Training in progress, step 1500

.gitattributes

@@ -25,3 +25,4 @@ 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
+wandb/run-20220308_102248-3h7e1p8o/run-3h7e1p8o.wandb filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

config.json

@@ -0,0 +1,252 @@
+{
+  "_name_or_path": "./",
+  "architectures": [
+    "SpeechEncoderDecoderModel"
+  ],
+  "decoder": {
+    "_name_or_path": "decoder",
+    "add_cross_attention": true,
+    "architectures": [
+      "BertLMHeadModel"
+    ],
+    "attention_probs_dropout_prob": 0.0,
+    "bad_words_ids": null,
+    "bos_token_id": null,
+    "chunk_size_feed_forward": 0,
+    "classifier_dropout": null,
+    "cross_attention_hidden_size": null,
+    "decoder_start_token_id": null,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "eos_token_id": null,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "gradient_checkpointing": false,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.1,
+    "hidden_size": 1024,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "is_decoder": true,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "layer_norm_eps": 1e-12,
+    "length_penalty": 1.0,
+    "max_length": 20,
+    "max_position_embeddings": 512,
+    "min_length": 0,
+    "model_type": "bert",
+    "no_repeat_ngram_size": 0,
+    "num_attention_heads": 16,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_hidden_layers": 2,
+    "num_return_sequences": 1,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": 0,
+    "position_embedding_type": "absolute",
+    "prefix": null,
+    "problem_type": null,
+    "pruned_heads": {},
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "sep_token_id": null,
+    "task_specific_params": null,
+    "temperature": 1.0,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": true,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": "float32",
+    "torchscript": false,
+    "transformers_version": "4.17.0.dev0",
+    "type_vocab_size": 2,
+    "use_bfloat16": false,
+    "use_cache": false,
+    "vocab_size": 30522
+  },
+  "decoder_start_token_id": 101,
+  "encoder": {
+    "_name_or_path": "facebook/wav2vec2-large-lv60",
+    "activation_dropout": 0.0,
+    "adapter_kernel_size": 3,
+    "adapter_stride": 2,
+    "add_adapter": true,
+    "add_cross_attention": false,
+    "apply_spec_augment": false,
+    "architectures": [
+      "Wav2Vec2ForPreTraining"
+    ],
+    "attention_dropout": 0.0,
+    "bad_words_ids": null,
+    "bos_token_id": 1,
+    "chunk_size_feed_forward": 0,
+    "classifier_proj_size": 256,
+    "codevector_dim": 768,
+    "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
+    ],
+    "cross_attention_hidden_size": null,
+    "ctc_loss_reduction": "sum",
+    "ctc_zero_infinity": false,
+    "decoder_start_token_id": null,
+    "diversity_loss_weight": 0.1,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "do_stable_layer_norm": true,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "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,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "gradient_checkpointing": false,
+    "hidden_act": "gelu",
+    "hidden_dropout": 0.0,
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 1024,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "is_decoder": false,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "layer_norm_eps": 1e-05,
+    "layerdrop": 0.0,
+    "length_penalty": 1.0,
+    "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": 0.0,
+    "max_length": 20,
+    "min_length": 0,
+    "model_type": "wav2vec2",
+    "no_repeat_ngram_size": 0,
+    "num_adapter_layers": 3,
+    "num_attention_heads": 16,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "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": 24,
+    "num_negatives": 100,
+    "num_return_sequences": 1,
+    "output_attentions": false,
+    "output_hidden_size": 1024,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": 0,
+    "prefix": null,
+    "problem_type": null,
+    "proj_codevector_dim": 768,
+    "pruned_heads": {},
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "sep_token_id": null,
+    "task_specific_params": null,
+    "tdnn_dilation": [
+      1,
+      2,
+      3,
+      1,
+      1
+    ],
+    "tdnn_dim": [
+      512,
+      512,
+      512,
+      512,
+      1500
+    ],
+    "tdnn_kernel": [
+      5,
+      3,
+      3,
+      1,
+      1
+    ],
+    "temperature": 1.0,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": true,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": null,
+    "torchscript": false,
+    "transformers_version": "4.17.0.dev0",
+    "use_bfloat16": false,
+    "use_weighted_layer_sum": false,
+    "vocab_size": 32,
+    "xvector_output_dim": 512
+  },
+  "eos_token_id": 102,
+  "is_encoder_decoder": true,
+  "max_length": 50,
+  "model_type": "speech-encoder-decoder",
+  "pad_token_id": 0,
+  "processor_class": "Wav2Vec2Processor",
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": null,
+  "use_cache": false
+}

