pushing separators

convtasnet1/CKPT.yaml

@@ -0,0 +1,4 @@
+# yamllint disable
+end-of-epoch: true
+si-snr: -0.42996728845946536
+unixtime: 1631299321.582555

convtasnet1/brain.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e24193f36931b7f57932532efbdcf64971f42732383ba6808825f77db258f6
+size 28

convtasnet1/counter.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b86b273ff34fce19d6b804eff5a3f5747ada4eaa22f1d49c01e52ddb7875b4b
+size 1

convtasnet1/decoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:623030a5a4317eabc555ec09254d6a05e5f3811933c429f90f06f903a22b808c
+size 17272

convtasnet1/encoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4314241647c3b6adece727c3ff8066f674c516ad0a531efaa0bbd89eb4786050
+size 17272

convtasnet1/hyperparams.yaml

@@ -0,0 +1,168 @@
+# Generated 2021-09-15 from:
+# /home/mila/s/subakany/speechbrain_new/recipes/WHAMandWHAMR/separation/hparams/convtasnet-whamr.yaml
+# yamllint disable
+# ################################
+# Model: SepFormer for source separation
+# https://arxiv.org/abs/2010.13154
+#
+# Dataset : WHAMR!
+# ################################
+# Basic parameters
+# Seed needs to be set at top of yaml, before objects with parameters are made
+#
+seed: 3
+__set_seed: !apply:torch.manual_seed [3]
+
+# Data params
+
+# the data folder for the wham dataset
+# data_folder needs to follow the format:  /yourpath/whamr.
+# make sure to use the name whamr at your top folder for the dataset!
+data_folder: /network/tmp1/subakany/whamr
+
+# the path for wsj0/si_tr_s/ folder -- only needed if dynamic mixing is used
+# e.g. /yourpath/wsj0-processed/si_tr_s/
+# you need to convert the original wsj0 to 8k
+# you can do this conversion with the script ../meta/preprocess_dynamic_mixing.py
+base_folder_dm: /network/tmp1/subakany/wsj0-processed/si_tr_s/
+
+experiment_name: convtasnet-whamr
+output_folder: results/convtasnet-whamr/3
+train_log: results/convtasnet-whamr/3/train_log.txt
+save_folder: results/convtasnet-whamr/3/save
+
+# the file names should start with whamr instead of whamorg
+train_data: results/convtasnet-whamr/3/save/whamr_tr.csv
+valid_data: results/convtasnet-whamr/3/save/whamr_cv.csv
+test_data: results/convtasnet-whamr/3/save/whamr_tt.csv
+skip_prep: false
+
+# Experiment params
+auto_mix_prec: false # Set it to True for mixed precision
+test_only: false
+num_spks: 2 # set to 3 for wsj0-3mix
+progressbar: true
+save_audio: false # Save estimated sources on disk
+sample_rate: 8000
+
+# Training parameters
+N_epochs: 200
+batch_size: 1
+lr: 0.00015
+clip_grad_norm: 5
+loss_upper_lim: 999999  # this is the upper limit for an acceptable loss
+# if True, the training sequences are cut to a specified length
+limit_training_signal_len: false
+# this is the length of sequences if we choose to limit
+# the signal length of training sequences
+training_signal_len: 32000000
+
+# Set it to True to dynamically create mixtures at training time
+dynamic_mixing: true
+
+# Parameters for data augmentation
+
+# rir_path variable points to the directory of the room impulse responses
+# e.g. /miniscratch/subakany/rir_wavs
+# If the path does not exist, it is created automatically.
+rir_path: /miniscratch/subakany/whamr_rirs_wav
+
+use_wavedrop: false
+use_speedperturb: true
+use_speedperturb_sameforeachsource: false
+use_rand_shift: false
+min_shift: -8000
+max_shift: 8000
+
+speedperturb: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 1.0
+  drop_freq_prob: 0.0
+  drop_chunk_prob: 0.0
+  sample_rate: 8000
+  speeds: [95, 100, 105]
+
+wavedrop: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 0.0
+  drop_freq_prob: 1.0
+  drop_chunk_prob: 1.0
+  sample_rate: 8000
+
+# loss thresholding -- this thresholds the training loss
+threshold_byloss: true
+threshold: -30
+
+# Encoder parameters
+N_encoder_out: 256
+out_channels: 256
+kernel_size: 16
+kernel_stride: 8
+
+# Dataloader options
+dataloader_opts:
+  batch_size: 1
+  num_workers: 3
+
+
+# Specifying the network
+Encoder: &id001 !new:speechbrain.lobes.models.dual_path.Encoder
+  kernel_size: 16
+  out_channels: 256
+
+
+MaskNet: &id003 !new:speechbrain.lobes.models.conv_tasnet.MaskNet
+
+  N: 256
+  B: 256
+  H: 512
+  P: 3
+  X: 6
+  R: 4
+  C: 2
+  norm_type: gLN
+  causal: false
+  mask_nonlinear: relu
+
+Decoder: &id002 !new:speechbrain.lobes.models.dual_path.Decoder
+  in_channels: 256
+  out_channels: 1
+  kernel_size: 16
+  stride: 8
+  bias: false
+
+
+optimizer: !name:torch.optim.Adam
+  lr: 0.00015
+  weight_decay: 0
+
+loss: !name:speechbrain.nnet.losses.get_si_snr_with_pitwrapper
+
+lr_scheduler: &id005 !new:speechbrain.nnet.schedulers.ReduceLROnPlateau
+
+  factor: 0.5
+  patience: 2
+  dont_halve_until_epoch: 85
+
+epoch_counter: &id004 !new:speechbrain.utils.epoch_loop.EpochCounter
+  limit: 200
+
+modules:
+  encoder: *id001
+  decoder: *id002
+  masknet: *id003
+save_all_checkpoints: true
+checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
+  checkpoints_dir: results/convtasnet-whamr/3/save
+  recoverables:
+    encoder: *id001
+    decoder: *id002
+    masknet: *id003
+    counter: *id004
+    lr_scheduler: *id005
+train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
+  save_file: results/convtasnet-whamr/3/train_log.txt
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+  loadables:
+      encoder: !ref <Encoder>
+      masknet: !ref <MaskNet>
+      decoder: !ref <Decoder>

