HTDemucs-ft / d12395a8.json
adefossez's picture
Upload folder using huggingface_hub
478be8a verified
Raw
History Blame Contribute Delete
11.1 kB
{
"klass": "demucs.htdemucs.HTDemucs",
"args": "[]",
"kwargs": "{\"sources\": [\"drums\", \"bass\", \"other\", \"vocals\"], \"audio_channels\": 2, \"samplerate\": 44100, \"segment\": {\"_type\": \"fraction\", \"numerator\": 39, \"denominator\": 5}, \"channels\": 48, \"channels_time\": null, \"growth\": 2, \"nfft\": 4096, \"wiener_iters\": 0, \"end_iters\": 0, \"wiener_residual\": false, \"cac\": true, \"depth\": 4, \"rewrite\": true, \"multi_freqs\": [], \"multi_freqs_depth\": 3, \"freq_emb\": 0.2, \"emb_scale\": 10, \"emb_smooth\": true, \"kernel_size\": 8, \"stride\": 4, \"time_stride\": 2, \"context\": 1, \"context_enc\": 0, \"norm_starts\": 4, \"norm_groups\": 4, \"dconv_mode\": 3, \"dconv_depth\": 2, \"dconv_comp\": 8, \"dconv_init\": 0.001, \"bottom_channels\": 512, \"t_layers\": 5, \"t_hidden_scale\": 4.0, \"t_heads\": 8, \"t_dropout\": 0.02, \"t_layer_scale\": true, \"t_gelu\": true, \"t_emb\": \"sin\", \"t_max_positions\": 10000, \"t_max_period\": 10000.0, \"t_weight_pos_embed\": 1.0, \"t_cape_mean_normalize\": true, \"t_cape_augment\": true, \"t_cape_glob_loc_scale\": [5000.0, 1.0, 1.4], \"t_sin_random_shift\": 0, \"t_norm_in\": true, \"t_norm_in_group\": false, \"t_group_norm\": false, \"t_norm_first\": true, \"t_norm_out\": true, \"t_weight_decay\": 0.05, \"t_lr\": null, \"t_sparse_self_attn\": false, \"t_sparse_cross_attn\": false, \"t_mask_type\": \"diag\", \"t_mask_random_seed\": 42, \"t_sparse_attn_window\": 400, \"t_global_window\": 100, \"t_sparsity\": 0.95, \"t_auto_sparsity\": false, \"t_cross_first\": false, \"rescale\": 0.1}",
"training_args": {
"dummy": null,
"dset": {
"musdb": "/checkpoint/defossez/datasets/musdbhq",
"musdb_samplerate": 44100,
"wav": "/checkpoint/defossez/datasets/allstems_44/",
"wav2": "/checkpoint/defossez/datasets/mmi44_goodclean",
"segment": 10,
"shift": 1,
"train_valid": false,
"full_cv": true,
"samplerate": 44100,
"channels": 2,
"normalize": true,
"metadata": "./metadata",
"sources": [
"drums",
"bass",
"other",
"vocals"
],
"valid_samples": 100,
"wav2_weight": null,
"wav2_valid": true
},
"test": {
"save": false,
"best": true,
"workers": 2,
"every": 20,
"split": true,
"shifts": 0,
"overlap": 0.25,
"sdr": true,
"metric": "loss",
"nonhq": null
},
"epochs": 50,
"batch_size": 8,
"max_batches": 800,
"optim": {
"lr": 0.0001,
"momentum": 0.9,
"beta2": 0.999,
"loss": "l1",
"optim": "adamw",
"weight_decay": 0,
"clip_grad": 5
},
"seed": 42,
"debug": false,
"valid_apply": true,
"flag": null,
"save_every": null,
"weights": [
0,
1,
0,
0
],
"augment": {
"shift_same": true,
"repitch": {
"proba": 0.2,
"max_tempo": 12
},
"remix": {
"proba": 0,
"group_size": 4
},
"scale": {
"proba": 0,
"min": 0.25,
"max": 1.25
},
"flip": true
},
"continue_from": "955717e8",
"continue_pretrained": null,
"pretrained_repo": null,
"continue_best": true,
"continue_opt": false,
"misc": {
"num_workers": 10,
