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:e953d290bf6fa77196bebdccebcf1449321dfb1e0785c52fad08116743eaa121
+size 577789320

checkpoint-2500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac921c6e5b5baf94cb84864ccefe951eb782be9d911539577b7f87581a2f2238
+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 sha256:17d91fb6e687c9e485a8b0f5b25e3d3502d313b85fb6cd013a370aa3ac81e5d9
+size 14244

checkpoint-2500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01c8f493f88d64182a2dfcee190fa9ab01bd14fa7f93a04b41d568c47a164cf9
+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": 0.35975897312164307,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b64_le4_s12000/checkpoint-2500",
+  "epoch": 2.094869109947644,
+  "eval_steps": 500,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.041884816753926704,
+      "grad_norm": 2.697540283203125,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 1.0526,
+      "step": 50
+    },
+    {
+      "epoch": 0.08376963350785341,
+      "grad_norm": 4.0351996421813965,
+      "learning_rate": 4.9000000000000005e-06,
+      "loss": 0.8484,
+      "step": 100
+    },
+    {
+      "epoch": 0.1256544502617801,
+      "grad_norm": 4.858840465545654,
+      "learning_rate": 7.4e-06,
+      "loss": 0.7505,
+      "step": 150
+    },
+    {
+      "epoch": 0.16753926701570682,
+      "grad_norm": 2.396157741546631,
+      "learning_rate": 9.900000000000002e-06,
+      "loss": 0.6889,
+      "step": 200
+    },
+    {
+      "epoch": 0.2094240837696335,
+      "grad_norm": 2.660891056060791,
+      "learning_rate": 1.24e-05,
+      "loss": 0.6363,
+      "step": 250
+    },
+    {
+      "epoch": 0.2513089005235602,
+      "grad_norm": 2.848365068435669,
+      "learning_rate": 1.49e-05,
+      "loss": 0.6107,
+      "step": 300
+    },
+    {
+      "epoch": 0.2931937172774869,
+      "grad_norm": 2.0031681060791016,
+      "learning_rate": 1.74e-05,
+      "loss": 0.5673,
+      "step": 350
+    },
+    {
+      "epoch": 0.33507853403141363,
+      "grad_norm": 4.771584510803223,
+      "learning_rate": 1.9900000000000003e-05,
+      "loss": 0.5566,
+      "step": 400
+    },
+    {
+      "epoch": 0.3769633507853403,
+      "grad_norm": 9.876974105834961,
+      "learning_rate": 2.2400000000000002e-05,
+      "loss": 0.5487,
+      "step": 450
+    },
+    {
+      "epoch": 0.418848167539267,
+      "grad_norm": 2.113546848297119,
+      "learning_rate": 2.4900000000000002e-05,
+      "loss": 0.5268,
+      "step": 500
+    },
+    {
+      "epoch": 0.418848167539267,
+      "eval_loss": 0.46557939052581787,
+      "eval_runtime": 297.9193,
+      "eval_samples_per_second": 28.494,
+      "eval_steps_per_second": 3.565,
+      "step": 500
+    },
+    {
+      "epoch": 0.4607329842931937,
+      "grad_norm": 2.7003655433654785,
+      "learning_rate": 2.7400000000000002e-05,
+      "loss": 0.5212,
+      "step": 550
+    },
+    {
+      "epoch": 0.5026178010471204,
+      "grad_norm": 5.649191379547119,
+      "learning_rate": 2.9900000000000002e-05,
+      "loss": 0.5138,
+      "step": 600
+    },
+    {
+      "epoch": 0.5445026178010471,
+      "grad_norm": 2.272181749343872,
+      "learning_rate": 3.24e-05,
+      "loss": 0.5121,
+      "step": 650
+    },
+    {
+      "epoch": 0.5863874345549738,
+      "grad_norm": 2.957671642303467,
+      "learning_rate": 3.49e-05,
+      "loss": 0.496,
+      "step": 700
+    },
+    {
+      "epoch": 0.6282722513089005,
+      "grad_norm": 3.577693462371826,
+      "learning_rate": 3.74e-05,
+      "loss": 0.4886,
