Training in progress, step 2500, checkpoint

checkpoint-2500/added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<ctc_blank>": 80,
+  "<mask>": 79
+}

checkpoint-2500/config.json

@@ -0,0 +1,92 @@
+{
+  "_name_or_path": "microsoft/speecht5_tts",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "SpeechT5ForTextToSpeech"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 0,
+  "conv_bias": false,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "decoder_attention_heads": 12,
+  "decoder_ffn_dim": 3072,
+  "decoder_layerdrop": 0.1,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 2,
+  "encoder_attention_heads": 12,
+  "encoder_ffn_dim": 3072,
+  "encoder_layerdrop": 0.1,
+  "encoder_layers": 12,
+  "encoder_max_relative_position": 160,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.0,
+  "guided_attention_loss_num_heads": 2,
+  "guided_attention_loss_scale": 10.0,
+  "guided_attention_loss_sigma": 0.4,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "is_encoder_decoder": true,
+  "layer_norm_eps": 1e-05,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "max_length": 1876,
+  "max_speech_positions": 1876,
+  "max_text_positions": 600,
+  "model_type": "speecht5",
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_mel_bins": 80,
+  "pad_token_id": 1,
+  "positional_dropout": 0.1,
+  "reduction_factor": 2,
+  "scale_embedding": false,
+  "speaker_embedding_dim": 512,
+  "speech_decoder_postnet_dropout": 0.5,
+  "speech_decoder_postnet_kernel": 5,
+  "speech_decoder_postnet_layers": 5,
+  "speech_decoder_postnet_units": 256,
+  "speech_decoder_prenet_dropout": 0.5,
+  "speech_decoder_prenet_layers": 2,
+  "speech_decoder_prenet_units": 256,
+  "torch_dtype": "float32",
+  "transformers_version": "4.41.0.dev0",
+  "use_cache": false,
+  "use_guided_attention_loss": true,
+  "vocab_size": 81
+}

checkpoint-2500/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "eos_token_id": 2,
+  "max_length": 1876,
+  "pad_token_id": 1,
+  "transformers_version": "4.41.0.dev0"
+}

checkpoint-2500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e27740e167330f964d33c6005b3ed2c425fb87deafab1f03d6b35dc2bc755632
+size 577789320

checkpoint-2500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5e0876fa6cc161cf74d3e04b86eaeabfdf2588337651e05fba4817d6b08242f
+size 1155772233

checkpoint-2500/preprocessor_config.json

@@ -0,0 +1,19 @@
+{
+  "do_normalize": false,
+  "feature_extractor_type": "SpeechT5FeatureExtractor",
+  "feature_size": 1,
+  "fmax": 7600,
+  "fmin": 80,
+  "frame_signal_scale": 1.0,
+  "hop_length": 16,
+  "mel_floor": 1e-10,
+  "num_mel_bins": 80,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "SpeechT5Processor",
+  "reduction_factor": 2,
+  "return_attention_mask": true,
+  "sampling_rate": 16000,
+  "win_function": "hann_window",
+  "win_length": 64
+}

checkpoint-2500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0a3df38c17b475ea7e47b6c462cb3838a91cdec2a3f47a69fc35d66481cfa21
+size 14244

checkpoint-2500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a6e1d45386e11e147f95efa198dfdb563b15409648099710720a92d2030eff3
+size 1064

checkpoint-2500/special_tokens_map.json

@@ -0,0 +1,13 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-2500/spm_char.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
+size 238473

checkpoint-2500/tokenizer_config.json

@@ -0,0 +1,63 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<ctc_blank>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 600,
+  "normalize": false,
+  "pad_token": "<pad>",
+  "processor_class": "SpeechT5Processor",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "SpeechT5Tokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-2500/trainer_state.json