"num_prints": 4,
"show": false,
"verbose": false
},
"ema": {
"epoch": [],
"batch": [
0.9995,
0.9999
]
},
"use_train_segment": true,
"model_segment": null,
"model": "htdemucs",
"demucs": {
"channels": 64,
"growth": 2,
"depth": 6,
"rewrite": true,
"lstm_layers": 0,
"kernel_size": 8,
"stride": 4,
"context": 1,
"gelu": true,
"glu": true,
"norm_groups": 4,
"norm_starts": 4,
"dconv_depth": 2,
"dconv_mode": 1,
"dconv_comp": 4,
"dconv_attn": 4,
"dconv_lstm": 4,
"dconv_init": 0.0001,
"resample": true,
"normalize": false,
"rescale": 0.1
},
"hdemucs": {
"channels": 48,
"channels_time": null,
"growth": 2,
"nfft": 4096,
"wiener_iters": 0,
"end_iters": 0,
"wiener_residual": false,
"cac": true,
"depth": 6,
"rewrite": true,
"hybrid": true,
"hybrid_old": false,
"multi_freqs": [],
"multi_freqs_depth": 3,
"freq_emb": 0.2,
"emb_scale": 10,
"emb_smooth": true,
"kernel_size": 8,
"stride": 4,
"time_stride": 2,
"context": 1,
"context_enc": 0,
"norm_starts": 4,
"norm_groups": 4,
"dconv_mode": 1,
"dconv_depth": 2,
"dconv_comp": 4,
"dconv_attn": 4,
"dconv_lstm": 4,
"dconv_init": 0.001,
"rescale": 0.1
},
"htdemucs": {
"channels": 48,
"channels_time": null,
"growth": 2,
"nfft": 4096,
"wiener_iters": 0,
"end_iters": 0,
"wiener_residual": false,
"cac": true,
"depth": 4,
"rewrite": true,
"multi_freqs": [],
"multi_freqs_depth": 3,
"freq_emb": 0.2,
"emb_scale": 10,
"emb_smooth": true,
"kernel_size": 8,
"stride": 4,
"time_stride": 2,
"context": 1,
"context_enc": 0,
"norm_starts": 4,
"norm_groups": 4,
"dconv_mode": 3,
"dconv_depth": 2,
"dconv_comp": 8,
"dconv_init": 0.001,
"bottom_channels": 512,
"t_layers": 5,
"t_hidden_scale": 4.0,
"t_heads": 8,
"t_dropout": 0.02,
"t_layer_scale": true,
"t_gelu": true,
"t_emb": "sin",
"t_max_positions": 10000,
"t_max_period": 10000.0,
"t_weight_pos_embed": 1.0,
"t_cape_mean_normalize": true,
"t_cape_augment": true,
"t_cape_glob_loc_scale": [
5000.0,
1.0,
1.4
],
"t_sin_random_shift": 0,
"t_norm_in": true,
"t_norm_in_group": false,
"t_group_norm": false,
"t_norm_first": true,
"t_norm_out": true,
"t_weight_decay": 0.05,
"t_lr": null,
"t_sparse_self_attn": false,
"t_sparse_cross_attn": false,
"t_mask_type": "diag",
"t_mask_random_seed": 42,
"t_sparse_attn_window": 400,
"t_global_window": 100,
"t_sparsity": 0.95,
"t_auto_sparsity": false,
"t_cross_first": false,
"rescale": 0.1
},
"svd": {
"penalty": 0,
"min_size": 0.1,
"dim": 1,
"niters": 2,
"powm": false,
"proba": 1,
"conv_only": false,
"convtr": false,
"bs": 1
},
"quant": {
"diffq": null,
"qat": null,
"min_size": 0.2,
"group_size": 8
},
"dora": {
"dir": "outputs",
"exclude": [
"misc.*",
"slurm.*",
"test.reval",
"flag"
]
},
"slurm": {
"time": 4320,
"constraint": "volta32gb",
"setup": [
"module load cudnn/v8.4.1.50-cuda.11.6 NCCL/2.11.4-6-cuda.11.6 cuda/11.6"
]
}
},
"metrics": [
{
"main": {
"loss": 0.07459118962287903,
"ms": 0.0,
"nsdr": 12.226367950439453,
"nsdr_bass": 12.226367950439453,
"nsdr_drums": 4.625657558441162,
"nsdr_other": 3.8199150562286377,
"nsdr_vocals": 5.249729156494141,
"reco": 0.07459118962287903,
