trained model 1

.gitattributes

@@ -25,3 +25,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+language_model/5gram.bin filter=lfs diff=lfs merge=lfs -text
+language_model/unigrams.txt filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,98 @@
+---
+language:
+- uk
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- mozilla-foundation/common_voice_8_0
+- generated_from_trainer
+- robust-speech-event
+datasets:
+- common_voice
+model-index:
+- name: wav2vec2-xls-r-1b-hy-cv
+  results:
+  - task: 
+      type: automatic-speech-recognition
+      name: Speech Recognition
+    dataset:
+      type: mozilla-foundation/common_voice_8_0
+      name: Common Voice uk
+      args: uk
+    metrics:
+      - type: wer
+        value: 12.246920571994903
+        name: WER LM
+      - type: cer
+        value: 2.513653497966816
+        name: CER LM
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# 
+
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - UK dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1747
+- Wer: 0.2107
+- Cer: 0.0408
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 8e-05
+- train_batch_size: 16
+- eval_batch_size: 64
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
+- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- training_steps: 8000
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
+| 1.3719        | 4.35  | 500  | 0.3389          | 0.4236 | 0.0833 |
+| 1.1361        | 8.7   | 1000 | 0.2309          | 0.3162 | 0.0630 |
+| 1.0517        | 13.04 | 1500 | 0.2166          | 0.3056 | 0.0597 |
+| 1.0118        | 17.39 | 2000 | 0.2141          | 0.2784 | 0.0557 |
+| 0.9922        | 21.74 | 2500 | 0.2231          | 0.2941 | 0.0594 |
+| 0.9929        | 26.09 | 3000 | 0.2171          | 0.2892 | 0.0587 |
+| 0.9485        | 30.43 | 3500 | 0.2236          | 0.2956 | 0.0599 |
+| 0.9573        | 34.78 | 4000 | 0.2314          | 0.3043 | 0.0616 |
+| 0.9195        | 39.13 | 4500 | 0.2169          | 0.2812 | 0.0580 |
+| 0.8915        | 43.48 | 5000 | 0.2109          | 0.2780 | 0.0560 |
+| 0.8449        | 47.83 | 5500 | 0.2050          | 0.2534 | 0.0514 |
+| 0.8028        | 52.17 | 6000 | 0.2032          | 0.2456 | 0.0492 |
+| 0.7881        | 56.52 | 6500 | 0.1890          | 0.2380 | 0.0469 |
+| 0.7423        | 60.87 | 7000 | 0.1816          | 0.2245 | 0.0442 |
+| 0.7248        | 65.22 | 7500 | 0.1789          | 0.2165 | 0.0422 |
+| 0.6993        | 69.57 | 8000 | 0.1747          | 0.2107 | 0.0408 |
+
+
+### Framework versions
+
+- Transformers 4.17.0.dev0
+- Pytorch 1.10.2+cu102
+- Datasets 1.18.2.dev0
+- Tokenizers 0.11.0

added_tokens.json

@@ -0,0 +1 @@
+{"<s>": 37, "</s>": 38}

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 69.57,
+    "eval_cer": 0.040792224817745956,
+    "eval_loss": 0.1746993511915207,
+    "eval_runtime": 197.4273,
+    "eval_samples": 5802,
+    "eval_samples_per_second": 29.388,
+    "eval_steps_per_second": 0.461,
+    "eval_wer": 0.21072254471659824,
+    "train_loss": 1.039798891067505,
+    "train_runtime": 49145.562,
+    "train_samples": 14717,
+    "train_samples_per_second": 20.836,
+    "train_steps_per_second": 0.163
+}

alphabet.json

@@ -0,0 +1 @@
+{"labels": [" ", "\u0430", "\u0431", "\u0432", "\u0433", "\u0434", "\u0435", "\u0436", "\u0437", "\u0438", "\u0439", "\u043a", "\u043b", "\u043c", "\u043d", "\u043e", "\u043f", "\u0440", "\u0441", "\u0442", "\u0443", "\u0444", "\u0445", "\u0446", "\u0447", "\u0448", "\u0449", "\u044c", "\u044e", "\u044f", "\u0454", "\u0456", "\u0457", "\u0491", "\u2019", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-1b",
+  "activation_dropout": 0.1,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 1024,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "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
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 1280,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_feature_length": 64,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.25,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.55,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 48,
+  "num_negatives": 100,
+  "output_hidden_size": 1280,
+  "pad_token_id": 36,
+  "proj_codevector_dim": 1024,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 39,
+  "xvector_output_dim": 512
+}

eval.py

@@ -0,0 +1,136 @@
+#!/usr/bin/env python3
+import argparse
+import re
+from typing import Dict
+
+import torch
+from datasets import Audio, Dataset, load_dataset, load_metric
+
+from transformers import AutoFeatureExtractor, pipeline, Wav2Vec2ProcessorWithLM
+
+
+def log_results(result: Dataset, args: Dict[str, str]):
+    """DO NOT CHANGE. This function computes and logs the result metrics."""
+
+    log_outputs = args.log_outputs
+    dataset_id = "_".join(args.dataset.split("/") + [args.config, args.split])
+
+    # load metric
+    wer = load_metric("wer")
+    cer = load_metric("cer")
+
+    # compute metrics
+    wer_result = wer.compute(references=result["target"], predictions=result["prediction"])
+    cer_result = cer.compute(references=result["target"], predictions=result["prediction"])
+
+    # print & log results
+    result_str = f"WER: {wer_result}\n" f"CER: {cer_result}"
+    print(result_str)
+
+    with open(f"{dataset_id}_eval_results.txt", "w") as f:
+        f.write(result_str)
+
+    # log all results in text file. Possibly interesting for analysis
+    if log_outputs is not None:
+        pred_file = f"log_{dataset_id}_predictions.txt"
+        target_file = f"log_{dataset_id}_targets.txt"
+
+        with open(pred_file, "w") as p, open(target_file, "w") as t:
+
+            # mapping function to write output
+            def write_to_file(batch, i):
+                p.write(f"{i}" + "\n")
+                p.write(batch["prediction"] + "\n")
+                t.write(f"{i}" + "\n")
+                t.write(batch["target"] + "\n")
+
+            result.map(write_to_file, with_indices=True)
+
+
+def normalize_text(text: str) -> str:
+    """This function normalizes the target text."""
+    text = re.sub("['՝՛՚‘`]", "’", text)
+    chars_to_remove_regex = re.compile('[\,\?\.\!\-\;\:\"\“\%\”\�«»\(\)–—…ý]')
+    text = re.sub(chars_to_remove_regex, '', text.lower())+" "
+    for a, b in zip("aceioxklmnpujы", "асеіохклмнпужи"):
+        text = re.sub(a, b, text)
+    text = " ".join(text.split())
+
+    return text
+
+
+def main(args):
+    # load dataset
+    dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
+
+    # for testing: only process the first two examples as a test
+    # dataset = dataset.select(range(10))
+
+    # load processor
+    # feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
+    # sampling_rate = feature_extractor.sampling_rate
+    processor = Wav2Vec2ProcessorWithLM.from_pretrained(args.model_id)
+
+    # resample audio
+    dataset = dataset.cast_column("audio", Audio(sampling_rate=processor.feature_extractor.sampling_rate))
+
+    # load eval pipeline
+    if args.device is None:
+        args.device = 0 if torch.cuda.is_available() else -1
+    asr = pipeline(
+        "automatic-speech-recognition", model=args.model_id, device=args.device, 
+        feature_extractor=processor.feature_extractor, decoder=processor.decoder
+    )
+
+    # map function to decode audio
+    def map_to_pred(batch):
+        prediction = asr(
+            batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
+        )
+
+        batch["prediction"] = prediction["text"]
+        batch["target"] = normalize_text(batch["sentence"])
+        return batch
+
+    # run inference on all examples
+    result = dataset.map(map_to_pred, remove_columns=dataset.column_names)
+
+    # compute and log_results
+    # do not change function below
+    log_results(result, args)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument(
+        "--model_id", type=str, required=True, help="Model identifier. Should be loadable with 🤗 Transformers"
+    )
+    parser.add_argument(
+        "--dataset",
+        type=str,
+        required=True,
+        help="Dataset name to evaluate the `model_id`. Should be loadable with 🤗 Datasets",
+    )
+    parser.add_argument(
+        "--config", type=str, required=True, help="Config of the dataset. *E.g.* `'en'`  for Common Voice"
+    )
+    parser.add_argument("--split", type=str, required=True, help="Split of the dataset. *E.g.* `'test'`")
+    parser.add_argument(
+        "--chunk_length_s", type=float, default=None, help="Chunk length in seconds. Defaults to 5 seconds."
+    )
+    parser.add_argument(
+        "--stride_length_s", type=float, default=None, help="Stride of the audio chunks. Defaults to 1 second."
+    )
+    parser.add_argument(
+        "--log_outputs", action="store_true", help="If defined, write outputs to log file for analysis."
+    )
+    parser.add_argument(
+        "--device",
+        type=int,
+        default=None,
+        help="The device to run the pipeline on. -1 for CPU (default), 0 for the first GPU and so on.",
+    )
+    args = parser.parse_args()
+
+    main(args)