@@ -0,0 +1,423 @@
+{
+  "best_metric": 0.3522670865058899,
+  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le4_s4000/checkpoint-2500",
+  "epoch": 4.18848167539267,
+  "eval_steps": 500,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08376963350785341,
+      "grad_norm": 2.9717624187469482,
+      "learning_rate": 2.4500000000000003e-06,
+      "loss": 1.0424,
+      "step": 50
+    },
+    {
+      "epoch": 0.16753926701570682,
+      "grad_norm": 2.9720630645751953,
+      "learning_rate": 4.950000000000001e-06,
+      "loss": 0.8474,
+      "step": 100
+    },
+    {
+      "epoch": 0.2513089005235602,
+      "grad_norm": 2.445929765701294,
+      "learning_rate": 7.45e-06,
+      "loss": 0.7336,
+      "step": 150
+    },
+    {
+      "epoch": 0.33507853403141363,
+      "grad_norm": 5.502955913543701,
+      "learning_rate": 9.950000000000001e-06,
+      "loss": 0.6492,
+      "step": 200
+    },
+    {
+      "epoch": 0.418848167539267,
+      "grad_norm": 2.3356130123138428,
+      "learning_rate": 1.2450000000000001e-05,
+      "loss": 0.6133,
+      "step": 250
+    },
+    {
+      "epoch": 0.5026178010471204,
+      "grad_norm": 1.937270164489746,
+      "learning_rate": 1.4950000000000001e-05,
+      "loss": 0.5889,
+      "step": 300
+    },
+    {
+      "epoch": 0.5863874345549738,
+      "grad_norm": 2.392244338989258,
+      "learning_rate": 1.745e-05,
+      "loss": 0.5694,
+      "step": 350
+    },
+    {
+      "epoch": 0.6701570680628273,
+      "grad_norm": 7.3209919929504395,
+      "learning_rate": 1.995e-05,
+      "loss": 0.5477,
+      "step": 400
+    },
+    {
+      "epoch": 0.7539267015706806,
+      "grad_norm": 3.415917158126831,
+      "learning_rate": 2.245e-05,
+      "loss": 0.5329,
+      "step": 450
+    },
+    {
+      "epoch": 0.837696335078534,
+      "grad_norm": 3.0256705284118652,
+      "learning_rate": 2.495e-05,
+      "loss": 0.5173,
+      "step": 500
+    },
+    {
+      "epoch": 0.837696335078534,
+      "eval_loss": 0.4566049873828888,
+      "eval_runtime": 261.3511,
+      "eval_samples_per_second": 32.481,
+      "eval_steps_per_second": 4.063,
+      "step": 500
+    },
+    {
+      "epoch": 0.9214659685863874,
+      "grad_norm": 1.9436837434768677,
+      "learning_rate": 2.7450000000000003e-05,
+      "loss": 0.5079,
+      "step": 550
+    },
+    {
+      "epoch": 1.0052356020942408,
+      "grad_norm": 1.819956660270691,
+      "learning_rate": 2.995e-05,
+      "loss": 0.4969,
+      "step": 600
+    },
+    {
+      "epoch": 1.0890052356020943,
+      "grad_norm": 5.457251071929932,
+      "learning_rate": 3.245e-05,
+      "loss": 0.4977,
+      "step": 650
+    },
+    {
+      "epoch": 1.1727748691099475,
+      "grad_norm": 3.183980703353882,
+      "learning_rate": 3.495e-05,
+      "loss": 0.4923,
+      "step": 700
+    },
+    {
+      "epoch": 1.256544502617801,
+      "grad_norm": 7.1660051345825195,
+      "learning_rate": 3.745e-05,
+      "loss": 0.4802,
+      "step": 750
+    },
+    {
+      "epoch": 1.3403141361256545,
+      "grad_norm": 5.499026775360107,
+      "learning_rate": 3.995e-05,
+      "loss": 0.4754,
+      "step": 800
+    },
+    {
+      "epoch": 1.4240837696335078,
+      "grad_norm": 2.8053908348083496,
+      "learning_rate": 4.245e-05,
+      "loss": 0.4669,
+      "step": 850
+    },
+    {