"reco_bass": 0.07459118962287903,
"reco_drums": 0.17048002779483795,
"reco_other": 0.19498775899410248,
"reco_vocals": 0.1458791196346283
},
"ema_batch_0": {
"loss": 0.07329171150922775,
"ms": 0.0,
"nsdr": 12.355029106140137,
"nsdr_bass": 12.355029106140137,
"nsdr_drums": 4.775204658508301,
"nsdr_other": 3.98699951171875,
"nsdr_vocals": 5.435001850128174,
"reco": 0.07329171150922775,
"reco_bass": 0.07329171150922775,
"reco_drums": 0.16768485307693481,
"reco_other": 0.19094298779964447,
"reco_vocals": 0.14255551993846893
},
"ema_batch_1": {
"loss": 0.07299935072660446,
"ms": 0.0,
"nsdr": 12.405416488647461,
"nsdr_bass": 12.405416488647461,
"nsdr_drums": 5.3668293952941895,
"nsdr_other": 4.57891845703125,
"nsdr_vocals": 6.068881034851074,
"reco": 0.07299935072660446,
"reco_bass": 0.07299935072660446,
"reco_drums": 0.15755127370357513,
"reco_other": 0.17828567326068878,
"reco_vocals": 0.1319224238395691
},
"loss": 0.07299935072660446,
"ms": 0.0,
"nsdr": 12.405416488647461,
"nsdr_bass": 12.405416488647461,
"nsdr_drums": 5.3668293952941895,
"nsdr_other": 4.57891845703125,
"nsdr_vocals": 6.068881034851074,
"reco": 0.07299935072660446,
"reco_bass": 0.07299935072660446,
"reco_drums": 0.15755127370357513,
"reco_other": 0.17828567326068878,
"reco_vocals": 0.1319224238395691,
"bname": "ema_batch_1",
"best": 0.07284779846668243
},
{
"nsdr_drums": 5.658610381882371,
"nsdr_med_drums": 5.80060424290383,
"nsdr_bass": 9.002055392391293,
"nsdr_med_bass": 9.07940117223616,
"nsdr_other": 4.611072469913693,
"nsdr_med_other": 5.1036910147710755,
"nsdr_vocals": 6.541697180184338,
"nsdr_med_vocals": 7.2168281807711825,
"nsdr": 6.453358856092924,
"nsdr_med": 6.800131152670563,
"sdr_drums": 5.54934189057237,
"sdr_med_drums": 5.333804068541108,
"sdr_bass": 9.254056514751642,
"sdr_med_bass": 10.389953922219558,
"sdr_other": 4.211750783644854,
"sdr_med_other": 4.611771763855684,
"sdr_vocals": 5.061763938348276,
"sdr_med_vocals": 6.220513431962674,
"sdr": 6.019228281829285,
"sdr_med": 6.639010796644756,
"sir_drums": 12.054832824473205,
"sir_med_drums": 17.430082777404607,
"sir_bass": 17.45758720732738,
"sir_med_bass": 17.680470512832308,
"sir_other": 1.3239675737972487,
"sir_med_other": 2.6533972153428587,
"sir_vocals": 6.425074427360061,
"sir_med_vocals": 8.100113366398997,
"sir": 9.315365508239474,
"sir_med": 11.466015967994693,
"isr_drums": 6.389546891107501,
"isr_med_drums": 6.01333182433255,
"isr_bass": 8.42766972491811,
"isr_med_bass": 9.372830721045464,
"isr_other": 11.117468470708078,
"isr_med_other": 11.353696768126351,
"isr_vocals": 11.803012557592133,
"isr_med_vocals": 12.695746216906267,
"isr": 9.434424411081455,
"isr_med": 9.858901382602658,
"sar_drums": 7.327009963891771,
"sar_med_drums": 7.714720001169638,
"sar_bass": 7.1366475349171035,
"sar_med_bass": 7.187720691109244,
"sar_other": 5.667175104415401,
"sar_med_other": 6.055248941377247,
"sar_vocals": 6.442210963593489,
"sar_med_vocals": 6.225578255311007,
"sar": 6.643260891704442,
"sar_med": 6.795816972241784
}
]
}