convtasnet1/lr_scheduler.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b633ee5d0a19696bcf1025be87e8e7ec9b783ad2b9adfa2077e057d18accaea6
+size 495

convtasnet1/masknet.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c8606649b45db4841e79ef35ec71bb7e0f79b1e5101fe690ef083dbb7c0c21c
+size 26404523

convtasnet1/optimizer.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eec7c786af8ea382d342b793fa3518fe337f72e114077b4755cfc0c521b74754
+size 52803531

convtasnet2/CKPT.yaml

@@ -0,0 +1,4 @@
+# yamllint disable
+end-of-epoch: true
+si-snr: -4.564481391887362
+unixtime: 1631590081.3795788

convtasnet2/brain.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e24193f36931b7f57932532efbdcf64971f42732383ba6808825f77db258f6
+size 28

convtasnet2/counter.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:031b4af5197ec30a926f48cf40e11a7dbc470048a21e4003b7a3c07c5dab1baa
+size 2

convtasnet2/decoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:680ae89a64815fb9ea7a9bf2e19682841772253084ba5db0fe30db96a62e16f1
+size 17272

convtasnet2/encoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5830174e7975d20ec1a82e6eb90521ad2359f58d69245864c6299d7fe89b94e3
+size 17272

convtasnet2/hyperparams.yaml

@@ -0,0 +1,168 @@
+# Generated 2021-09-15 from:
+# /home/mila/s/subakany/speechbrain_new/recipes/WHAMandWHAMR/separation/hparams/convtasnet-whamr.yaml
+# yamllint disable
+# ################################
+# Model: SepFormer for source separation
+# https://arxiv.org/abs/2010.13154
+#
+# Dataset : WHAMR!
+# ################################
+# Basic parameters
+# Seed needs to be set at top of yaml, before objects with parameters are made
+#
+seed: 3
+__set_seed: !apply:torch.manual_seed [3]
+
+# Data params
+
+# the data folder for the wham dataset
+# data_folder needs to follow the format:  /yourpath/whamr.
+# make sure to use the name whamr at your top folder for the dataset!
+data_folder: /network/tmp1/subakany/whamr
+
+# the path for wsj0/si_tr_s/ folder -- only needed if dynamic mixing is used
+# e.g. /yourpath/wsj0-processed/si_tr_s/
+# you need to convert the original wsj0 to 8k
+# you can do this conversion with the script ../meta/preprocess_dynamic_mixing.py
+base_folder_dm: /network/tmp1/subakany/wsj0-processed/si_tr_s/
+
+experiment_name: convtasnet-whamr
+output_folder: results/convtasnet-whamr/3
+train_log: results/convtasnet-whamr/3/train_log.txt
+save_folder: results/convtasnet-whamr/3/save
+
+# the file names should start with whamr instead of whamorg
+train_data: results/convtasnet-whamr/3/save/whamr_tr.csv
+valid_data: results/convtasnet-whamr/3/save/whamr_cv.csv
+test_data: results/convtasnet-whamr/3/save/whamr_tt.csv
+skip_prep: false
+
+# Experiment params
+auto_mix_prec: false # Set it to True for mixed precision
+test_only: false
+num_spks: 2 # set to 3 for wsj0-3mix
+progressbar: true
+save_audio: false # Save estimated sources on disk
+sample_rate: 8000
+
+# Training parameters
+N_epochs: 200
+batch_size: 1
+lr: 0.00015
+clip_grad_norm: 5
+loss_upper_lim: 999999  # this is the upper limit for an acceptable loss