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 69.57,
+    "eval_cer": 0.040792224817745956,
+    "eval_loss": 0.1746993511915207,
+    "eval_runtime": 197.4273,
+    "eval_samples": 5802,
+    "eval_samples_per_second": 29.388,
+    "eval_steps_per_second": 0.461,
+    "eval_wer": 0.21072254471659824
+}

language_model/5gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:019cbcf92023babd3238ae8c5b0f93b82a4d727ef602571b6c7e4a96a5eb812c
+size 1848640830

language_model/attrs.json

@@ -0,0 +1 @@
+{"alpha": 0.5, "beta": 1.5, "unk_score_offset": -10.0, "score_boundary": true}

language_model/unigrams.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:899fb26be366407f0b198087853d5945df7f627680c3aebbd54a7b182e481e01
+size 18515524

mozilla-foundation_common_voice_8_0_uk_test_eval_results.txt

@@ -0,0 +1,2 @@
+WER: 0.12246920571994903
+CER: 0.02513653497966816

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "Wav2Vec2ProcessorWithLM",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:365416933b95258481e889019db831b92860ad1820c509b05f6d9f38ec68ab92
+size 3850512561

run.sh

@@ -0,0 +1,40 @@
+python run_speech_recognition_ctc.py \
+	--dataset_name="mozilla-foundation/common_voice_8_0" \
+	--dataset_config_name="uk" \
+	--model_name_or_path="facebook/wav2vec2-xls-r-1b" \
+	--output_dir="./" \
+	--overwrite_output_dir \
+	--max_steps 8000 \
+	--per_device_train_batch_size="16" \
+	--per_device_eval_batch_size="64" \
+	--gradient_accumulation_steps="8" \
+	--dataloader_num_workers 8 \
+	--learning_rate="8e-5" \
+	--adam_beta2 0.98 \
+	--lr_scheduler_type cosine \
+	--warmup_ratio 0.1 \
+	--evaluation_strategy="steps" \
+	--text_column_name="sentence" \
+	--chars_to_ignore \, \? \. \! \- \; \: \" \“ \% \‘ \” \� \' « » \( \) ՝ ՛ ՚ \– \— \… ý \
+	--save_steps="500" \
+	--eval_steps="500" \
+	--logging_steps="100" \
+	--save_total_limit 5 \
+	--freeze_feature_encoder \
+	--layerdrop="0.1" \
+	--activation_dropout="0.1" \
+	--feat_proj_dropout="0.0" \
+	--mask_time_prob="0.55" \
+	--mask_time_length="10" \
+	--mask_feature_prob="0.25" \
+	--mask_feature_length="64" \
+	--gradient_checkpointing \
+	--use_auth_token \
+	--fp16 \
+	--group_by_length \
+	--do_train --do_eval \
+	--load_best_model_at_end \
+	--report_to all \
+	--run_name="xlsr-uk-cv-1b-1" \
+	--wandb_project="xlsr-uk" \
+	--bnb --tristage_sched