create_model.py

@@ -0,0 +1,57 @@
+from transformers import SpeechEncoderDecoderModel, AutoFeatureExtractor, AutoTokenizer, Wav2Vec2Processor, BertConfig, BertLMHeadModel
+import torch
+
+# checkpoints to leverage
+encoder_id = "facebook/wav2vec2-large-lv60"
+decoder_id = "bert-large-uncased"
+
+feature_extractor = AutoFeatureExtractor.from_pretrained(encoder_id)
+feature_extractor.save_pretrained("./")
+tokenizer = AutoTokenizer.from_pretrained(decoder_id)
+tokenizer.save_pretrained("./")
+
+decoder_config = BertConfig.from_pretrained(decoder_id, is_decoder=True)
+decoder_config.num_hidden_layers = 2
+decoder = BertLMHeadModel(decoder_config)
+decoder.save_pretrained("decoder") # save the decoder
+
+model = SpeechEncoderDecoderModel.from_encoder_decoder_pretrained(encoder_id, "decoder", encoder_add_adapter=True)
+
+# verify the decoder
+model.decoder.config.is_decoder
+model.decoder.config.add_cross_attention
+
+# set all encoder regularisation to zero
+model.config.encoder.feat_proj_dropout = 0.0
+model.config.encoder.final_dropout = 0.0
+model.config.encoder.activation_dropout = 0.0
+model.config.encoder.apply_spec_augment = False
+model.config.encoder.attention_dropout = 0.0
+model.config.encoder.feat_extract_dropout = 0.0
+model.config.encoder.feat_proj_dropout = 0.0
+model.config.encoder.hidden_dropout = 0.0
+model.config.encoder.hidden_dropout_prob = 0.0
+model.config.encoder.layerdrop = 0.0
+model.config.encoder.mask_feature_prob = 0.0
+model.config.encoder.mask_time_prob = 0.0
+
+# set all decoder regularisation to zero
+model.config.decoder.attention_probs_dropout_prob = 0.0
+
+# set special token ids
+model.config.decoder_start_token_id = tokenizer.cls_token_id
+model.config.pad_token_id = tokenizer.pad_token_id
+model.config.eos_token_id = tokenizer.sep_token_id
+model.config.max_length = 50
+model.config.num_beams = 1
+model.config.use_cache = False
+model.config.decoder.use_cache = False
+model.config.processor_class = "Wav2Vec2Processor"
+
+# check if generation works
+out = model.generate(torch.ones((1, 2000)))
+
+model.save_pretrained("./")
+
+
+

decoder/config.json

@@ -0,0 +1,26 @@
+{
+  "architectures": [
+    "BertLMHeadModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "is_decoder": true,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 2,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

decoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08e952bd7ce8c58af9176e9ca0d5dbd0000f8cec52144a1c4c1d6f1ec837515e
+size 232250084

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f3a7260c97b2a7fe59e2961a806b5870cd20da57098aabe1b950586c95a4396
+size 1603306797

run_librispeech.sh

@@ -0,0 +1,36 @@
+#!/usr/bin/env bash
+CUDA_VISIBLE_DEVICES=1 python run_speech_recognition_seq2seq.py \
+	--dataset_name="librispeech_asr" \
+	--model_name_or_path="./" \
+	--tokenizer_name="./" \
+	--dataset_config_name="clean" \
+	--train_split_name="train.100" \
+	--eval_split_name="validation" \
+	--output_dir="./" \
+	--preprocessing_num_workers="1" \
+	--length_column_name="input_length" \
+	--overwrite_output_dir \
+	--num_train_epochs="20" \
+	--per_device_train_batch_size="16" \
+	--per_device_eval_batch_size="16" \
+	--gradient_accumulation_steps="8" \
+	--generation_max_length="40" \
+	--generation_num_beams="1" \
+	--learning_rate="3e-5" \
+	--warmup_steps="1000" \
+	--evaluation_strategy="steps" \
+	--text_column_name="text" \
+	--save_steps="1500" \
+	--eval_steps="1500" \
+	--logging_steps="1" \
+	--save_total_limit="1" \
+	--freeze_feature_encoder \
+	--gradient_checkpointing \
+	--fp16 \
+	--group_by_length \
+	--predict_with_generate \
+	--do_lower_case \
+	--do_eval --do_train \
+	--push_to_hub \
+	--use_auth_token 
+

