Training in progress, step 1500, checkpoint

checkpoint-1500/added_tokens.json

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

checkpoint-1500/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-1500/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-1500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89156d472a3fe41e0dff6ba9f0f276cd4f49fba4b1f1f8d5416ee7c4b2018ef1
+size 577789320

checkpoint-1500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a0f98b115d072b6b5b4fa87cc8ed90d0e8a4711ddd299caa8a7a3bfab600930
+size 1155772233

checkpoint-1500/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-1500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03a6c12dbbc6e6898ce1589d586631bf0117e1b74d14f57ce4fc66d31edc6637
+size 14244

checkpoint-1500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d536ffd212ec2d6c7b7e2554122dbdcf42d544b077a72c397931a02ddb8f6ab4
+size 1064

checkpoint-1500/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-1500/spm_char.model

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

checkpoint-1500/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-1500/trainer_state.json

@@ -0,0 +1,267 @@
+{
+  "best_metric": 0.40509456396102905,
+  "best_model_checkpoint": "mikhail_panzo/fil_b128_le4_s4000/checkpoint-1500",
+  "epoch": 133.33333333333334,
+  "eval_steps": 500,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 2.152308940887451,
+      "learning_rate": 2.5e-06,
+      "loss": 0.7738,
+      "step": 50
+    },
+    {
+      "epoch": 8.88888888888889,
+      "grad_norm": 1.326554536819458,
+      "learning_rate": 5e-06,
+      "loss": 0.6968,
+      "step": 100
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 1.4387542009353638,
+      "learning_rate": 7.5e-06,
+      "loss": 0.636,
+      "step": 150
+    },
+    {
+      "epoch": 17.77777777777778,
+      "grad_norm": 1.033190131187439,
+      "learning_rate": 1e-05,
+      "loss": 0.5486,
+      "step": 200
+    },
+    {
+      "epoch": 22.22222222222222,
+      "grad_norm": 1.4383914470672607,
+      "learning_rate": 1.25e-05,
+      "loss": 0.516,
+      "step": 250
+    },
+    {
+      "epoch": 26.666666666666668,
+      "grad_norm": 1.8267806768417358,
+      "learning_rate": 1.5e-05,
+      "loss": 0.4989,
+      "step": 300
+    },
+    {
+      "epoch": 31.11111111111111,
+      "grad_norm": 1.2525017261505127,
+      "learning_rate": 1.75e-05,
+      "loss": 0.4908,
+      "step": 350
+    },
+    {
+      "epoch": 35.55555555555556,
+      "grad_norm": 1.1662768125534058,
+      "learning_rate": 2e-05,
+      "loss": 0.4766,
+      "step": 400
+    },
+    {
+      "epoch": 40.0,
+      "grad_norm": 1.110721230506897,
+      "learning_rate": 2.25e-05,
+      "loss": 0.4716,
+      "step": 450
+    },
+    {
+      "epoch": 44.44444444444444,
+      "grad_norm": 1.6693321466445923,
+      "learning_rate": 2.5e-05,
+      "loss": 0.4635,
+      "step": 500
+    },
+    {
+      "epoch": 44.44444444444444,
+      "eval_loss": 0.42072969675064087,
+      "eval_runtime": 9.5188,
+      "eval_samples_per_second": 16.704,
+      "eval_steps_per_second": 2.101,
+      "step": 500
+    },
+    {
+      "epoch": 48.888888888888886,
+      "grad_norm": 1.0238850116729736,
+      "learning_rate": 2.7500000000000004e-05,
+      "loss": 0.4585,
+      "step": 550
+    },
+    {
+      "epoch": 53.333333333333336,
+      "grad_norm": 0.8729687929153442,
+      "learning_rate": 3e-05,
+      "loss": 0.4572,
+      "step": 600
+    },
+    {
+      "epoch": 57.77777777777778,
+      "grad_norm": 1.4291303157806396,
+      "learning_rate": 3.2500000000000004e-05,