+# if True, the training sequences are cut to a specified length
+limit_training_signal_len: false
+# this is the length of sequences if we choose to limit
+# the signal length of training sequences
+training_signal_len: 32000000
+
+# Set it to True to dynamically create mixtures at training time
+dynamic_mixing: true
+
+# Parameters for data augmentation
+
+# rir_path variable points to the directory of the room impulse responses
+# e.g. /miniscratch/subakany/rir_wavs
+# If the path does not exist, it is created automatically.
+rir_path: /miniscratch/subakany/whamr_rirs_wav
+
+use_wavedrop: false
+use_speedperturb: true
+use_speedperturb_sameforeachsource: false
+use_rand_shift: false
+min_shift: -8000
+max_shift: 8000
+
+speedperturb: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 1.0
+  drop_freq_prob: 0.0
+  drop_chunk_prob: 0.0
+  sample_rate: 8000
+  speeds: [95, 100, 105]
+
+wavedrop: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 0.0
+  drop_freq_prob: 1.0
+  drop_chunk_prob: 1.0
+  sample_rate: 8000
+
+# loss thresholding -- this thresholds the training loss
+threshold_byloss: true
+threshold: -30
+
+# Encoder parameters
+N_encoder_out: 256
+out_channels: 256
+kernel_size: 16
+kernel_stride: 8
+
+# Dataloader options
+dataloader_opts:
+  batch_size: 1
+  num_workers: 3
+
+
+# Specifying the network
+Encoder: &id001 !new:speechbrain.lobes.models.dual_path.Encoder
+  kernel_size: 16
+  out_channels: 256
+
+
+MaskNet: &id003 !new:speechbrain.lobes.models.conv_tasnet.MaskNet
+
+  N: 256
+  B: 256
+  H: 512
+  P: 3
+  X: 6
+  R: 4
+  C: 2
+  norm_type: gLN
+  causal: false
+  mask_nonlinear: relu
+
+Decoder: &id002 !new:speechbrain.lobes.models.dual_path.Decoder
+  in_channels: 256
+  out_channels: 1
+  kernel_size: 16
+  stride: 8
+  bias: false
+
+
+optimizer: !name:torch.optim.Adam
+  lr: 0.00015
+  weight_decay: 0
+
+loss: !name:speechbrain.nnet.losses.get_si_snr_with_pitwrapper
+
+lr_scheduler: &id005 !new:speechbrain.nnet.schedulers.ReduceLROnPlateau
+
+  factor: 0.5
+  patience: 2
+  dont_halve_until_epoch: 85
+
+epoch_counter: &id004 !new:speechbrain.utils.epoch_loop.EpochCounter
+  limit: 200
+
+modules:
+  encoder: *id001
+  decoder: *id002
+  masknet: *id003
+save_all_checkpoints: true
+checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
+  checkpoints_dir: results/convtasnet-whamr/3/save
+  recoverables:
+    encoder: *id001
+    decoder: *id002
+    masknet: *id003
+    counter: *id004
+    lr_scheduler: *id005
+train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
+  save_file: results/convtasnet-whamr/3/train_log.txt
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+  loadables:
+      encoder: !ref <Encoder>
+      masknet: !ref <MaskNet>
+      decoder: !ref <Decoder>

convtasnet2/lr_scheduler.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7efd3c42e7652fa5529e1d01e70d8faebd9d5b3939d0d94c719990ab12e5318f
+size 943

convtasnet2/masknet.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a43214e59d3f5b23936a9be023790a4920806186c2e814dc8f4dea13f3a5a2df
+size 26404523

