Download 04573f0d.json from MisterCD/HTDemucs-ft: direct link, hf CLI and curl.
- Browser
- Download file 11.1 kB
-
https://huggingface.co/MisterCD/HTDemucs-ft/resolve/main/04573f0d.json
- Command line
-
hf download hf://MisterCD/HTDemucs-ft/04573f0d.json
-
curl -L -o 04573f0d.json https://huggingface.co/MisterCD/HTDemucs-ft/resolve/main/04573f0d.json
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, | |
| 0, | |
| 0, | |
| 1 | |
| ], | |
| "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": 0.08635193854570389, | |
| "ms": 0.0, | |
| "nsdr": 9.887733459472656, | |
| "nsdr_bass": 1.9094270467758179, | |
| "nsdr_drums": -1.4275637865066528, | |
| "nsdr_other": 1.3280521631240845, | |
| "nsdr_vocals": 9.887733459472656, | |
| "reco": 0.08635193854570389, | |
| "reco_bass": 0.27492132782936096, | |
| "reco_drums": 0.29225555062294006, | |
| "reco_other": 0.2490427941083908, | |
| "reco_vocals": 0.08635193854570389 | |
| }, | |
| "ema_batch_0": { | |
| "loss": 0.08506794273853302, | |
| "ms": 0.0, | |
| "nsdr": 10.032146453857422, | |
| "nsdr_bass": 2.023679733276367, | |
| "nsdr_drums": -1.3376169204711914, | |
| "nsdr_other": 1.3965520858764648, | |
| "nsdr_vocals": 10.032146453857422, | |
| "reco": 0.08506794273853302, | |
| "reco_bass": 0.27160191535949707, | |
| "reco_drums": 0.2892371118068695, | |
| "reco_other": 0.24691219627857208, | |
| "reco_vocals": 0.08506794273853302 | |
| }, | |
| "ema_batch_1": { | |
| "loss": 0.085032157599926, | |
| "ms": 0.0, | |
| "nsdr": 10.01611042022705, | |
| "nsdr_bass": 2.701692819595337, | |
| "nsdr_drums": -0.03992346301674843, | |
| "nsdr_other": 2.335426092147827, | |
| "nsdr_vocals": 10.01611042022705, | |
| "reco": 0.085032157599926, | |
| "reco_bass": 0.2510550022125244, | |
| "reco_drums": 0.258406400680542, | |
| "reco_other": 0.2280634641647339, | |
| "reco_vocals": 0.085032157599926 | |
| }, | |
| "loss": 0.085032157599926, | |
| "ms": 0.0, | |
| "nsdr": 10.01611042022705, | |
| "nsdr_bass": 2.701692819595337, | |
| "nsdr_drums": -0.03992346301674843, | |
| "nsdr_other": 2.335426092147827, | |
| "nsdr_vocals": 10.01611042022705, | |
| "reco": 0.085032157599926, | |
| "reco_bass": 0.2510550022125244, | |
| "reco_drums": 0.258406400680542, | |
| "reco_other": 0.2280634641647339, | |
| "reco_vocals": 0.085032157599926, | |
| "bname": "ema_batch_1", | |
| "best": 0.08491263538599014 | |
| }, | |
| { | |
| "nsdr_drums": -0.6409806271595753, | |
| "nsdr_med_drums": -0.4546512620871669, | |
| "nsdr_bass": 1.901309260012953, | |
| "nsdr_med_bass": 1.99247581790831, | |
| "nsdr_other": 1.8001048182240853, | |
| "nsdr_med_other": 2.25102747995457, | |
| "nsdr_vocals": 8.965026613502095, | |
| "nsdr_med_vocals": 9.708247677177978, | |
| "nsdr": 3.0063650161448896, | |
| "nsdr_med": 3.374274928238423, | |
| "sdr_drums": -0.9858202949154312, | |
| "sdr_med_drums": -0.7586626432388546, | |
| "sdr_bass": 1.8360585395109574, | |
| "sdr_med_bass": 1.9923042432558375, | |
| "sdr_other": 1.5559783147409378, | |
| "sdr_med_other": 2.0184491417303843, | |
| "sdr_vocals": 8.276195251269366, | |
| "sdr_med_vocals": 9.206614327457075, | |
| "sdr": 2.6706029526514574, | |
| "sdr_med": 3.1146762673011104, | |
| "sir_drums": -3.0703757442433983, | |
| "sir_med_drums": 3.7515712829766548, | |
| "sir_bass": 14.378058671572825, | |
| "sir_med_bass": 13.046233457444732, | |
| "sir_other": -1.268877787576836, | |
| "sir_med_other": 0.47189632386594926, | |
| "sir_vocals": 13.207182759818481, | |
| "sir_med_vocals": 15.4689984820013, | |
| "sir": 5.811496974892768, | |
| "sir_med": 8.184674886572159, | |
| "isr_drums": 1.5977322153466869, | |
| "isr_med_drums": 1.6044336761340112, | |
| "isr_bass": -2.7818493299655405, | |
| "isr_med_bass": 1.7088512018851911, | |
| "isr_other": 4.336090816781649, | |
| "isr_med_other": 4.25185412894745, | |
| "isr_vocals": 15.43228601785809, | |
| "isr_med_vocals": 15.255793981166773, | |
| "isr": 4.646064930005221, | |
| "isr_med": 5.705233247033356, | |
| "sar_drums": -6.886988773866586, | |
| "sar_med_drums": -8.267265823203246, | |
| "sar_bass": 3.499480084038568, | |
| "sar_med_bass": 2.626303437936244, | |
| "sar_other": 0.09047195213537478, | |
| "sar_med_other": -0.004790688026467377, | |
| "sar_vocals": 7.007389282932945, | |
| "sar_med_vocals": 7.066486437650068, | |
| "sar": 0.9275881363100756, | |
| "sar_med": 0.3551833410891494 | |
| } | |
| ] | |
| } |