Training in progress, step 2000, checkpoint

checkpoint-2000/added_tokens.json

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

checkpoint-2000/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-2000/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-2000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52912110366f353648456d94db4eb3eededa1a7bd1e1ee948a08978fa689fc86
+size 577789320

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc09339c0743ae93f2c6d529ea7fd41ef8b4d7558d5651b00ff3f3def1959bb5
+size 1155772233

checkpoint-2000/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-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:024d0cf30a7dc2cd699fa88d8139043fbfd6dbc8be6ed6931372b1fb1049dcda
+size 14244

checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b0b2a4fb4f9beb82c2cdc2dd6c09abca2bfc108bd9784a5fcb4addd2fba2f91
+size 1064

checkpoint-2000/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-2000/spm_char.model

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

checkpoint-2000/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-2000/trainer_state.json

@@ -0,0 +1,345 @@
+{
+  "best_metric": 0.4031977355480194,
+  "best_model_checkpoint": "mikhail-panzo/ceb_b64_le5_s4000/checkpoint-2000",
+  "epoch": 79.20792079207921,
+  "eval_steps": 500,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 2.883972644805908,
+      "learning_rate": 2.5000000000000004e-07,
+      "loss": 0.8004,
+      "step": 50
+    },
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 2.5420947074890137,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 0.77,
+      "step": 100
+    },
+    {
+      "epoch": 5.9405940594059405,
+      "grad_norm": 1.3930779695510864,
+      "learning_rate": 7.5e-07,
+      "loss": 0.7525,
+      "step": 150
+    },
+    {
+      "epoch": 7.920792079207921,
+      "grad_norm": 1.8414520025253296,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.722,
+      "step": 200
+    },
+    {
+      "epoch": 9.900990099009901,
+      "grad_norm": 2.0665407180786133,
+      "learning_rate": 1.25e-06,
+      "loss": 0.7076,
+      "step": 250
+    },
+    {
+      "epoch": 11.881188118811881,
+      "grad_norm": 1.2595587968826294,
+      "learning_rate": 1.5e-06,
+      "loss": 0.6757,
+      "step": 300
+    },
+    {
+      "epoch": 13.861386138613861,
+      "grad_norm": 1.4587587118148804,
+      "learning_rate": 1.75e-06,
+      "loss": 0.6562,
+      "step": 350
+    },
+    {
+      "epoch": 15.841584158415841,
+      "grad_norm": 2.241546154022217,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.6016,
+      "step": 400
+    },
+    {
+      "epoch": 17.821782178217823,
+      "grad_norm": 1.4243369102478027,
+      "learning_rate": 2.25e-06,
+      "loss": 0.5593,
+      "step": 450
+    },
+    {
+      "epoch": 19.801980198019802,
+      "grad_norm": 1.6889928579330444,
+      "learning_rate": 2.5e-06,
+      "loss": 0.536,
+      "step": 500
+    },
+    {
+      "epoch": 19.801980198019802,
+      "eval_loss": 0.4640245735645294,
+      "eval_runtime": 6.7454,
+      "eval_samples_per_second": 26.685,
+      "eval_steps_per_second": 3.41,
+      "step": 500
+    },
+    {
+      "epoch": 21.782178217821784,
+      "grad_norm": 1.1601781845092773,
+      "learning_rate": 2.7500000000000004e-06,
+      "loss": 0.5198,
+      "step": 550
+    },
+    {
+      "epoch": 23.762376237623762,
+      "grad_norm": 1.1969332695007324,
+      "learning_rate": 3e-06,
+      "loss": 0.5114,
+      "step": 600
+    },
+    {
+      "epoch": 25.742574257425744,
+      "grad_norm": 1.1022064685821533,
+      "learning_rate": 3.2500000000000002e-06,
+      "loss": 0.5092,
+      "step": 650
+    },
+    {
+      "epoch": 27.722772277227723,
+      "grad_norm": 1.2418344020843506,
+      "learning_rate": 3.5e-06,
+      "loss": 0.4951,
+      "step": 700
+    },
+    {
+      "epoch": 29.702970297029704,
+      "grad_norm": 1.1083300113677979,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.4911,
+      "step": 750
+    },
+    {
+      "epoch": 31.683168316831683,
+      "grad_norm": 1.2230323553085327,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.4866,
+      "step": 800
+    },
+    {
+      "epoch": 33.663366336633665,
+      "grad_norm": 1.1232423782348633,
+      "learning_rate": 4.25e-06,
+      "loss": 0.4876,
+      "step": 850
+    },
+    {
+      "epoch": 35.64356435643565,
+      "grad_norm": 0.9751402139663696,
+      "learning_rate": 4.5e-06,
+      "loss": 0.4772,
+      "step": 900
+    },
+    {
+      "epoch": 37.62376237623762,
+      "grad_norm": 1.0373197793960571,