run_speech_recognition_seq2seq.py

@@ -0,0 +1,539 @@
+#!/usr/bin/env python
+# coding=utf-8
+# Copyright 2021 The HuggingFace 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
+# limitations under the License.
+"""
+Fine-tuning the library models for sequence to sequence speech recognition.
+"""
+# You can also adapt this script on your own sequence to sequence speech
+# recognition task. Pointers for this are left as comments.
+
+import logging
+import os
+import sys
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Optional, Union
+
+import datasets
+import torch
+from datasets import DatasetDict, load_dataset, load_metric
+
+import bitsandbytes as bnb
+import transformers
+from transformers import (
+    AutoConfig,
+    AutoFeatureExtractor,
+    AutoModelForSpeechSeq2Seq,
+    AutoProcessor,
+    AutoTokenizer,
+    HfArgumentParser,
+    Seq2SeqTrainer,
+    Seq2SeqTrainingArguments,
+    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
+from transformers.optimization import Adafactor
+
+
+# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
+check_min_version("4.17.0.dev0")
+
+require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/summarization/requirements.txt")
+
+logger = logging.getLogger(__name__)
+
+
+@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"}
+    )
+    config_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
+    )
+    tokenizer_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
+    )
+    feature_extractor_name: Optional[str] = field(
+        default=None, metadata={"help": "feature extractor name or path if not the same as model_name"}
+    )
+    cache_dir: Optional[str] = field(
+        default=None,
+        metadata={"help": "Where to store the pretrained models downloaded from huggingface.co"},
+    )
+    use_fast_tokenizer: bool = field(
+        default=True,
+        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
+    )
+    model_revision: str = field(
+        default="main",
+        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
+    )
+    use_auth_token: bool = field(
+        default=False,
+        metadata={
+            "help": "Will use the token generated when running `transformers-cli login` (necessary to use this script "
+            "with private models)."
+        },
+    )
+    freeze_feature_encoder: bool = field(
+        default=True, metadata={"help": "Whether to freeze the feature encoder layers of the model."}
+    )
+
+
+@dataclass
+class DataTrainingArguments:
+    """
+    Arguments pertaining to what data we are going to input our model for training and eval.
+    """
+
+    dataset_name: str = field(
+        default=None, metadata={"help": "The name of the dataset to use (via the datasets library)."}
+    )
+    dataset_config_name: Optional[str] = field(
+        default=None, metadata={"help": "The configuration name of the dataset to use (via the datasets library)."}
+    )
+    text_column: Optional[str] = field(
+        default=None,
+        metadata={"help": "The name of the column in the datasets containing the full texts (for summarization)."},
+    )
+    overwrite_cache: bool = field(
+        default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
+    )
+    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 evaluation examples to this "
+            "value if set."
+        },
+    )
+    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'"},
+    )
+    max_duration_in_seconds: float = field(
+        default=20.0,
+        metadata={
+            "help": "Truncate 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"
+        },
+    )
+    train_split_name: str = field(
+        default="train",
+        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'"
+        },
+    )
+    do_lower_case: bool = field(
+        default=True,
+        metadata={"help": "Whether the target text should be lower cased."},
+    )
+
+
+@dataclass
+class DataCollatorSpeechSeq2SeqWithPadding:
+    """
+    Data collator that will dynamically pad the inputs received.
+    Args:
+        processor ([`Wav2Vec2Processor`])
+            The processor used for proccessing the data.
+        decoder_start_token_id (`int`)
+            The begin-of-sentence of the decoder.
+    """
+
+    processor: Any
+    decoder_start_token_id: int
+
+    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.feature_extractor.pad(input_features, return_tensors="pt")
+
+        labels_batch = self.processor.tokenizer.pad(label_features, 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)
+
+        # if bos token is appended in previous tokenization step,
+        # cut bos token here as it's append later anyways
+        if (labels[:, 0] == self.decoder_start_token_id).all().cpu().item():
+            labels = labels[:, 1:]
+
+        batch["labels"] = labels
+
+        return batch
+
+
+def main():
+    # 1. Parse input arguments
+    # 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, Seq2SeqTrainingArguments))
+
+    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 = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args, data_args, training_args = parser.parse_args_into_dataclasses()
+
+    # 2. 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)],
+    )
+    log_level = training_args.get_process_log_level()
+    logger.setLevel(log_level)
+    datasets.utils.logging.set_verbosity(log_level)
+    transformers.utils.logging.set_verbosity(log_level)
+    transformers.utils.logging.enable_default_handler()
+    transformers.utils.logging.enable_explicit_format()
+
+    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}"
+    )
+    logger.info(f"Training/evaluation parameters {training_args}")
+
+    # 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)
+
+    # 3. Detecting last checkpoint and eventualy continue from 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 and training_args.resume_from_checkpoint is 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."
+            )
+
+    # Set seed before initializing model.
+    set_seed(training_args.seed)
+
+    # 4. Load 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
+        )
+
+    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
+        )
+
+    if data_args.audio_column_name not in next(iter(raw_datasets.values())).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(next(iter(raw_datasets.values())).column_names)}."
+        )
+
+    if data_args.text_column_name not in next(iter(raw_datasets.values())).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(next(iter(raw_datasets.values())).column_names)}."
+        )
+
+    # 5. Load pretrained model, tokenizer, and feature extractor
+    #
+    # Distributed training:
+    # The .from_pretrained methods guarantee that only one local process can concurrently
+    config = AutoConfig.from_pretrained(
+        model_args.config_name if model_args.config_name else model_args.model_name_or_path,