convtasnet2/optimizer.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34e64b2e266108cd74b2ebd7a5039b1d6e39ca53ae03eab20e8dc123402e6c8e
+size 52803979

convtasnet3/CKPT.yaml

@@ -0,0 +1,4 @@
+# yamllint disable
+end-of-epoch: true
+si-snr: -5.491251103860138
+unixtime: 1631850893.625292

convtasnet3/brain.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e24193f36931b7f57932532efbdcf64971f42732383ba6808825f77db258f6
+size 28

convtasnet3/counter.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16dc368a89b428b2485484313ba67a3912ca03f2b2b42429174a4f8b3dc84e44
+size 3

convtasnet3/decoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35b76309d4da7851d953ad333fe2214c5f52625f09bd786c546f4152789c7710
+size 17272

convtasnet3/encoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9dcf0cc01b62e9bf3545eeccbc34547c718f6ae626d6f91bb6d8540ce418e15
+size 17272

convtasnet3/hyperparams.yaml

@@ -0,0 +1,168 @@
+# Generated 2021-09-15 from:
+# /home/mila/s/subakany/speechbrain_new/recipes/WHAMandWHAMR/separation/hparams/convtasnet-whamr.yaml
+# yamllint disable
+# ################################
+# Model: SepFormer for source separation
+# https://arxiv.org/abs/2010.13154
+#
+# Dataset : WHAMR!
+# ################################
+# Basic parameters
+# Seed needs to be set at top of yaml, before objects with parameters are made
+#
+seed: 3
+__set_seed: !apply:torch.manual_seed [3]
+
+# Data params
+
+# the data folder for the wham dataset
+# data_folder needs to follow the format:  /yourpath/whamr.
+# make sure to use the name whamr at your top folder for the dataset!
+data_folder: /network/tmp1/subakany/whamr
+
+# the path for wsj0/si_tr_s/ folder -- only needed if dynamic mixing is used
+# e.g. /yourpath/wsj0-processed/si_tr_s/
+# you need to convert the original wsj0 to 8k
+# you can do this conversion with the script ../meta/preprocess_dynamic_mixing.py
+base_folder_dm: /network/tmp1/subakany/wsj0-processed/si_tr_s/
+
+experiment_name: convtasnet-whamr
+output_folder: results/convtasnet-whamr/3
+train_log: results/convtasnet-whamr/3/train_log.txt
+save_folder: results/convtasnet-whamr/3/save
+
+# the file names should start with whamr instead of whamorg
+train_data: results/convtasnet-whamr/3/save/whamr_tr.csv
+valid_data: results/convtasnet-whamr/3/save/whamr_cv.csv
+test_data: results/convtasnet-whamr/3/save/whamr_tt.csv
+skip_prep: false
+
+# Experiment params
+auto_mix_prec: false # Set it to True for mixed precision
+test_only: false
+num_spks: 2 # set to 3 for wsj0-3mix
+progressbar: true
+save_audio: false # Save estimated sources on disk
+sample_rate: 8000
+
+# Training parameters
+N_epochs: 200
+batch_size: 1
+lr: 0.00015
+clip_grad_norm: 5
+loss_upper_lim: 999999  # this is the upper limit for an acceptable loss
+# if True, the training sequences are cut to a specified length
+limit_training_signal_len: false
+# this is the length of sequences if we choose to limit
+# the signal length of training sequences
+training_signal_len: 32000000
+
+# Set it to True to dynamically create mixtures at training time
+dynamic_mixing: true
+
+# Parameters for data augmentation
+
+# rir_path variable points to the directory of the room impulse responses
+# e.g. /miniscratch/subakany/rir_wavs
+# If the path does not exist, it is created automatically.
+rir_path: /miniscratch/subakany/whamr_rirs_wav
+
+use_wavedrop: false
+use_speedperturb: true
+use_speedperturb_sameforeachsource: false
+use_rand_shift: false
+min_shift: -8000
+max_shift: 8000
+
+speedperturb: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 1.0
+  drop_freq_prob: 0.0
+  drop_chunk_prob: 0.0
+  sample_rate: 8000
+  speeds: [95, 100, 105]
+
+wavedrop: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 0.0
+  drop_freq_prob: 1.0
+  drop_chunk_prob: 1.0
+  sample_rate: 8000
+
+# loss thresholding -- this thresholds the training loss
+threshold_byloss: true
+threshold: -30
+
+# Encoder parameters
+N_encoder_out: 256
+out_channels: 256
+kernel_size: 16
+kernel_stride: 8
+
+# Dataloader options
+dataloader_opts:
+  batch_size: 1
+  num_workers: 3
+
+
+# Specifying the network
+Encoder: &id001 !new:speechbrain.lobes.models.dual_path.Encoder
+  kernel_size: 16
+  out_channels: 256
+
+
+MaskNet: &id003 !new:speechbrain.lobes.models.conv_tasnet.MaskNet
+
+  N: 256
+  B: 256
+  H: 512
+  P: 3
+  X: 6
+  R: 4
+  C: 2
+  norm_type: gLN
+  causal: false
+  mask_nonlinear: relu
+
+Decoder: &id002 !new:speechbrain.lobes.models.dual_path.Decoder
+  in_channels: 256
+  out_channels: 1
+  kernel_size: 16
+  stride: 8
+  bias: false
+
+
+optimizer: !name:torch.optim.Adam
+  lr: 0.00015
+  weight_decay: 0
+
+loss: !name:speechbrain.nnet.losses.get_si_snr_with_pitwrapper
+
+lr_scheduler: &id005 !new:speechbrain.nnet.schedulers.ReduceLROnPlateau
+
+  factor: 0.5
+  patience: 2
+  dont_halve_until_epoch: 85
+
+epoch_counter: &id004 !new:speechbrain.utils.epoch_loop.EpochCounter
+  limit: 200
+
+modules:
+  encoder: *id001
+  decoder: *id002
+  masknet: *id003
+save_all_checkpoints: true
+checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
+  checkpoints_dir: results/convtasnet-whamr/3/save
+  recoverables:
+    encoder: *id001
+    decoder: *id002
+    masknet: *id003
+    counter: *id004
+    lr_scheduler: *id005
+train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
+  save_file: results/convtasnet-whamr/3/train_log.txt
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+  loadables:
+      encoder: !ref <Encoder>
+      masknet: !ref <MaskNet>
+      decoder: !ref <Decoder>