+      "step": 750
+    },
+    {
+      "epoch": 0.6701570680628273,
+      "grad_norm": 2.466113328933716,
+      "learning_rate": 3.99e-05,
+      "loss": 0.4879,
+      "step": 800
+    },
+    {
+      "epoch": 0.7120418848167539,
+      "grad_norm": 2.941208600997925,
+      "learning_rate": 4.24e-05,
+      "loss": 0.4865,
+      "step": 850
+    },
+    {
+      "epoch": 0.7539267015706806,
+      "grad_norm": 2.3843142986297607,
+      "learning_rate": 4.49e-05,
+      "loss": 0.4689,
+      "step": 900
+    },
+    {
+      "epoch": 0.7958115183246073,
+      "grad_norm": 2.232208728790283,
+      "learning_rate": 4.74e-05,
+      "loss": 0.4768,
+      "step": 950
+    },
+    {
+      "epoch": 0.837696335078534,
+      "grad_norm": 2.1702969074249268,
+      "learning_rate": 4.99e-05,
+      "loss": 0.4583,
+      "step": 1000
+    },
+    {
+      "epoch": 0.837696335078534,
+      "eval_loss": 0.4044812321662903,
+      "eval_runtime": 296.3081,
+      "eval_samples_per_second": 28.649,
+      "eval_steps_per_second": 3.584,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8795811518324608,
+      "grad_norm": 2.7613942623138428,
+      "learning_rate": 5.2400000000000007e-05,
+      "loss": 0.4664,
+      "step": 1050
+    },
+    {
+      "epoch": 0.9214659685863874,
+      "grad_norm": 2.4752321243286133,
+      "learning_rate": 5.4900000000000006e-05,
+      "loss": 0.442,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9633507853403142,
+      "grad_norm": 2.27407169342041,
+      "learning_rate": 5.74e-05,
+      "loss": 0.4498,
+      "step": 1150
+    },
+    {
+      "epoch": 1.0052356020942408,
+      "grad_norm": 1.8998862504959106,
+      "learning_rate": 5.99e-05,
+      "loss": 0.4433,
+      "step": 1200
+    },
+    {
+      "epoch": 1.0471204188481675,
+      "grad_norm": 2.8479738235473633,
+      "learning_rate": 6.24e-05,
+      "loss": 0.4417,
+      "step": 1250
+    },
+    {
+      "epoch": 1.0890052356020943,
+      "grad_norm": 1.9095953702926636,
+      "learning_rate": 6.49e-05,
+      "loss": 0.444,
+      "step": 1300
+    },
+    {
+      "epoch": 1.130890052356021,
+      "grad_norm": 1.368077278137207,
+      "learning_rate": 6.740000000000001e-05,
+      "loss": 0.4519,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1727748691099475,
+      "grad_norm": 2.0456721782684326,
+      "learning_rate": 6.99e-05,
+      "loss": 0.4411,
+      "step": 1400
+    },
+    {
+      "epoch": 1.2146596858638743,
+      "grad_norm": 1.6217619180679321,
+      "learning_rate": 7.24e-05,
+      "loss": 0.4324,
+      "step": 1450
+    },
+    {
+      "epoch": 1.256544502617801,
+      "grad_norm": 2.961646795272827,
+      "learning_rate": 7.49e-05,
+      "loss": 0.4321,
+      "step": 1500
+    },
+    {
+      "epoch": 1.256544502617801,
+      "eval_loss": 0.3930450975894928,
+      "eval_runtime": 301.1428,
+      "eval_samples_per_second": 28.189,
+      "eval_steps_per_second": 3.527,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2984293193717278,
+      "grad_norm": 2.543201446533203,
+      "learning_rate": 7.740000000000001e-05,
+      "loss": 0.4364,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3403141361256545,
+      "grad_norm": 1.5807560682296753,
+      "learning_rate": 7.99e-05,
+      "loss": 0.433,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3821989528795813,
+      "grad_norm": 1.587385892868042,
+      "learning_rate": 8.24e-05,
+      "loss": 0.4285,
+      "step": 1650
+    },
+    {
+      "epoch": 1.4240837696335078,