+        cache_dir=model_args.cache_dir,
+        revision=model_args.model_revision,
+        use_auth_token=True if model_args.use_auth_token else None,
+    )
+
+    feature_extractor = AutoFeatureExtractor.from_pretrained(
+        model_args.feature_extractor_name if model_args.feature_extractor_name else model_args.model_name_or_path,
+        cache_dir=model_args.cache_dir,
+        revision=model_args.model_revision,
+        use_auth_token=True if model_args.use_auth_token else None,
+    )
+    tokenizer = AutoTokenizer.from_pretrained(
+        model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
+        cache_dir=model_args.cache_dir,
+        use_fast=model_args.use_fast_tokenizer,
+        revision=model_args.model_revision,
+        use_auth_token=True if model_args.use_auth_token else None,
+    )
+    model = AutoModelForSpeechSeq2Seq.from_pretrained(
+        model_args.model_name_or_path,
+        config=config,
+        cache_dir=model_args.cache_dir,
+        revision=model_args.model_revision,
+        use_auth_token=True if model_args.use_auth_token else None,
+    )
+
+    if model.config.decoder_start_token_id is None:
+        raise ValueError("Make sure that `config.decoder_start_token_id` is correctly defined")
+
+    if model_args.freeze_feature_encoder:
+        model.freeze_feature_encoder()
+
+    # 6. Resample speech dataset if necassary
+    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)
+        )
+
+    # 7. Preprocessing the datasets.
+    # We need to read the audio files as arrays and tokenize the targets.
+    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
+    text_column_name = data_args.text_column_name
+    model_input_name = feature_extractor.model_input_names[0]
+    do_lower_case = data_args.do_lower_case
+
+    if data_args.max_train_samples is not None:
+        raw_datasets["train"] = raw_datasets["train"].select(range(data_args.max_train_samples))
+
+    if data_args.max_eval_samples is not None:
+        raw_datasets["eval"] = raw_datasets["eval"].select(range(data_args.max_eval_samples))
+
+    def prepare_dataset(batch):
+        # process audio
+        sample = batch[audio_column_name]
+        inputs = feature_extractor(sample["array"], sampling_rate=sample["sampling_rate"])
+        # process audio length
+        batch[model_input_name] = inputs.input_values[0]
+        batch["input_length"] = len(batch["input_values"])
+
+        # process targets
+        input_str = batch[text_column_name].lower() if do_lower_case else batch[text_column_name]
+        batch["labels"] = tokenizer(input_str).input_ids
+        return batch
+
+    with training_args.main_process_first(desc="dataset map pre-processing"):
+        vectorized_datasets = raw_datasets.map(
+            prepare_dataset,
+            remove_columns=next(iter(raw_datasets.values())).column_names,
+            num_proc=data_args.preprocessing_num_workers,
+            desc="preprocess train dataset",
+        )
+
+    # filter data that is shorter than min_input_length or longer than
+    # max_input_length
+    def is_audio_in_length_range(length):
+        return length > min_input_length and length < max_input_length
+
+    vectorized_datasets = vectorized_datasets.filter(
+        is_audio_in_length_range,
+        num_proc=num_workers,
+        input_columns=["input_length"],
+    )
+
+    # 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:
+        cache = {k: v.cache_files for k, v in vectorized_datasets.items()}
+        logger.info(f"Data preprocessing finished. Files cached at {cache}.")
+        return
+
+    # 8. Load Metric
+    metric = load_metric("wer")
+
+    def compute_metrics(pred):
+        pred_ids = pred.predictions
+
+        pred.label_ids[pred.label_ids == -100] = tokenizer.pad_token_id
+
+        pred_str = tokenizer.batch_decode(pred_ids, skip_special_tokens=True)
+        # we do not want to group tokens when computing the metrics
+        label_str = tokenizer.batch_decode(pred.label_ids, skip_special_tokens=True)
+
+        wer = metric.compute(predictions=pred_str, references=label_str)
+
+        return {"wer": wer}
+
+    # 9. Create a single speech processor
+    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)
+
+    processor = AutoProcessor.from_pretrained(training_args.output_dir)
+
+    # 10. Define data collator
+    data_collator = DataCollatorSpeechSeq2SeqWithPadding(
+        processor=processor, decoder_start_token_id=model.config.decoder_start_token_id
+    )
+	
+    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,
+        },
+    ]    
+
+    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,
+    )
+
+    """optimizer = Adafactor(
+        params=optimizer_grouped_parameters,
+        lr=training_args.learning_rate,
+        beta1=training_args.adam_beta1,
+        eps=training_args.adam_epsilon,
+        relative_step=False,
+    )"""
+
+
+    optimizers = (optimizer, None)
+	
+
+    #11. Initialize Trainer
+
+    trainer = Seq2SeqTrainer(
+        model=model,
+        args=training_args,
+        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,
+        data_collator=data_collator,
+        compute_metrics=compute_metrics if training_args.predict_with_generate else None,
+		optimizers=optimizers,
+    )
+
+    # 12. Training
+    if training_args.do_train:
+        checkpoint = None
+        if training_args.resume_from_checkpoint is not None:
+            checkpoint = training_args.resume_from_checkpoint
+        elif last_checkpoint is not None:
+            checkpoint = last_checkpoint
+        train_result = trainer.train(resume_from_checkpoint=checkpoint)
+        trainer.save_model()  # Saves the feature extractor too for easy upload
+
+        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()
+
+    # 13. Evaluation
+    results = {}
+    if training_args.do_eval:
+        logger.info("*** Evaluate ***")
+        metrics = trainer.evaluate(
+            metric_key_prefix="eval", max_length=model.config.max_length, num_beams=model.config.num_beams
+        )
+        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)
+
+    # 14. Write Training Stats
+    kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "speech recognition"}
+    if data_args.dataset_name is not None:
+        kwargs["dataset_tags"] = data_args.dataset_name
+        if data_args.dataset_config_name is not None:
+            kwargs["dataset_args"] = data_args.dataset_config_name
+            kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}"
+        else:
+            kwargs["dataset"] = data_args.dataset_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/Mar08_10-20-58_sanchit--v100/1646734887.1129148/events.out.tfevents.1646734887.sanchit--v100.1584958.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6172cf4ba2b22d9e1131c5eed4704444f81feebe9a63f8298ea4b73d6a67f8d
+size 4973