convtasnet3/lr_scheduler.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca2047f7f167981a8323fedc60c915590115468bd222ce33c19871f495b811f2
+size 1391

convtasnet3/masknet.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f14d10e6f2e7d4eb7770f3938fac171dfb86d46bce7082fbfee400d95dfe5cc5
+size 26404523

convtasnet3/optimizer.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17f618f7f980ba67fb00fe0fb3e13dafb0e9ec5111ffb945e94c53fdc7c7aac3
+size 52803979

dprnn1/CKPT.yaml

@@ -0,0 +1,4 @@
+# yamllint disable
+end-of-epoch: true
+si-snr: -0.5297146898724144
+unixtime: 1631292299.804519

dprnn1/brain.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e24193f36931b7f57932532efbdcf64971f42732383ba6808825f77db258f6
+size 28

dprnn1/counter.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b86b273ff34fce19d6b804eff5a3f5747ada4eaa22f1d49c01e52ddb7875b4b
+size 1

dprnn1/dataloader-TRAIN.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:876c9b16254e157d1eb645390dcfae6f29b9d3cd394e73a91de8ee5d0e67ee43
+size 5

dprnn1/decoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2e73c51dc6b7bdcc3a546af3e2f682dc4d3f7357d1497b5e545401a3cb517b1
+size 17272

dprnn1/encoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdf0dbef1da288e831b4facdfa428de1c3bac0dfefe6c08a127c6e09c60d148b
+size 17272