+      "learning_rate": 4.75e-06,
+      "loss": 0.4791,
+      "step": 950
+    },
+    {
+      "epoch": 39.603960396039604,
+      "grad_norm": 1.0164889097213745,
+      "learning_rate": 5e-06,
+      "loss": 0.4674,
+      "step": 1000
+    },
+    {
+      "epoch": 39.603960396039604,
+      "eval_loss": 0.4247289001941681,
+      "eval_runtime": 6.5527,
+      "eval_samples_per_second": 27.47,
+      "eval_steps_per_second": 3.51,
+      "step": 1000
+    },
+    {
+      "epoch": 41.584158415841586,
+      "grad_norm": 1.087668538093567,
+      "learning_rate": 5.2500000000000006e-06,
+      "loss": 0.4754,
+      "step": 1050
+    },
+    {
+      "epoch": 43.56435643564357,
+      "grad_norm": 0.891859769821167,
+      "learning_rate": 5.500000000000001e-06,
+      "loss": 0.4677,
+      "step": 1100
+    },
+    {
+      "epoch": 45.54455445544554,
+      "grad_norm": 1.018160343170166,
+      "learning_rate": 5.75e-06,
+      "loss": 0.4642,
+      "step": 1150
+    },
+    {
+      "epoch": 47.524752475247524,
+      "grad_norm": 1.4765816926956177,
+      "learning_rate": 6e-06,
+      "loss": 0.4602,
+      "step": 1200
+    },
+    {
+      "epoch": 49.504950495049506,
+      "grad_norm": 1.3534387350082397,
+      "learning_rate": 6.25e-06,
+      "loss": 0.4634,
+      "step": 1250
+    },
+    {
+      "epoch": 51.48514851485149,
+      "grad_norm": 1.031537413597107,
+      "learning_rate": 6.5000000000000004e-06,
+      "loss": 0.4567,
+      "step": 1300
+    },
+    {
+      "epoch": 53.46534653465346,
+      "grad_norm": 1.0172531604766846,
+      "learning_rate": 6.750000000000001e-06,
+      "loss": 0.4592,
+      "step": 1350
+    },
+    {
+      "epoch": 55.445544554455445,
+      "grad_norm": 0.8680336475372314,
+      "learning_rate": 7e-06,
+      "loss": 0.4583,
+      "step": 1400
+    },
+    {
+      "epoch": 57.42574257425743,
+      "grad_norm": 1.2397912740707397,
+      "learning_rate": 7.25e-06,
+      "loss": 0.4459,
+      "step": 1450
+    },
+    {
+      "epoch": 59.40594059405941,
+      "grad_norm": 1.021185040473938,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.4479,
+      "step": 1500
+    },
+    {
+      "epoch": 59.40594059405941,
+      "eval_loss": 0.4102235734462738,
+      "eval_runtime": 7.6974,
+      "eval_samples_per_second": 23.384,
+      "eval_steps_per_second": 2.988,
+      "step": 1500
+    },
+    {
+      "epoch": 61.386138613861384,
+      "grad_norm": 1.005591630935669,
+      "learning_rate": 7.75e-06,
+      "loss": 0.4469,
+      "step": 1550
+    },
+    {
+      "epoch": 63.366336633663366,
+      "grad_norm": 0.8193312287330627,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.4425,
+      "step": 1600
+    },
+    {
+      "epoch": 65.34653465346534,
+      "grad_norm": 0.9143276810646057,
+      "learning_rate": 8.25e-06,
+      "loss": 0.4424,
+      "step": 1650
+    },
+    {
+      "epoch": 67.32673267326733,
+      "grad_norm": 1.0127445459365845,
+      "learning_rate": 8.5e-06,
+      "loss": 0.4446,
+      "step": 1700
+    },
+    {
+      "epoch": 69.3069306930693,
+      "grad_norm": 1.045551061630249,
+      "learning_rate": 8.750000000000001e-06,
+      "loss": 0.4417,
+      "step": 1750
+    },
+    {
+      "epoch": 71.2871287128713,
+      "grad_norm": 0.9629473090171814,
+      "learning_rate": 9e-06,
+      "loss": 0.4362,
+      "step": 1800
+    },
+    {
+      "epoch": 73.26732673267327,
+      "grad_norm": 0.9655708074569702,
+      "learning_rate": 9.250000000000001e-06,
+      "loss": 0.4393,
+      "step": 1850
+    },
+    {
+      "epoch": 75.24752475247524,
+      "grad_norm": 1.167992115020752,
+      "learning_rate": 9.5e-06,
+      "loss": 0.4373,
+      "step": 1900
+    },
+    {
+      "epoch": 77.22772277227723,
+      "grad_norm": 0.9866817593574524,
+      "learning_rate": 9.75e-06,
+      "loss": 0.4369,
+      "step": 1950
+    },
+    {
+      "epoch": 79.20792079207921,
+      "grad_norm": 0.8923723101615906,
+      "learning_rate": 1e-05,
+      "loss": 0.4356,
+      "step": 2000
+    },
+    {
+      "epoch": 79.20792079207921,
+      "eval_loss": 0.4031977355480194,
+      "eval_runtime": 7.0669,
+      "eval_samples_per_second": 25.471,
+      "eval_steps_per_second": 3.255,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 4000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 160,
+  "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": 2.161155031107257e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:893b77dc1258076eac6823109e549347547b5ddcab0a290335f22ad5e88711bb
+size 5304