+      "loss": 0.452,
+      "step": 650
+    },
+    {
+      "epoch": 62.22222222222222,
+      "grad_norm": 1.3374826908111572,
+      "learning_rate": 3.5e-05,
+      "loss": 0.447,
+      "step": 700
+    },
+    {
+      "epoch": 66.66666666666667,
+      "grad_norm": 1.7066700458526611,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.4424,
+      "step": 750
+    },
+    {
+      "epoch": 71.11111111111111,
+      "grad_norm": 2.1874194145202637,
+      "learning_rate": 4e-05,
+      "loss": 0.4443,
+      "step": 800
+    },
+    {
+      "epoch": 75.55555555555556,
+      "grad_norm": 1.4660208225250244,
+      "learning_rate": 4.25e-05,
+      "loss": 0.4346,
+      "step": 850
+    },
+    {
+      "epoch": 80.0,
+      "grad_norm": 1.2421706914901733,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4328,
+      "step": 900
+    },
+    {
+      "epoch": 84.44444444444444,
+      "grad_norm": 0.8712289929389954,
+      "learning_rate": 4.75e-05,
+      "loss": 0.4335,
+      "step": 950
+    },
+    {
+      "epoch": 88.88888888888889,
+      "grad_norm": 1.161516547203064,
+      "learning_rate": 5e-05,
+      "loss": 0.4317,
+      "step": 1000
+    },
+    {
+      "epoch": 88.88888888888889,
+      "eval_loss": 0.40811634063720703,
+      "eval_runtime": 9.5158,
+      "eval_samples_per_second": 16.709,
+      "eval_steps_per_second": 2.102,
+      "step": 1000
+    },
+    {
+      "epoch": 93.33333333333333,
+      "grad_norm": 4.811057090759277,
+      "learning_rate": 5.25e-05,
+      "loss": 0.4311,
+      "step": 1050
+    },
+    {
+      "epoch": 97.77777777777777,
+      "grad_norm": 1.5318878889083862,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 0.4253,
+      "step": 1100
+    },
+    {
+      "epoch": 102.22222222222223,
+      "grad_norm": 1.7489895820617676,
+      "learning_rate": 5.7499999999999995e-05,
+      "loss": 0.4229,
+      "step": 1150
+    },
+    {
+      "epoch": 106.66666666666667,
+      "grad_norm": 1.162777304649353,
+      "learning_rate": 6e-05,
+      "loss": 0.4214,
+      "step": 1200
+    },
+    {
+      "epoch": 111.11111111111111,
+      "grad_norm": 1.2216763496398926,
+      "learning_rate": 6.25e-05,
+      "loss": 0.4168,
+      "step": 1250
+    },
+    {
+      "epoch": 115.55555555555556,
+      "grad_norm": 1.368053674697876,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 0.4197,
+      "step": 1300
+    },
+    {
+      "epoch": 120.0,
+      "grad_norm": 1.3240705728530884,
+      "learning_rate": 6.750000000000001e-05,
+      "loss": 0.4143,
+      "step": 1350
+    },
+    {
+      "epoch": 124.44444444444444,
+      "grad_norm": 3.4118428230285645,
+      "learning_rate": 7e-05,
+      "loss": 0.4159,
+      "step": 1400
+    },
+    {
+      "epoch": 128.88888888888889,
+      "grad_norm": 2.0438427925109863,
+      "learning_rate": 7.25e-05,
+      "loss": 0.4135,
+      "step": 1450
+    },
+    {
+      "epoch": 133.33333333333334,
+      "grad_norm": 3.8993186950683594,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.412,
+      "step": 1500
+    },
+    {
+      "epoch": 133.33333333333334,
+      "eval_loss": 0.40509456396102905,
+      "eval_runtime": 8.6764,
+      "eval_samples_per_second": 18.326,
+      "eval_steps_per_second": 2.305,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 4000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 364,
+  "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": 4.122316318184064e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe21f66ae29351cf4a52c96ecd128baf630207d21c90cee80fdb06c7fcd7437
+size 5304