dprnn1/hyperparams.yaml

@@ -0,0 +1,183 @@
+# Generated 2021-09-18 from:
+# /home/mila/s/subakany/speechbrain_new/recipes/WHAMandWHAMR/separation/hparams/dprnn-whamr.yaml
+# yamllint disable
+# ################################
+# Model: SepFormer for source separation
+# https://arxiv.org/abs/2010.13154
+#
+# Dataset : WHAMR!
+# ################################
+# Basic parameters
+# Seed needs to be set at top of yaml, before objects with parameters are made
+#
+seed: 3
+__set_seed: !apply:torch.manual_seed [3]
+
+# Data params
+
+# the data folder for the wham dataset
+# data_folder needs to follow the format:  /yourpath/whamr.
+# make sure to use the name whamr at your top folder for the dataset!
+data_folder: /network/tmp1/subakany/whamr/
+
+# the path for wsj0/si_tr_s/ folder -- only needed if dynamic mixing is used
+# e.g. /yourpath/wsj0-processed/si_tr_s/
+# you need to convert the original wsj0 to 8k
+# you can do this conversion with the script ../meta/preprocess_dynamic_mixing.py
+base_folder_dm: /network/tmp1/subakany/wsj0-processed/si_tr_s/
+
+experiment_name: dprnn-whamr
+output_folder: results/dprnn-whamr/3
+train_log: results/dprnn-whamr/3/train_log.txt
+save_folder: results/dprnn-whamr/3/save
+
+# the file names should start with whamr instead of whamorg
+train_data: results/dprnn-whamr/3/save/whamr_tr.csv
+valid_data: results/dprnn-whamr/3/save/whamr_cv.csv
+test_data: results/dprnn-whamr/3/save/whamr_tt.csv
+skip_prep: false
+
+# Experiment params
+auto_mix_prec: true # Set it to True for mixed precision
+test_only: false
+num_spks: 2 # set to 3 for wsj0-3mix
+progressbar: true
+save_audio: false # Save estimated sources on disk
+sample_rate: 8000
+
+# Training parameters
+N_epochs: 200
+batch_size: 1
+lr: 0.00015
+clip_grad_norm: 5
+loss_upper_lim: 999999  # this is the upper limit for an acceptable loss
+# if True, the training sequences are cut to a specified length
+limit_training_signal_len: false
+# this is the length of sequences if we choose to limit
+# the signal length of training sequences
+training_signal_len: 32000000
+
+# Set it to True to dynamically create mixtures at training time
+dynamic_mixing: true
+
+# Parameters for data augmentation
+
+# rir_path variable points to the directory of the room impulse responses
+# e.g. /miniscratch/subakany/rir_wavs
+# If the path does not exist, it is created automatically.
+rir_path: /miniscratch/subakany/whamr_rirs_wav
+
+use_wavedrop: false
+use_speedperturb: true
+use_speedperturb_sameforeachsource: false
+use_rand_shift: false
+min_shift: -8000
+max_shift: 8000
+
+speedperturb: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 1.0
+  drop_freq_prob: 0.0
+  drop_chunk_prob: 0.0
+  sample_rate: 8000
+  speeds: [95, 100, 105]
+
+wavedrop: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 0.0
+  drop_freq_prob: 1.0
+  drop_chunk_prob: 1.0
+  sample_rate: 8000
+
+# loss thresholding -- this thresholds the training loss
+threshold_byloss: true
+threshold: -30
+
+# Encoder parameters
+N_encoder_out: 256
+out_channels: 256
+kernel_size: 16
+kernel_stride: 8
+
+# Dataloader options
+dataloader_opts:
+  batch_size: 1
+  num_workers: 3
+
+
+# Specifying the network
+Encoder: &id003 !new:speechbrain.lobes.models.dual_path.Encoder
+  kernel_size: 16
+  out_channels: 256
+
+
+intra: &id001 !new:speechbrain.lobes.models.dual_path.SBRNNBlock
+  num_layers: 1
+  input_size: 256
+  hidden_channels: 256
+  dropout: 0
+  bidirectional: true
+
+inter: &id002 !new:speechbrain.lobes.models.dual_path.SBRNNBlock
+  num_layers: 1
+  input_size: 256
+  hidden_channels: 256
+  dropout: 0
+  bidirectional: true
+
+MaskNet: &id005 !new:speechbrain.lobes.models.dual_path.Dual_Path_Model
+
+  num_spks: 2
+  in_channels: 256
+  out_channels: 256
+  num_layers: 6
+  K: 250
+  intra_model: *id001
+  inter_model: *id002
+  norm: ln
+  linear_layer_after_inter_intra: true
+  skip_around_intra: true
+
+Decoder: &id004 !new:speechbrain.lobes.models.dual_path.Decoder
+  in_channels: 256
+  out_channels: 1
+  kernel_size: 16
+  stride: 8
+  bias: false
+
+
+
+optimizer: !name:torch.optim.Adam
+  lr: 0.00015
+  weight_decay: 0
+
+loss: !name:speechbrain.nnet.losses.get_si_snr_with_pitwrapper
+
+lr_scheduler: &id007 !new:speechbrain.nnet.schedulers.ReduceLROnPlateau
+
+  factor: 0.5
+  patience: 2
+  dont_halve_until_epoch: 85
+
+epoch_counter: &id006 !new:speechbrain.utils.epoch_loop.EpochCounter
+  limit: 200
+
+modules:
+  encoder: *id003
+  decoder: *id004
+  masknet: *id005
+save_all_checkpoints: true
+checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
+  checkpoints_dir: results/dprnn-whamr/3/save
+  recoverables:
+    encoder: *id003
+    decoder: *id004
+    masknet: *id005
+    counter: *id006
+    lr_scheduler: *id007
+train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
+  save_file: results/dprnn-whamr/3/train_log.txt
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+   loadables:
+      encoder: !ref <Encoder>
+      masknet: !ref <MaskNet>
+      decoder: !ref <Decoder>