run_speech_recognition_ctc.py

@@ -0,0 +1,819 @@
+#!/usr/bin/env python
+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+
+""" Fine-tuning a 🤗 Transformers CTC model for automatic speech recognition"""
+
+import functools
+import json
+import logging
+import os
+import re
+import sys
+import warnings
+from dataclasses import dataclass, field
+from typing import Dict, List, Optional, Union
+
+import datasets
+import numpy as np
+import torch
+from torch.optim.lr_scheduler import LambdaLR
+from datasets import DatasetDict, load_dataset, load_metric
+
+import bitsandbytes as bnb
+import transformers
+from transformers import (
+    AutoConfig,
+    AutoFeatureExtractor,
+    AutoModelForCTC,
+    AutoProcessor,
+    AutoTokenizer,
+    HfArgumentParser,
+    Trainer,
+    TrainingArguments,
+    Wav2Vec2Processor,
+    set_seed,
+)
+from transformers.trainer_pt_utils import get_parameter_names
+from transformers.trainer_utils import get_last_checkpoint, is_main_process
+from transformers.utils import check_min_version
+from transformers.utils.versions import require_version
+
+
+# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
+check_min_version("4.16.0.dev0")
+
+require_version("datasets>=1.13.3", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt")
+
+
+logger = logging.getLogger(__name__)
+
+
+def list_field(default=None, metadata=None):
+    return field(default_factory=lambda: default, metadata=metadata)
+
+
+@dataclass
+class ModelArguments:
+    """
+    Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
+    """
+
+    model_name_or_path: str = field(
+        metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"}
+    )
+    tokenizer_name_or_path: Optional[str] = field(
+        default=None,
+        metadata={"help": "Path to pretrained tokenizer or tokenizer identifier from huggingface.co/models"},
+    )
+    cache_dir: Optional[str] = field(
+        default=None,
+        metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"},
+    )
+    freeze_feature_encoder: bool = field(
+        default=True, metadata={"help": "Whether to freeze the feature encoder layers of the model."}
+    )
+    attention_dropout: float = field(
+        default=0.0, metadata={"help": "The dropout ratio for the attention probabilities."}
+    )
+    activation_dropout: float = field(
+        default=0.0, metadata={"help": "The dropout ratio for activations inside the fully connected layer."}
+    )
+    feat_proj_dropout: float = field(default=0.0, metadata={"help": "The dropout ratio for the projected features."})
+    hidden_dropout: float = field(
+        default=0.0,
+        metadata={
+            "help": "The dropout probability for all fully connected layers in the embeddings, encoder, and pooler."
+        },
+    )
+    final_dropout: float = field(
+        default=0.0,
+        metadata={"help": "The dropout probability for the final projection layer."},
+    )
+    mask_time_prob: float = field(
+        default=0.05,
+        metadata={
+            "help": "Probability of each feature vector along the time axis to be chosen as the start of the vector"
+            "span to be masked. Approximately ``mask_time_prob * sequence_length // mask_time_length`` feature"
+            "vectors will be masked along the time axis."
+        },
+    )
+    mask_time_length: int = field(
+        default=10,
+        metadata={"help": "Length of vector span to mask along the time axis."},
+    )
+    mask_feature_prob: float = field(
+        default=0.0,
+        metadata={
+            "help": "Probability of each feature vector along the feature axis to be chosen as the start of the vector"
+            "span to be masked. Approximately ``mask_feature_prob * sequence_length // mask_feature_length`` feature bins will be masked along the time axis."
+        },
+    )
+    mask_feature_length: int = field(
+        default=10,
+        metadata={"help": "Length of vector span to mask along the feature axis."},
+    )
+    layerdrop: float = field(default=0.0, metadata={"help": "The LayerDrop probability."})
+    ctc_loss_reduction: Optional[str] = field(
+        default="mean", metadata={"help": "The way the ctc loss should be reduced. Should be one of 'mean' or 'sum'."}
+    )
+
+
+@dataclass
+class DataTrainingArguments:
+    """
+    Arguments pertaining to what data we are going to input our model for training and eval.
+
+    Using `HfArgumentParser` we can turn this class
+    into argparse arguments to be able to specify them on
+    the command line.
+    """
+
+    dataset_name: str = field(
+        metadata={"help": "The configuration name of the dataset to use (via the datasets library)."}
+    )
+    dataset_config_name: str = field(
+        default=None, metadata={"help": "The configuration name of the dataset to use (via the datasets library)."}
+    )
+    train_split_name: str = field(
+        default="train+validation",
+        metadata={
+            "help": "The name of the training data set split to use (via the datasets library). Defaults to 'train'"
+        },
+    )
+    eval_split_name: str = field(
+        default="test",
+        metadata={
+            "help": "The name of the training data set split to use (via the datasets library). Defaults to 'train'"
+        },
+    )
+    audio_column_name: str = field(
+        default="audio",
+        metadata={"help": "The name of the dataset column containing the audio data. Defaults to 'audio'"},
+    )
+    text_column_name: str = field(
+        default="text",
+        metadata={"help": "The name of the dataset column containing the text data. Defaults to 'text'"},
+    )
+    overwrite_cache: bool = field(
+        default=False, metadata={"help": "Overwrite the cached preprocessed datasets or not."}
+    )
+    preprocessing_num_workers: Optional[int] = field(
+        default=None,
+        metadata={"help": "The number of processes to use for the preprocessing."},
+    )
+    max_train_samples: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": "For debugging purposes or quicker training, truncate the number of training examples to this "
+            "value if set."
+        },
+    )
+    max_eval_samples: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": "For debugging purposes or quicker training, truncate the number of validation examples to this "
+            "value if set."
+        },
+    )
+    chars_to_ignore: Optional[List[str]] = list_field(
+        default=None,
+        metadata={"help": "A list of characters to remove from the transcripts."},
+    )
+    eval_metrics: List[str] = list_field(
+        default=["wer", "cer"],
+        metadata={"help": "A list of metrics the model should be evaluated on. E.g. `'wer cer'`"},
+    )
+    max_duration_in_seconds: float = field(
+        default=20.0,
+        metadata={
+            "help": "Filter audio files that are longer than `max_duration_in_seconds` seconds to 'max_duration_in_seconds`"
+        },
+    )
+    min_duration_in_seconds: float = field(
+        default=0.0, metadata={"help": "Filter audio files that are shorter than `min_duration_in_seconds` seconds"}
+    )
+    preprocessing_only: bool = field(
+        default=False,
+        metadata={
+            "help": "Whether to only do data preprocessing and skip training. "
+            "This is especially useful when data preprocessing errors out in distributed training due to timeout. "
+            "In this case, one should run the preprocessing in a non-distributed setup with `preprocessing_only=True` "
+            "so that the cached datasets can consequently be loaded in distributed training"
+        },
+    )
+    use_auth_token: bool = field(
+        default=False,
+        metadata={
+            "help": "If :obj:`True`, will use the token generated when running"
+            ":obj:`transformers-cli login` as HTTP bearer authorization for remote files."
+        },
+    )
+    unk_token: str = field(
+        default="[UNK]",
+        metadata={"help": "The unk token for the tokenizer"},
+    )
+    pad_token: str = field(
+        default="[PAD]",
+        metadata={"help": "The padding token for the tokenizer"},
+    )
+    word_delimiter_token: str = field(
+        default="|",
+        metadata={"help": "The word delimiter token for the tokenizer"},
+    )
+    phoneme_language: Optional[str] = field(
+        default=None,
+        metadata={
+            "help": "The target language that should be used be"
+            " passed to the tokenizer for tokenization. Note that"
+            " this is only relevant if the model classifies the"
+            " input audio to a sequence of phoneme sequences."
+        },
+    )
+
+@dataclass
+class ExtraArguments:
+    "Additional training arguments"
+    bnb: bool = field(
+        default=False, 
+        metadata = {"help":"If true uses 8bit Adam"}
+    )
+    tristage_sched: bool = field(
+        default=False, 
+        metadata = {"help":"If true uses tristage LR scheduler (refer to XLS-R paper)"}
+    )
+    wandb_project: str = field(
+        default=None,
+        metadata = {"help":"Name of wandb project to log into"}
+    )
+
+
+@dataclass
+class DataCollatorCTCWithPadding:
+    """
+    Data collator that will dynamically pad the inputs received.
+    Args:
+        processor (:class:`~transformers.AutoProcessor`)
+            The processor used for proccessing the data.
+        padding (:obj:`bool`, :obj:`str` or :class:`~transformers.tokenization_utils_base.PaddingStrategy`, `optional`, defaults to :obj:`True`):
+            Select a strategy to pad the returned sequences (according to the model's padding side and padding index)
+            among:
+            * :obj:`True` or :obj:`'longest'`: Pad to the longest sequence in the batch (or no padding if only a single
+              sequence if provided).
+            * :obj:`'max_length'`: Pad to a maximum length specified with the argument :obj:`max_length` or to the
+              maximum acceptable input length for the model if that argument is not provided.
+            * :obj:`False` or :obj:`'do_not_pad'` (default): No padding (i.e., can output a batch with sequences of
+              different lengths).
+        max_length (:obj:`int`, `optional`):
+            Maximum length of the ``input_values`` of the returned list and optionally padding length (see above).
+        max_length_labels (:obj:`int`, `optional`):
+            Maximum length of the ``labels`` returned list and optionally padding length (see above).
+        pad_to_multiple_of (:obj:`int`, `optional`):
+            If set will pad the sequence to a multiple of the provided value.
+            This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability >=
+            7.5 (Volta).
+    """
+
+    processor: AutoProcessor
+    padding: Union[bool, str] = "longest"
+    pad_to_multiple_of: Optional[int] = None
+    pad_to_multiple_of_labels: Optional[int] = None
+
+    def __call__(self, features: List[Dict[str, Union[List[int], torch.Tensor]]]) -> Dict[str, torch.Tensor]:
+        # split inputs and labels since they have to be of different lenghts and need
+        # different padding methods
+        input_features = [{"input_values": feature["input_values"]} for feature in features]
+        label_features = [{"input_ids": feature["labels"]} for feature in features]
+
+        batch = self.processor.pad(
+            input_features,
+            padding=self.padding,
+            pad_to_multiple_of=self.pad_to_multiple_of,
+            return_tensors="pt",
+        )
+
+        with self.processor.as_target_processor():
+            labels_batch = self.processor.pad(
+                label_features,
+                padding=self.padding,
+                pad_to_multiple_of=self.pad_to_multiple_of_labels,
+                return_tensors="pt",
+            )
+
+        # replace padding with -100 to ignore loss correctly
+        labels = labels_batch["input_ids"].masked_fill(labels_batch.attention_mask.ne(1), -100)
+
+        batch["labels"] = labels
+
+        return batch
+
+
+def get_tri_stage_schedule(
+    optimizer, num_training_steps, ratios=[0.1, 0.4, 0.5], num_warmup_steps=None, num_hold_steps=None, start_ratio=0.01, end_ratio=0.05 
+):
+    assert (num_warmup_steps is None) == (num_hold_steps is None)
+    if num_warmup_steps is None:
+        num_warmup_steps = int(ratios[0]*num_training_steps)
+        num_hold_steps = int(ratios[1]*num_training_steps)
+        start_decay_step = num_warmup_steps + num_hold_steps
+        a_w, b_w = (1-start_ratio)/num_warmup_steps, start_ratio
+        num_decay_steps = num_training_steps - start_decay_step
+        a_d, b_d = (end_ratio-1)/num_decay_steps, 1.
+        
+    def lr_lambda(current_step):
+        if current_step < num_warmup_steps:
+            return a_w * float(current_step) + b_w
+        if current_step < start_decay_step:
+            return 1.
+        return max(end_ratio, a_d * float(current_step - start_decay_step) + b_d )
+    
+    return LambdaLR(optimizer, lr_lambda)
+
+def create_vocabulary_from_data(
+    datasets: DatasetDict,
+    word_delimiter_token: Optional[str] = None,
+    unk_token: Optional[str] = None,
+    pad_token: Optional[str] = None,
+):
+    # Given training and test labels create vocabulary
+    def extract_all_chars(batch):
+        all_text = " ".join(batch["target_text"])
+        vocab = list(set(all_text))
+        return {"vocab": [vocab], "all_text": [all_text]}
+
+    vocabs = datasets.map(
+        extract_all_chars,
+        batched=True,
+        batch_size=-1,
+        keep_in_memory=True,
+        remove_columns=datasets["train"].column_names,
+    )
+
+    # take union of all unique characters in each dataset
+    vocab_set = functools.reduce(
+        lambda vocab_1, vocab_2: set(vocab_1["vocab"][0]) | set(vocab_2["vocab"][0]), vocabs.values()
+    )
+
+    vocab_dict = {v: k for k, v in enumerate(sorted(list(vocab_set)))}
+
+    # replace white space with delimiter token
+    if word_delimiter_token is not None:
+        vocab_dict[word_delimiter_token] = vocab_dict[" "]
+        del vocab_dict[" "]
+
+    # add unk and pad token
+    if unk_token is not None:
+        vocab_dict[unk_token] = len(vocab_dict)
+
+    if pad_token is not None:
+        vocab_dict[pad_token] = len(vocab_dict)
+
+    return vocab_dict
+
+
+def main():
+    # See all possible arguments in src/transformers/training_args.py
+    # or by passing the --help flag to this script.
+    # We now keep distinct sets of args, for a cleaner separation of concerns.
+
+    parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments, ExtraArguments))
+    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
+        # If we pass only one argument to the script and it's the path to a json file,
+        # let's parse it to get our arguments.
+        model_args, data_args, training_args, extra_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args, data_args, training_args, extra_args = parser.parse_args_into_dataclasses()
+
+    # Detecting last checkpoint.
+    last_checkpoint = None
+    if os.path.isdir(training_args.output_dir) and training_args.do_train and not training_args.overwrite_output_dir:
+        last_checkpoint = get_last_checkpoint(training_args.output_dir)
+        if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0:
+            raise ValueError(
+                f"Output directory ({training_args.output_dir}) already exists and is not empty. "
+                "Use --overwrite_output_dir to overcome."
+            )
+        elif last_checkpoint is not None:
+            logger.info(
+                f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change "
+                "the `--output_dir` or add `--overwrite_output_dir` to train from scratch."
+            )
+
+    # Setup logging
+    logging.basicConfig(
+        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
+        datefmt="%m/%d/%Y %H:%M:%S",
+        handlers=[logging.StreamHandler(sys.stdout)],
+    )
+    logger.setLevel(logging.INFO if is_main_process(training_args.local_rank) else logging.WARN)
+
+    # Log on each process the small summary:
+    logger.warning(
+        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+        f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
+    )
+    # Set the verbosity to info of the Transformers logger (on main process only):
+    if is_main_process(training_args.local_rank):
+        transformers.utils.logging.set_verbosity_info()
+    logger.info("Training/evaluation parameters %s", training_args)
+
+    # Set seed before initializing model.
+    set_seed(training_args.seed)
+
+    # configure wandb run
+    os.environ["WANDB_PROJECT"] = extra_args.wandb_project
+
+    # 1. First, let's load the dataset
+    raw_datasets = DatasetDict()
+
+    if training_args.do_train:
+        raw_datasets["train"] = load_dataset(
+            data_args.dataset_name,
+            data_args.dataset_config_name,
+            split=data_args.train_split_name,
+            use_auth_token=data_args.use_auth_token,
+        )
+
+        if data_args.audio_column_name not in raw_datasets["train"].column_names:
+            raise ValueError(
+                f"--audio_column_name '{data_args.audio_column_name}' not found in dataset '{data_args.dataset_name}'. "
+                "Make sure to set `--audio_column_name` to the correct audio column - one of "
+                f"{', '.join(raw_datasets['train'].column_names)}."
+            )
+
+        if data_args.text_column_name not in raw_datasets["train"].column_names:
+            raise ValueError(
+                f"--text_column_name {data_args.text_column_name} not found in dataset '{data_args.dataset_name}'. "
+                "Make sure to set `--text_column_name` to the correct text column - one of "
+                f"{', '.join(raw_datasets['train'].column_names)}."
+            )
+
+        if data_args.max_train_samples is not None:
+            raw_datasets["train"] = raw_datasets["train"].select(range(data_args.max_train_samples))
+
+    if training_args.do_eval:
+        raw_datasets["eval"] = load_dataset(
+            data_args.dataset_name,
+            data_args.dataset_config_name,
+            split=data_args.eval_split_name,
+            use_auth_token=data_args.use_auth_token,
+        )
+
+        if data_args.max_eval_samples is not None:
+            raw_datasets["eval"] = raw_datasets["eval"].select(range(data_args.max_eval_samples))
+
+    # 2. We remove some special characters from the datasets
+    # that make training complicated and do not help in transcribing the speech
+    # E.g. characters, such as `,` and `.` do not really have an acoustic characteristic
+    # that could be easily picked up by the model
+    chars_to_ignore_regex = (
+        f'[{"".join(data_args.chars_to_ignore)}]' if data_args.chars_to_ignore is not None else None
+    )
+    text_column_name = data_args.text_column_name
+    
+    def remove_special_characters(batch):
+        text = batch[text_column_name]
+        # normalize apostrophe first
+        text = re.sub("['՝՛՚‘`]", "’", text)
+        if chars_to_ignore_regex is not None:
+            text = re.sub(chars_to_ignore_regex, "", text.lower()) + " "
+        else:
+            text = text.lower() + " "
+        for a, b in zip("aceioxklmnpujы", "асеіохклмнпужи"):
+            text = re.sub(a, b, text)
+        batch["target_text"] = text
+        return batch
+
+    with training_args.main_process_first(desc="dataset map special characters removal"):
+        raw_datasets = raw_datasets.map(
+            remove_special_characters,
+            remove_columns=[text_column_name],
+            desc="remove special characters from datasets",
+        )
+
+    # save special tokens for tokenizer
+    word_delimiter_token = data_args.word_delimiter_token
+    unk_token = data_args.unk_token
+    pad_token = data_args.pad_token
+
+    # 3. Next, let's load the config as we might need it to create
+    # the tokenizer
+    # load config
+    config = AutoConfig.from_pretrained(
+        model_args.model_name_or_path, cache_dir=model_args.cache_dir, use_auth_token=data_args.use_auth_token
+    )
+
+    # 4. Next, if no tokenizer file is defined,
+    # we create the vocabulary of the model by extracting all unique characters from
+    # the training and evaluation datasets
+    # We need to make sure that only first rank saves vocabulary
+    # make sure all processes wait until vocab is created
+    tokenizer_name_or_path = model_args.tokenizer_name_or_path
+    tokenizer_kwargs = {}
+    if tokenizer_name_or_path is None:
+        # save vocab in training output dir
+        tokenizer_name_or_path = training_args.output_dir
+
+        vocab_file = os.path.join(tokenizer_name_or_path, "vocab.json")
+
+        with training_args.main_process_first():
+            if training_args.overwrite_output_dir and os.path.isfile(vocab_file):
+                os.remove(vocab_file)
+
+        with training_args.main_process_first(desc="dataset map vocabulary creation"):
+            if not os.path.isfile(vocab_file):
+                os.makedirs(tokenizer_name_or_path, exist_ok=True)
+                vocab_dict = create_vocabulary_from_data(
+                    raw_datasets,
+                    word_delimiter_token=word_delimiter_token,
+                    unk_token=unk_token,
+                    pad_token=pad_token,
+                )
+
+                # save vocab dict to be loaded into tokenizer
+                with open(vocab_file, "w") as file:
+                    json.dump(vocab_dict, file)
+
+        # if tokenizer has just been created
+        # it is defined by `tokenizer_class` if present in config else by `model_type`
+        tokenizer_kwargs = {
+            "config": config if config.tokenizer_class is not None else None,
+            "tokenizer_type": config.model_type if config.tokenizer_class is None else None,
+            "unk_token": unk_token,
+            "pad_token": pad_token,