+      "epoch": 1.5078534031413613,
+      "grad_norm": 3.017005443572998,
+      "learning_rate": 4.495e-05,
+      "loss": 0.4604,
+      "step": 900
+    },
+    {
+      "epoch": 1.5916230366492146,
+      "grad_norm": 2.7971177101135254,
+      "learning_rate": 4.745e-05,
+      "loss": 0.4565,
+      "step": 950
+    },
+    {
+      "epoch": 1.675392670157068,
+      "grad_norm": 3.1588356494903564,
+      "learning_rate": 4.995e-05,
+      "loss": 0.455,
+      "step": 1000
+    },
+    {
+      "epoch": 1.675392670157068,
+      "eval_loss": 0.40312233567237854,
+      "eval_runtime": 256.4334,
+      "eval_samples_per_second": 33.104,
+      "eval_steps_per_second": 4.141,
+      "step": 1000
+    },
+    {
+      "epoch": 1.7591623036649215,
+      "grad_norm": 2.2053232192993164,
+      "learning_rate": 5.245e-05,
+      "loss": 0.4543,
+      "step": 1050
+    },
+    {
+      "epoch": 1.8429319371727748,
+      "grad_norm": 2.0562164783477783,
+      "learning_rate": 5.495e-05,
+      "loss": 0.4456,
+      "step": 1100
+    },
+    {
+      "epoch": 1.9267015706806283,
+      "grad_norm": 2.730119466781616,
+      "learning_rate": 5.745e-05,
+      "loss": 0.4355,
+      "step": 1150
+    },
+    {
+      "epoch": 2.0104712041884816,
+      "grad_norm": 1.7484283447265625,
+      "learning_rate": 5.995000000000001e-05,
+      "loss": 0.4299,
+      "step": 1200
+    },
+    {
+      "epoch": 2.094240837696335,
+      "grad_norm": 1.1786061525344849,
+      "learning_rate": 6.245000000000001e-05,
+      "loss": 0.4305,
+      "step": 1250
+    },
+    {
+      "epoch": 2.1780104712041886,
+      "grad_norm": 1.98978590965271,
+      "learning_rate": 6.494999999999999e-05,
+      "loss": 0.4295,
+      "step": 1300
+    },
+    {
+      "epoch": 2.261780104712042,
+      "grad_norm": 2.818659782409668,
+      "learning_rate": 6.745e-05,
+      "loss": 0.4235,
+      "step": 1350
+    },
+    {
+      "epoch": 2.345549738219895,
+      "grad_norm": 2.3864262104034424,
+      "learning_rate": 6.995e-05,
+      "loss": 0.4271,
+      "step": 1400
+    },
+    {
+      "epoch": 2.4293193717277486,
+      "grad_norm": 1.3647903203964233,
+      "learning_rate": 7.245000000000001e-05,
+      "loss": 0.4208,
+      "step": 1450
+    },
+    {
+      "epoch": 2.513089005235602,
+      "grad_norm": 2.2144172191619873,
+      "learning_rate": 7.495e-05,
+      "loss": 0.4175,
+      "step": 1500
+    },
+    {
+      "epoch": 2.513089005235602,
+      "eval_loss": 0.3777858018875122,
+      "eval_runtime": 260.5025,
+      "eval_samples_per_second": 32.587,
+      "eval_steps_per_second": 4.077,
+      "step": 1500
+    },
+    {
+      "epoch": 2.5968586387434556,
+      "grad_norm": 1.6483193635940552,
+      "learning_rate": 7.745e-05,
+      "loss": 0.414,
+      "step": 1550
+    },
+    {
+      "epoch": 2.680628272251309,
+      "grad_norm": 1.7688554525375366,
+      "learning_rate": 7.995e-05,
+      "loss": 0.4153,
+      "step": 1600
+    },
+    {
+      "epoch": 2.7643979057591626,
+      "grad_norm": 1.2314317226409912,
+      "learning_rate": 8.245e-05,
+      "loss": 0.4089,
+      "step": 1650
+    },
+    {
+      "epoch": 2.8481675392670156,
+      "grad_norm": 1.6623793840408325,
+      "learning_rate": 8.495e-05,
+      "loss": 0.4124,
+      "step": 1700
+    },
+    {
+      "epoch": 2.931937172774869,