dprnn1/lr_scheduler.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e124452ee7deeb68cdaa636fbcb85a7654dfda37ac2b38ccebf02d5604de85dc
+size 495

dprnn1/masknet.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:427819af9b78e9700010f6cbe84af6c33cb5b229762018c76c3ba9117c4f8761
+size 58477294

dprnn1/optimizer.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52cdd192d0a27ca7a9a065505c30cb0aa025abf4afdc88fbcf006249929514e7
+size 117035321

dprnn2/CKPT.yaml

@@ -0,0 +1,4 @@
+# yamllint disable
+end-of-epoch: true
+si-snr: -5.874772780568888
+unixtime: 1631748976.9005034

dprnn2/brain.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e24193f36931b7f57932532efbdcf64971f42732383ba6808825f77db258f6
+size 28

dprnn2/counter.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fca346db656187102ce806ac732e06a62df0dbb2829e511a770556d398e1a6e
+size 2

dprnn2/decoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74fac8e4c0330aca71afef77553f5da11fa5a89bff026e9fe8e5fd529ddc3b9a
+size 17272

dprnn2/encoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aef9291a6d0d0a74955bf539c81d36a35bdc6900306d5d1c69dca57b0b315e4c
+size 17272

dprnn2/hyperparams.yaml

@@ -0,0 +1,183 @@
+# Generated 2021-09-18 from:
+# /home/mila/s/subakany/speechbrain_new/recipes/WHAMandWHAMR/separation/hparams/dprnn-whamr.yaml
+# yamllint disable
+# ################################
+# Model: SepFormer for source separation
+# https://arxiv.org/abs/2010.13154
+#
+# Dataset : WHAMR!
+# ################################
+# Basic parameters
+# Seed needs to be set at top of yaml, before objects with parameters are made
+#
+seed: 3
+__set_seed: !apply:torch.manual_seed [3]
+
+# Data params
+
+# the data folder for the wham dataset
+# data_folder needs to follow the format:  /yourpath/whamr.
+# make sure to use the name whamr at your top folder for the dataset!
+data_folder: /network/tmp1/subakany/whamr/
+
+# the path for wsj0/si_tr_s/ folder -- only needed if dynamic mixing is used
+# e.g. /yourpath/wsj0-processed/si_tr_s/
+# you need to convert the original wsj0 to 8k
+# you can do this conversion with the script ../meta/preprocess_dynamic_mixing.py
+base_folder_dm: /network/tmp1/subakany/wsj0-processed/si_tr_s/
+
+experiment_name: dprnn-whamr
+output_folder: results/dprnn-whamr/3
+train_log: results/dprnn-whamr/3/train_log.txt
+save_folder: results/dprnn-whamr/3/save
+
+# the file names should start with whamr instead of whamorg
+train_data: results/dprnn-whamr/3/save/whamr_tr.csv
+valid_data: results/dprnn-whamr/3/save/whamr_cv.csv
+test_data: results/dprnn-whamr/3/save/whamr_tt.csv
+skip_prep: false
+
+# Experiment params
+auto_mix_prec: true # Set it to True for mixed precision
+test_only: false
+num_spks: 2 # set to 3 for wsj0-3mix
+progressbar: true
+save_audio: false # Save estimated sources on disk
+sample_rate: 8000
+
+# Training parameters
+N_epochs: 200
+batch_size: 1
+lr: 0.00015
+clip_grad_norm: 5
+loss_upper_lim: 999999  # this is the upper limit for an acceptable loss
+# if True, the training sequences are cut to a specified length
+limit_training_signal_len: false
+# this is the length of sequences if we choose to limit
+# the signal length of training sequences
+training_signal_len: 32000000
+
+# Set it to True to dynamically create mixtures at training time
+dynamic_mixing: true
+
+# Parameters for data augmentation
+
+# rir_path variable points to the directory of the room impulse responses
+# e.g. /miniscratch/subakany/rir_wavs
+# If the path does not exist, it is created automatically.
+rir_path: /miniscratch/subakany/whamr_rirs_wav
+
+use_wavedrop: false