+            "word_delimiter_token": word_delimiter_token,
+        }
+
+    # 5. Now we can instantiate the feature extractor, tokenizer and model
+    # Note for distributed training, the .from_pretrained methods guarantee that only
+    # one local process can concurrently download model & vocab.
+
+    # load feature_extractor and tokenizer
+    tokenizer = AutoTokenizer.from_pretrained(
+        tokenizer_name_or_path,
+        use_auth_token=data_args.use_auth_token,
+        **tokenizer_kwargs,
+    )
+    feature_extractor = AutoFeatureExtractor.from_pretrained(
+        model_args.model_name_or_path, cache_dir=model_args.cache_dir, use_auth_token=data_args.use_auth_token
+    )
+
+    # adapt config
+    config.update(
+        {
+            "feat_proj_dropout": model_args.feat_proj_dropout,
+            "attention_dropout": model_args.attention_dropout,
+            "hidden_dropout": model_args.hidden_dropout,
+            "final_dropout": model_args.final_dropout,
+            "mask_time_prob": model_args.mask_time_prob,
+            "mask_time_length": model_args.mask_time_length,
+            "mask_feature_prob": model_args.mask_feature_prob,
+            "mask_feature_length": model_args.mask_feature_length,
+            "gradient_checkpointing": training_args.gradient_checkpointing,
+            "layerdrop": model_args.layerdrop,
+            "ctc_loss_reduction": model_args.ctc_loss_reduction,
+            "pad_token_id": tokenizer.pad_token_id,
+            "vocab_size": len(tokenizer),
+            "activation_dropout": model_args.activation_dropout,
+        }
+    )
+
+    # create model
+    model = AutoModelForCTC.from_pretrained(
+        model_args.model_name_or_path,
+        cache_dir=model_args.cache_dir,
+        config=config,
+        use_auth_token=data_args.use_auth_token,
+    )
+
+    # freeze encoder
+    if model_args.freeze_feature_encoder:
+        model.freeze_feature_encoder()
+
+    # 6. Now we preprocess the datasets including loading the audio, resampling and normalization
+    # Thankfully, `datasets` takes care of automatically loading and resampling the audio,
+    # so that we just need to set the correct target sampling rate and normalize the input
+    # via the `feature_extractor`
+
+    # make sure that dataset decodes audio with correct sampling rate
+    dataset_sampling_rate = next(iter(raw_datasets.values())).features[data_args.audio_column_name].sampling_rate
+    if dataset_sampling_rate != feature_extractor.sampling_rate:
+        raw_datasets = raw_datasets.cast_column(
+            data_args.audio_column_name, datasets.features.Audio(sampling_rate=feature_extractor.sampling_rate)
+        )
+
+    # derive max & min input length for sample rate & max duration
+    max_input_length = data_args.max_duration_in_seconds * feature_extractor.sampling_rate
+    min_input_length = data_args.min_duration_in_seconds * feature_extractor.sampling_rate
+    audio_column_name = data_args.audio_column_name
+    num_workers = data_args.preprocessing_num_workers
+
+    # `phoneme_language` is only relevant if the model is fine-tuned on phoneme classification
+    phoneme_language = data_args.phoneme_language
+
+    # Preprocessing the datasets.
+    # We need to read the audio files as arrays and tokenize the targets.
+    def prepare_dataset(batch):
+        # load audio
+        sample = batch[audio_column_name]
+
+        inputs = feature_extractor(sample["array"], sampling_rate=sample["sampling_rate"])
+        batch["input_values"] = inputs.input_values[0]
+        batch["length"] = len(batch["input_values"])
+
+        # encode targets
+        additional_kwargs = {}
+        if phoneme_language is not None:
+            additional_kwargs["phonemizer_lang"] = phoneme_language
+
+        batch["labels"] = tokenizer(batch["target_text"], **additional_kwargs).input_ids
+        return batch
+
+    with training_args.main_process_first(desc="dataset map preprocessing"):
+        vectorized_datasets = raw_datasets.map(
+            prepare_dataset,
+            remove_columns=next(iter(raw_datasets.values())).column_names,
+            num_proc=num_workers,
+            desc="preprocess datasets",
+        )
+
+        def is_audio_in_length_range(length):
+            return length > min_input_length and length < max_input_length
+
+        # filter data that is shorter than min_input_length
+        vectorized_datasets = vectorized_datasets.filter(
+            is_audio_in_length_range,
+            num_proc=num_workers,
+            input_columns=["length"],
+        )
+
+    # 7. Next, we can prepare the training.
+    # Let's use word error rate (WER) as our evaluation metric,
+    # instantiate a data collator and the trainer
+
+    # Define evaluation metrics during training, *i.e.* word error rate, character error rate
+    eval_metrics = {metric: load_metric(metric) for metric in data_args.eval_metrics}
+
+    # for large datasets it is advised to run the preprocessing on a
+    # single machine first with ``args.preprocessing_only`` since there will mostly likely
+    # be a timeout when running the script in distributed mode.
+    # In a second step ``args.preprocessing_only`` can then be set to `False` to load the
+    # cached dataset
+    if data_args.preprocessing_only:
+        logger.info(f"Data preprocessing finished. Files cached at {vectorized_datasets.cache_files}")
+        return
+
+    def compute_metrics(pred):
+        pred_logits = pred.predictions
+        pred_ids = np.argmax(pred_logits, axis=-1)
+
+        pred.label_ids[pred.label_ids == -100] = tokenizer.pad_token_id
+
+        pred_str = tokenizer.batch_decode(pred_ids)
+        # we do not want to group tokens when computing the metrics
+        label_str = tokenizer.batch_decode(pred.label_ids, group_tokens=False)
+
+        metrics = {k: v.compute(predictions=pred_str, references=label_str) for k, v in eval_metrics.items()}
+
+        return metrics
+
+    # Now save everything to be able to create a single processor later
+    if is_main_process(training_args.local_rank):
+        # save feature extractor, tokenizer and config
+        feature_extractor.save_pretrained(training_args.output_dir)
+        tokenizer.save_pretrained(training_args.output_dir)
+        config.save_pretrained(training_args.output_dir)
+
+    try:
+        processor = AutoProcessor.from_pretrained(training_args.output_dir)
+    except (OSError, KeyError):
+        warnings.warn(
+            "Loading a processor from a feature extractor config that does not"
+            " include a `processor_class` attribute is deprecated and will be removed in v5. Please add the following "
+            " attribute to your `preprocessor_config.json` file to suppress this warning: "
+            " `'processor_class': 'Wav2Vec2Processor'`",
+            FutureWarning,
+        )
+        processor = Wav2Vec2Processor.from_pretrained(training_args.output_dir)
+
+    # Instantiate custom data collator
+    data_collator = DataCollatorCTCWithPadding(processor=processor)
+
+    decay_parameters = get_parameter_names(model, [torch.nn.LayerNorm])
+    decay_parameters = [name for name in decay_parameters if "bias" not in name]
+    optimizer_grouped_parameters = [
+        {
+            "params": [p for n, p in model.named_parameters() if n in decay_parameters],
+            "weight_decay": training_args.weight_decay,
+        },
+        {
+            "params": [p for n, p in model.named_parameters() if n not in decay_parameters],
+            "weight_decay": 0.0,
+        },
+    ]
+    if extra_args.bnb:
+        optimizer = bnb.optim.Adam8bit(
+            params=optimizer_grouped_parameters,
+            lr=training_args.learning_rate,
+            betas=(training_args.adam_beta1, training_args.adam_beta2),
+            eps=training_args.adam_epsilon,
+        )
+    else:
+        optimizer = torch.optim.AdamW(
+            params=optimizer_grouped_parameters,
+            lr=training_args.learning_rate,
+            betas=(training_args.adam_beta1, training_args.adam_beta2),
+            eps=training_args.adam_epsilon,
+        )
+    if extra_args.tristage_sched:
+        scheduler = get_tri_stage_schedule(optimizer, training_args.max_steps)
+    else:
+        scheduler = None
+    optimizers = (optimizer, scheduler)
+
+    # Initialize Trainer
+    trainer = Trainer(
+        model=model,
+        data_collator=data_collator,
+        args=training_args,
+        compute_metrics=compute_metrics,
+        train_dataset=vectorized_datasets["train"] if training_args.do_train else None,
+        eval_dataset=vectorized_datasets["eval"] if training_args.do_eval else None,
+        tokenizer=feature_extractor,
+        optimizers=optimizers,
+    )
+
+    # 8. Finally, we can start training
+
+    # Training
+    if training_args.do_train:
+
+        # use last checkpoint if exist
+        if last_checkpoint is not None:
+            checkpoint = last_checkpoint
+        elif os.path.isdir(model_args.model_name_or_path):
+            checkpoint = model_args.model_name_or_path
+        else:
+            checkpoint = None
+
+        train_result = trainer.train(resume_from_checkpoint=checkpoint)
+        trainer.save_model()
+
+        metrics = train_result.metrics
+        max_train_samples = (
+            data_args.max_train_samples
+            if data_args.max_train_samples is not None
+            else len(vectorized_datasets["train"])
+        )
+        metrics["train_samples"] = min(max_train_samples, len(vectorized_datasets["train"]))
+
+        trainer.log_metrics("train", metrics)
+        trainer.save_metrics("train", metrics)
+        trainer.save_state()
+
+    # Evaluation
+    results = {}
+    if training_args.do_eval:
+        logger.info("*** Evaluate ***")
+        metrics = trainer.evaluate()
+        max_eval_samples = (
+            data_args.max_eval_samples if data_args.max_eval_samples is not None else len(vectorized_datasets["eval"])
+        )
+        metrics["eval_samples"] = min(max_eval_samples, len(vectorized_datasets["eval"]))
+
+        trainer.log_metrics("eval", metrics)
+        trainer.save_metrics("eval", metrics)
+
+    # Write model card and (optionally) push to hub
+    config_name = data_args.dataset_config_name if data_args.dataset_config_name is not None else "na"
+    kwargs = {
+        "finetuned_from": model_args.model_name_or_path,
+        "tasks": "speech-recognition",
+        "tags": ["automatic-speech-recognition", data_args.dataset_name],
+        "dataset_args": f"Config: {config_name}, Training split: {data_args.train_split_name}, Eval split: {data_args.eval_split_name}",
+        "dataset": f"{data_args.dataset_name.upper()} - {config_name.upper()}",
+    }
+    if "common_voice" in data_args.dataset_name:
+        kwargs["language"] = config_name
+
+    if training_args.push_to_hub:
+        trainer.push_to_hub(**kwargs)
+    else:
+        trainer.create_model_card(**kwargs)
+
+    return results
+
+
+if __name__ == "__main__":
+    main()

