.5433863792122453,0.5071676508818519,42.0,0.5433863792122453,0.5071676508818519,0.5080213903743316,18.0,2484.0,501.0,748.0,,, |
|
|