+use_speedperturb: true
+use_speedperturb_sameforeachsource: false
+use_rand_shift: false
+min_shift: -8000
+max_shift: 8000
+
+speedperturb: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 1.0
+  drop_freq_prob: 0.0
+  drop_chunk_prob: 0.0
+  sample_rate: 8000
+  speeds: [95, 100, 105]
+
+wavedrop: !new:speechbrain.lobes.augment.TimeDomainSpecAugment
+  perturb_prob: 0.0
+  drop_freq_prob: 1.0
+  drop_chunk_prob: 1.0
+  sample_rate: 8000
+
+# loss thresholding -- this thresholds the training loss
+threshold_byloss: true
+threshold: -30
+
+# Encoder parameters
+N_encoder_out: 256
+out_channels: 256
+kernel_size: 16
+kernel_stride: 8
+
+# Dataloader options
+dataloader_opts:
+  batch_size: 1
+  num_workers: 3
+
+
+# Specifying the network
+Encoder: &id003 !new:speechbrain.lobes.models.dual_path.Encoder
+  kernel_size: 16
+  out_channels: 256
+
+
+intra: &id001 !new:speechbrain.lobes.models.dual_path.SBRNNBlock
+  num_layers: 1
+  input_size: 256
+  hidden_channels: 256
+  dropout: 0
+  bidirectional: true
+
+inter: &id002 !new:speechbrain.lobes.models.dual_path.SBRNNBlock
+  num_layers: 1
+  input_size: 256
+  hidden_channels: 256
+  dropout: 0
+  bidirectional: true
+
+MaskNet: &id005 !new:speechbrain.lobes.models.dual_path.Dual_Path_Model
+
+  num_spks: 2
+  in_channels: 256
+  out_channels: 256
+  num_layers: 6
+  K: 250
+  intra_model: *id001
+  inter_model: *id002
+  norm: ln
+  linear_layer_after_inter_intra: true
+  skip_around_intra: true
+
+Decoder: &id004 !new:speechbrain.lobes.models.dual_path.Decoder
+  in_channels: 256
+  out_channels: 1
+  kernel_size: 16
+  stride: 8
+  bias: false
+
+
+
+optimizer: !name:torch.optim.Adam
+  lr: 0.00015
+  weight_decay: 0
+
+loss: !name:speechbrain.nnet.losses.get_si_snr_with_pitwrapper
+
+lr_scheduler: &id007 !new:speechbrain.nnet.schedulers.ReduceLROnPlateau
+
+  factor: 0.5
+  patience: 2
+  dont_halve_until_epoch: 85
+
+epoch_counter: &id006 !new:speechbrain.utils.epoch_loop.EpochCounter
+  limit: 200
+
+modules:
+  encoder: *id003
+  decoder: *id004
+  masknet: *id005
+save_all_checkpoints: true
+checkpointer: !new:speechbrain.utils.checkpoints.Checkpointer
+  checkpoints_dir: results/dprnn-whamr/3/save
+  recoverables:
+    encoder: *id003
+    decoder: *id004
+    masknet: *id005
+    counter: *id006
+    lr_scheduler: *id007
+train_logger: !new:speechbrain.utils.train_logger.FileTrainLogger
+  save_file: results/dprnn-whamr/3/train_log.txt
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+   loadables:
+      encoder: !ref <Encoder>
+      masknet: !ref <MaskNet>
+      decoder: !ref <Decoder>

dprnn2/lr_scheduler.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb3b7d061e814f1fa9b59ec858d9f1c346671ce3d81c15b3a1a176a371b52037
+size 943

dprnn2/masknet.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dbfce012dc85b68b3cc9f8e1d10b1fc63ef54248b34d562f61bb8c0f7a23030
+size 58477294

dprnn2/optimizer.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7323ab9ed7747b6bb0dd116f2e7f6271d1f6f9d28971644210e5016fa7ee033
+size 117035641

dprnn3/CKPT.yaml

@@ -0,0 +1,4 @@
+# yamllint disable
+end-of-epoch: true
+si-snr: -7.044497849920388
+unixtime: 1632151547.7435942

dprnn3/brain.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e24193f36931b7f57932532efbdcf64971f42732383ba6808825f77db258f6
+size 28

dprnn3/counter.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3346f2bbf6c34bd2dbe28bd1bb657d0e9c37392a1d5ec9929e6a5df4763ddc2d
+size 3

dprnn3/decoder.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d26e5412c04d918938a2b5a0e310d814fcf1451c3c7e9dda9f52b549f553f64
+size 17272