runs/Feb01_12-24-43_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/1643718576.9397461/events.out.tfevents.1643718576.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.42275.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97e0ce2b9c9e8bbc043dbd4024a232c732938d83558db8e8111016dde634d0ed
+size 4772

runs/Feb01_12-24-43_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/events.out.tfevents.1643718576.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.42275.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c0db5e99e87f736d9955ce0ddcec8880697a0eef171e40c9d48fcf33ebf568e
+size 23453

runs/Feb01_12-24-43_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/events.out.tfevents.1643767924.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.42275.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:552b8002384acb7a5bd2ea9b6290e9ad5a2557387b41f0cc540fd8b8be1e7e35
+size 405

runs/Jan29_23-24-58_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/1643498904.5496016/events.out.tfevents.1643498904.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.15864.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18ae84a878579643857afaae9439fe039c9b895752884493b5fbe6b4383906b5
+size 4772

runs/Jan29_23-24-58_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/events.out.tfevents.1643498904.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.15864.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5b40435508986a6b67e10ffd32361114b87a6bb7bdc7dee41614a7c909aa4af
+size 20706

runs/Jan29_23-24-58_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/events.out.tfevents.1643522448.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.15864.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5355c8d035b3de8c259861d7c442129b5944e8e0a4a014dceaaa7353ab61a26a
+size 405

runs/Jan31_02-14-35_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/1643595489.2781765/events.out.tfevents.1643595489.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.49449.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15841f99516fb0eaf5801b1bcbe20f6542d9aca95ae7905edfd44d894130a8ac
+size 4772

runs/Jan31_02-14-35_job-b1f4681b-d20d-47f2-af64-0c1734f4ff64/events.out.tfevents.1643595489.job-b1f4681b-d20d-47f2-af64-0c1734f4ff64.49449.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:349a3f5226ed0989582e3efe46eea28a55a2182165bb76f57ff79588f851067e
+size 30792

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "/workspace/output/uk/models/wav2vec2-xls-r-1b-uk-cv/", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2ProcessorWithLM"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 69.57,
+    "train_loss": 1.039798891067505,
+    "train_runtime": 49145.562,
+    "train_samples": 14717,
+    "train_samples_per_second": 20.836,
+    "train_steps_per_second": 0.163
+}