+      "grad_norm": 3.812507390975952,
+      "learning_rate": 8.745000000000001e-05,
+      "loss": 0.4112,
+      "step": 1750
+    },
+    {
+      "epoch": 3.0157068062827226,
+      "grad_norm": 2.141019821166992,
+      "learning_rate": 8.995e-05,
+      "loss": 0.4081,
+      "step": 1800
+    },
+    {
+      "epoch": 3.099476439790576,
+      "grad_norm": 1.8928133249282837,
+      "learning_rate": 9.245e-05,
+      "loss": 0.4067,
+      "step": 1850
+    },
+    {
+      "epoch": 3.183246073298429,
+      "grad_norm": 2.322817087173462,
+      "learning_rate": 9.495e-05,
+      "loss": 0.4088,
+      "step": 1900
+    },
+    {
+      "epoch": 3.2670157068062826,
+      "grad_norm": 2.1984918117523193,
+      "learning_rate": 9.745000000000001e-05,
+      "loss": 0.3976,
+      "step": 1950
+    },
+    {
+      "epoch": 3.350785340314136,
+      "grad_norm": 2.0455121994018555,
+      "learning_rate": 9.995e-05,
+      "loss": 0.4022,
+      "step": 2000
+    },
+    {
+      "epoch": 3.350785340314136,
+      "eval_loss": 0.3677983582019806,
+      "eval_runtime": 257.3111,
+      "eval_samples_per_second": 32.991,
+      "eval_steps_per_second": 4.127,
+      "step": 2000
+    },
+    {
+      "epoch": 3.4345549738219896,
+      "grad_norm": 2.0729174613952637,
+      "learning_rate": 9.755000000000001e-05,
+      "loss": 0.4017,
+      "step": 2050
+    },
+    {
+      "epoch": 3.518324607329843,
+      "grad_norm": 1.4496002197265625,
+      "learning_rate": 9.505e-05,
+      "loss": 0.3999,
+      "step": 2100
+    },
+    {
+      "epoch": 3.6020942408376966,
+      "grad_norm": 1.313783884048462,
+      "learning_rate": 9.255e-05,
+      "loss": 0.3971,
+      "step": 2150
+    },
+    {
+      "epoch": 3.6858638743455496,
+      "grad_norm": 2.4938676357269287,
+      "learning_rate": 9.005000000000001e-05,
+      "loss": 0.3958,
+      "step": 2200
+    },
+    {
+      "epoch": 3.769633507853403,
+      "grad_norm": 1.6081351041793823,
+      "learning_rate": 8.755e-05,
+      "loss": 0.389,
+      "step": 2250
+    },
+    {
+      "epoch": 3.8534031413612566,
+      "grad_norm": 1.5322096347808838,
+      "learning_rate": 8.505000000000001e-05,
+      "loss": 0.3887,
+      "step": 2300
+    },
+    {
+      "epoch": 3.93717277486911,
+      "grad_norm": 2.3951408863067627,
+      "learning_rate": 8.26e-05,
+      "loss": 0.392,
+      "step": 2350
+    },
+    {
+      "epoch": 4.020942408376963,
+      "grad_norm": 1.0958179235458374,
+      "learning_rate": 8.010000000000001e-05,
+      "loss": 0.3846,
+      "step": 2400
+    },
+    {
+      "epoch": 4.104712041884817,
+      "grad_norm": 1.2315720319747925,
+      "learning_rate": 7.76e-05,
+      "loss": 0.3867,
+      "step": 2450
+    },
+    {
+      "epoch": 4.18848167539267,
+      "grad_norm": 1.3186556100845337,
+      "learning_rate": 7.510000000000001e-05,
+      "loss": 0.3848,
+      "step": 2500
+    },
+    {
+      "epoch": 4.18848167539267,
+      "eval_loss": 0.3522670865058899,
+      "eval_runtime": 269.3867,
+      "eval_samples_per_second": 31.512,
+      "eval_steps_per_second": 3.942,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 4000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.477753664307475e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5484e04f6ef3fa6c1a7fb861e2b263b28356830486ed692cff98e0b32f9f8b6f
+size 5304