+      "grad_norm": 2.1177687644958496,
+      "learning_rate": 8.49e-05,
+      "loss": 0.4209,
+      "step": 1700
+    },
+    {
+      "epoch": 1.4659685863874345,
+      "grad_norm": 1.6078702211380005,
+      "learning_rate": 8.740000000000001e-05,
+      "loss": 0.426,
+      "step": 1750
+    },
+    {
+      "epoch": 1.5078534031413613,
+      "grad_norm": 2.10492205619812,
+      "learning_rate": 8.99e-05,
+      "loss": 0.4191,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5497382198952878,
+      "grad_norm": 2.169309616088867,
+      "learning_rate": 9.240000000000001e-05,
+      "loss": 0.4178,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5916230366492146,
+      "grad_norm": 1.559006929397583,
+      "learning_rate": 9.49e-05,
+      "loss": 0.4133,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6335078534031413,
+      "grad_norm": 2.0436346530914307,
+      "learning_rate": 9.74e-05,
+      "loss": 0.4234,
+      "step": 1950
+    },
+    {
+      "epoch": 1.675392670157068,
+      "grad_norm": 2.089486837387085,
+      "learning_rate": 9.99e-05,
+      "loss": 0.4092,
+      "step": 2000
+    },
+    {
+      "epoch": 1.675392670157068,
+      "eval_loss": 0.37686464190483093,
+      "eval_runtime": 310.3113,
+      "eval_samples_per_second": 27.356,
+      "eval_steps_per_second": 3.422,
+      "step": 2000
+    },
+    {
+      "epoch": 1.7179057591623037,
+      "grad_norm": 2.684563398361206,
+      "learning_rate": 9.952e-05,
+      "loss": 0.4099,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7597905759162304,
+      "grad_norm": 1.7815254926681519,
+      "learning_rate": 9.902e-05,
+      "loss": 0.4093,
+      "step": 2100
+    },
+    {
+      "epoch": 1.8016753926701572,
+      "grad_norm": 1.9900892972946167,
+      "learning_rate": 9.852e-05,
+      "loss": 0.4116,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8435602094240837,
+      "grad_norm": 1.8637460470199585,
+      "learning_rate": 9.802e-05,
+      "loss": 0.4089,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8854450261780105,
+      "grad_norm": 1.58171546459198,
+      "learning_rate": 9.752e-05,
+      "loss": 0.4053,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9273298429319372,
+      "grad_norm": 1.4995630979537964,
+      "learning_rate": 9.702e-05,
+      "loss": 0.4049,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9692146596858637,
+      "grad_norm": 2.447049379348755,
+      "learning_rate": 9.652e-05,
+      "loss": 0.4029,
+      "step": 2350
+    },
+    {
+      "epoch": 2.0110994764397905,
+      "grad_norm": 1.8776074647903442,
+      "learning_rate": 9.602e-05,
+      "loss": 0.3988,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0529842931937172,
+      "grad_norm": 3.1785056591033936,
+      "learning_rate": 9.552000000000001e-05,
+      "loss": 0.4023,
+      "step": 2450
+    },
+    {
+      "epoch": 2.094869109947644,
+      "grad_norm": 1.4416707754135132,
+      "learning_rate": 9.502000000000001e-05,
+      "loss": 0.4008,
+      "step": 2500
+    },
+    {
+      "epoch": 2.094869109947644,
+      "eval_loss": 0.35975897312164307,
+      "eval_runtime": 275.2855,
+      "eval_samples_per_second": 30.837,
+      "eval_steps_per_second": 3.858,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 12000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 11,
+  "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": 2.245101969215635e+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:fa8f41edfc5db87ff9f21c8561c4e1b9e9e17c2b5ee34213317807a5e0e41371
+size 5304