trainer_state.json

@@ -0,0 +1,665 @@
+{
+  "best_metric": 0.1746993511915207,
+  "best_model_checkpoint": "./checkpoint-8000",
+  "epoch": 69.56521739130434,
+  "global_step": 8000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.0304e-05,
+      "loss": 6.0688,
+      "step": 100
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.0204000000000002e-05,
+      "loss": 3.1889,
+      "step": 200
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 3.0104e-05,
+      "loss": 2.2092,
+      "step": 300
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 4.000399999999999e-05,
+      "loss": 1.5574,
+      "step": 400
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 4.9903999999999995e-05,
+      "loss": 1.3719,
+      "step": 500
+    },
+    {
+      "epoch": 4.35,
+      "eval_cer": 0.08328164981226813,
+      "eval_loss": 0.3388712704181671,
+      "eval_runtime": 199.6026,
+      "eval_samples_per_second": 29.068,
+      "eval_steps_per_second": 0.456,
+      "eval_wer": 0.42359243003445185,
+      "step": 500
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 5.9804e-05,
+      "loss": 1.3634,
+      "step": 600
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 6.970400000000001e-05,
+      "loss": 1.3345,
+      "step": 700
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 7.960399999999999e-05,
+      "loss": 1.2407,
+      "step": 800
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 8e-05,
+      "loss": 1.172,
+      "step": 900
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 8e-05,
+      "loss": 1.1361,
+      "step": 1000
+    },
+    {
+      "epoch": 8.7,
+      "eval_cer": 0.06299695098173931,
+      "eval_loss": 0.23091697692871094,
+      "eval_runtime": 209.236,
+      "eval_samples_per_second": 27.729,
+      "eval_steps_per_second": 0.435,
+      "eval_wer": 0.31617820567275473,
+      "step": 1000
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8e-05,
+      "loss": 1.0819,
+      "step": 1100
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 8e-05,
+      "loss": 1.0923,
+      "step": 1200
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 8e-05,
+      "loss": 1.0803,
+      "step": 1300
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 8e-05,
+      "loss": 1.0719,
+      "step": 1400
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 8e-05,
+      "loss": 1.0517,
+      "step": 1500
+    },
+    {
+      "epoch": 13.04,
+      "eval_cer": 0.059732533136255095,
+      "eval_loss": 0.21664097905158997,
+      "eval_runtime": 198.1012,
+      "eval_samples_per_second": 29.288,
+      "eval_steps_per_second": 0.459,
+      "eval_wer": 0.3056066827127283,
+      "step": 1500
+    },
+    {
+      "epoch": 13.91,
+      "learning_rate": 8e-05,
+      "loss": 1.0523,
+      "step": 1600
+    },
+    {
+      "epoch": 14.78,
+      "learning_rate": 8e-05,
+      "loss": 1.0414,
+      "step": 1700
+    },
+    {
+      "epoch": 15.65,
+      "learning_rate": 8e-05,
+      "loss": 1.031,
+      "step": 1800
+    },
+    {
+      "epoch": 16.52,
+      "learning_rate": 8e-05,
+      "loss": 1.0126,
+      "step": 1900
+    },
+    {
+      "epoch": 17.39,
+      "learning_rate": 8e-05,
+      "loss": 1.0118,
+      "step": 2000
+    },
+    {
+      "epoch": 17.39,
+      "eval_cer": 0.055684506456416864,
+      "eval_loss": 0.21414823830127716,
+      "eval_runtime": 204.5017,
+      "eval_samples_per_second": 28.371,
+      "eval_steps_per_second": 0.445,
+      "eval_wer": 0.2783991693803389,
+      "step": 2000
+    },
+    {
+      "epoch": 18.26,
+      "learning_rate": 8e-05,
+      "loss": 1.0243,
+      "step": 2100
+    },
+    {
+      "epoch": 19.13,
+      "learning_rate": 8e-05,
+      "loss": 1.0084,
+      "step": 2200
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 8e-05,
+      "loss": 1.0074,
+      "step": 2300
+    },
+    {
+      "epoch": 20.87,
+      "learning_rate": 8e-05,
+      "loss": 0.9933,
+      "step": 2400
+    },
+    {
+      "epoch": 21.74,
+      "learning_rate": 8e-05,
+      "loss": 0.9922,
+      "step": 2500
+    },
+    {
+      "epoch": 21.74,
+      "eval_cer": 0.05935372696988491,
+      "eval_loss": 0.22312845289707184,
+      "eval_runtime": 198.4087,
+      "eval_samples_per_second": 29.243,
+      "eval_steps_per_second": 0.459,
+      "eval_wer": 0.29411487092359245,
+      "step": 2500
+    },
+    {
+      "epoch": 22.61,
+      "learning_rate": 8e-05,
+      "loss": 0.9838,
+      "step": 2600
+    },
+    {
+      "epoch": 23.48,
+      "learning_rate": 8e-05,
+      "loss": 1.019,
+      "step": 2700
+    },
+    {
+      "epoch": 24.35,
+      "learning_rate": 8e-05,
+      "loss": 0.9897,
+      "step": 2800
+    },
+    {
+      "epoch": 25.22,
+      "learning_rate": 8e-05,
+      "loss": 0.9897,
+      "step": 2900
+    },
+    {
+      "epoch": 26.09,
+      "learning_rate": 8e-05,
+      "loss": 0.9929,
+      "step": 3000
+    },
+    {
+      "epoch": 26.09,
+      "eval_cer": 0.05865182142631663,
+      "eval_loss": 0.21711210906505585,
+      "eval_runtime": 199.7486,
+      "eval_samples_per_second": 29.047,
+      "eval_steps_per_second": 0.456,
+      "eval_wer": 0.28915946953608,
+      "step": 3000
+    },
+    {
+      "epoch": 26.96,
+      "learning_rate": 8e-05,
+      "loss": 0.9971,
+      "step": 3100
+    },
+    {
+      "epoch": 27.83,
+      "learning_rate": 8e-05,
+      "loss": 0.9621,
+      "step": 3200
+    },
+    {
+      "epoch": 28.7,
+      "learning_rate": 8e-05,
+      "loss": 0.9744,
+      "step": 3300
+    },
+    {
+      "epoch": 29.57,
+      "learning_rate": 8e-05,
+      "loss": 0.9587,
+      "step": 3400
+    },
+    {
+      "epoch": 30.43,
+      "learning_rate": 8e-05,
+      "loss": 0.9485,
+      "step": 3500
+    },
+    {
+      "epoch": 30.43,
+      "eval_cer": 0.05992565000538499,
+      "eval_loss": 0.2236333191394806,
+      "eval_runtime": 197.9762,
+      "eval_samples_per_second": 29.307,
+      "eval_steps_per_second": 0.46,
+      "eval_wer": 0.29560149133984615,
+      "step": 3500
+    },
+    {
+      "epoch": 31.3,
+      "learning_rate": 8e-05,
+      "loss": 0.9468,
+      "step": 3600
+    },
+    {
+      "epoch": 32.17,
+      "learning_rate": 8e-05,
+      "loss": 0.9598,
+      "step": 3700
+    },
+    {
+      "epoch": 33.04,
+      "learning_rate": 8e-05,
+      "loss": 0.9575,
+      "step": 3800
+    },
+    {
+      "epoch": 33.91,
+      "learning_rate": 8e-05,
+      "loss": 0.9491,
+      "step": 3900
+    },
+    {
+      "epoch": 34.78,
+      "learning_rate": 8e-05,
+      "loss": 0.9573,
+      "step": 4000
+    },
+    {
+      "epoch": 34.78,
+      "eval_cer": 0.061589426108657946,
+      "eval_loss": 0.2313707023859024,
+      "eval_runtime": 198.2807,
+      "eval_samples_per_second": 29.262,
+      "eval_steps_per_second": 0.459,
+      "eval_wer": 0.304285242342725,
+      "step": 4000
+    },
+    {
+      "epoch": 35.65,
+      "learning_rate": 7.8176e-05,
+      "loss": 0.9251,
+      "step": 4100
+    },
+    {
+      "epoch": 36.52,
+      "learning_rate": 7.627600000000001e-05,
+      "loss": 0.9299,
+      "step": 4200
+    },
+    {
+      "epoch": 37.39,
+      "learning_rate": 7.437600000000001e-05,
+      "loss": 0.928,
+      "step": 4300
+    },
+    {
+      "epoch": 38.26,
+      "learning_rate": 7.2476e-05,
+      "loss": 0.9303,
+      "step": 4400
+    },
+    {
+      "epoch": 39.13,
+      "learning_rate": 7.057600000000001e-05,
+      "loss": 0.9195,
+      "step": 4500
+    },
+    {
+      "epoch": 39.13,
+      "eval_cer": 0.058024191601644466,
+      "eval_loss": 0.21694457530975342,
+      "eval_runtime": 198.1273,
+      "eval_samples_per_second": 29.284,
+      "eval_steps_per_second": 0.459,
+      "eval_wer": 0.2812308273160602,
+      "step": 4500
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 6.8676e-05,
+      "loss": 0.9182,
+      "step": 4600
+    },
+    {
+      "epoch": 40.87,
+      "learning_rate": 6.6776e-05,
+      "loss": 0.8885,
+      "step": 4700
+    },
+    {
+      "epoch": 41.74,
+      "learning_rate": 6.487600000000001e-05,
+      "loss": 0.9037,
+      "step": 4800
+    },
+    {
+      "epoch": 42.61,
+      "learning_rate": 6.2976e-05,
+      "loss": 0.8929,
+      "step": 4900
+    },
+    {
+      "epoch": 43.48,
+      "learning_rate": 6.1076e-05,
+      "loss": 0.8915,
+      "step": 5000
+    },
+    {
+      "epoch": 43.48,
+      "eval_cer": 0.05601503340550457,
+      "eval_loss": 0.21094831824302673,
+      "eval_runtime": 198.7631,
+      "eval_samples_per_second": 29.191,
+      "eval_steps_per_second": 0.458,
+      "eval_wer": 0.2779508235405163,
+      "step": 5000
+    },
+    {
+      "epoch": 44.35,
+      "learning_rate": 5.9176000000000004e-05,
+      "loss": 0.8687,
+      "step": 5100
+    },
+    {
+      "epoch": 45.22,
+      "learning_rate": 5.7276000000000005e-05,
+      "loss": 0.8635,
+      "step": 5200
+    },
+    {
+      "epoch": 46.09,
+      "learning_rate": 5.5376e-05,
+      "loss": 0.859,
+      "step": 5300
+    },
+    {
+      "epoch": 46.96,
+      "learning_rate": 5.3476e-05,
+      "loss": 0.8586,
+      "step": 5400
+    },
+    {
+      "epoch": 47.83,
+      "learning_rate": 5.157600000000001e-05,
+      "loss": 0.8449,
+      "step": 5500
+    },
+    {
+      "epoch": 47.83,
+      "eval_cer": 0.05136537340260782,
+      "eval_loss": 0.20504631102085114,
+      "eval_runtime": 198.4879,
+      "eval_samples_per_second": 29.231,
+      "eval_steps_per_second": 0.458,
+      "eval_wer": 0.25338619094813347,
+      "step": 5500
+    },
+    {
+      "epoch": 48.7,
+      "learning_rate": 4.9676000000000003e-05,
+      "loss": 0.8345,
+      "step": 5600
+    },
+    {
+      "epoch": 49.57,
+      "learning_rate": 4.7776e-05,
+      "loss": 0.8229,
+      "step": 5700
+    },
+    {
+      "epoch": 50.43,
+      "learning_rate": 4.5876000000000006e-05,
+      "loss": 0.8203,
+      "step": 5800
+    },
+    {
+      "epoch": 51.3,
+      "learning_rate": 4.397600000000001e-05,
+      "loss": 0.8084,
+      "step": 5900
+    },
+    {
+      "epoch": 52.17,
+      "learning_rate": 4.207600000000001e-05,
+      "loss": 0.8028,
+      "step": 6000
+    },
+    {
+      "epoch": 52.17,
+      "eval_cer": 0.04915567076544842,
+      "eval_loss": 0.2032497674226761,
+      "eval_runtime": 199.0329,
+      "eval_samples_per_second": 29.151,
+      "eval_steps_per_second": 0.457,
+      "eval_wer": 0.24562272877436406,
+      "step": 6000
+    },
+    {
+      "epoch": 53.04,
+      "learning_rate": 4.0176e-05,
+      "loss": 0.7975,
+      "step": 6100
+    },
+    {
+      "epoch": 53.91,
+      "learning_rate": 3.8295000000000005e-05,
+      "loss": 0.7942,
+      "step": 6200
+    },
+    {
+      "epoch": 54.78,
+      "learning_rate": 3.6395e-05,
+      "loss": 0.7867,
+      "step": 6300
+    },
+    {
+      "epoch": 55.65,
+      "learning_rate": 3.4495e-05,
+      "loss": 0.787,
+      "step": 6400
+    },
+    {
+      "epoch": 56.52,
+      "learning_rate": 3.2595e-05,
+      "loss": 0.7881,
+      "step": 6500
+    },
+    {
+      "epoch": 56.52,
+      "eval_cer": 0.04694596812828902,
+      "eval_loss": 0.18896546959877014,
+      "eval_runtime": 199.9061,
+      "eval_samples_per_second": 29.024,
+      "eval_steps_per_second": 0.455,
+      "eval_wer": 0.2380008494973807,
+      "step": 6500
+    },
+    {
+      "epoch": 57.39,
+      "learning_rate": 3.0695e-05,
+      "loss": 0.7608,
+      "step": 6600
+    },
+    {
+      "epoch": 58.26,
+      "learning_rate": 2.8795000000000005e-05,
+      "loss": 0.7542,
+      "step": 6700
+    },
+    {
+      "epoch": 59.13,
+      "learning_rate": 2.6895e-05,
+      "loss": 0.755,
+      "step": 6800
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 2.4995000000000004e-05,
+      "loss": 0.7569,
+      "step": 6900
+    },
+    {
+      "epoch": 60.87,
+      "learning_rate": 2.3095e-05,
+      "loss": 0.7423,
+      "step": 7000
+    },
+    {
+      "epoch": 60.87,
+      "eval_cer": 0.04418662517129838,
+      "eval_loss": 0.18159770965576172,
+      "eval_runtime": 198.6364,
+      "eval_samples_per_second": 29.209,
+      "eval_steps_per_second": 0.458,
+      "eval_wer": 0.22452687715323988,
+      "step": 7000
+    },
+    {
+      "epoch": 61.74,
+      "learning_rate": 2.1195000000000006e-05,
+      "loss": 0.7409,
+      "step": 7100
+    },
+    {
+      "epoch": 62.61,
+      "learning_rate": 1.9295e-05,
+      "loss": 0.7208,
+      "step": 7200
+    },
+    {
+      "epoch": 63.48,
+      "learning_rate": 1.7395e-05,
+      "loss": 0.7488,
+      "step": 7300
+    },
+    {
+      "epoch": 64.35,
+      "learning_rate": 1.5495000000000003e-05,
+      "loss": 0.7134,
+      "step": 7400
+    },
+    {
+      "epoch": 65.22,
+      "learning_rate": 1.3594999999999998e-05,
+      "loss": 0.7248,
+      "step": 7500
+    },
+    {
+      "epoch": 65.22,
+      "eval_cer": 0.042203463476772125,
+      "eval_loss": 0.17892056703567505,
+      "eval_runtime": 197.8292,
+      "eval_samples_per_second": 29.328,
+      "eval_steps_per_second": 0.46,
+      "eval_wer": 0.21650384633536268,
+      "step": 7500
+    },
+    {
+      "epoch": 66.09,
+      "learning_rate": 1.1695000000000002e-05,
+      "loss": 0.7076,
+      "step": 7600
+    },
+    {
+      "epoch": 66.96,
+      "learning_rate": 9.794999999999999e-06,
+      "loss": 0.7132,
+      "step": 7700
+    },
+    {
+      "epoch": 67.83,
+      "learning_rate": 7.895000000000003e-06,
+      "loss": 0.6972,
+      "step": 7800
+    },
+    {
+      "epoch": 68.7,
+      "learning_rate": 5.994999999999999e-06,
+      "loss": 0.7019,
+      "step": 7900
+    },
+    {
+      "epoch": 69.57,
+      "learning_rate": 4.095000000000005e-06,
+      "loss": 0.6993,
+      "step": 8000
+    },
+    {
+      "epoch": 69.57,
+      "eval_cer": 0.040792224817745956,
+      "eval_loss": 0.1746993511915207,
+      "eval_runtime": 198.0151,
+      "eval_samples_per_second": 29.301,
+      "eval_steps_per_second": 0.46,
+      "eval_wer": 0.21072254471659824,
+      "step": 8000
+    },
+    {
+      "epoch": 69.57,
+      "step": 8000,
+      "total_flos": 3.5433976605211066e+20,
+      "train_loss": 1.039798891067505,
+      "train_runtime": 49145.562,
+      "train_samples_per_second": 20.836,
+      "train_steps_per_second": 0.163
+    }
+  ],
+  "max_steps": 8000,
+  "num_train_epochs": 70,
+  "total_flos": 3.5433976605211066e+20,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41664458c700f011b982fd12459ef4f67a27b7782f82eb5a754452d5e0174acb
+size 3055

vocab.json

@@ -0,0 +1 @@
+{"а": 1, "б": 2, "в": 3, "г": 4, "д": 5, "е": 6, "ж": 7, "з": 8, "и": 9, "й": 10, "к": 11, "л": 12, "м": 13, "н": 14, "о": 15, "п": 16, "р": 17, "с": 18, "т": 19, "у": 20, "ф": 21, "х": 22, "ц": 23, "ч": 24, "ш": 25, "щ": 26, "ь": 27, "ю": 28, "я": 29, "є": 30, "і": 31, "ї": 32, "ґ": 33, "’": 34, "|": 0, "[UNK]": 35, "[PAD]": 36}