runs/Mar08_10-20-58_sanchit--v100/events.out.tfevents.1646734887.sanchit--v100.1584958.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6926e685d5f451f5d400822a71d866d2a99aeee6721bcb625391d3edb4d4d81c
+size 9822

runs/Mar08_10-22-25_sanchit--v100/1646734967.8329628/events.out.tfevents.1646734967.sanchit--v100.1586051.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfec89e6027d2720d4bba75b5e9ce57669928d1b53e22d2a02608115a0307378
+size 4973

runs/Mar08_10-22-25_sanchit--v100/events.out.tfevents.1646734967.sanchit--v100.1586051.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdde1fff2be605c39ac9aa8e42b02d1a5c8e489d82994cba36540ec379451225
+size 244337

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "./", "tokenizer_class": "BertTokenizer"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:273c599674abc4f492178d278fe9451f7ef368bc7a4b89d6b6d78ce4cdb3ae3c
+size 3119

wandb/debug-internal.log

@@ -0,0 +1 @@
+run-20220308_102248-3h7e1p8o/logs/debug-internal.log

wandb/debug.log

@@ -0,0 +1 @@
+run-20220308_102248-3h7e1p8o/logs/debug.log

wandb/latest-run

@@ -0,0 +1 @@
+run-20220308_102248-3h7e1p8o

wandb/run-20220308_102127-2todcy4h/files/config.yaml

@@ -0,0 +1,708 @@
+wandb_version: 1
+
+_n_gpu:
+  desc: null
+  value: 1
+_name_or_path:
+  desc: null
+  value: ./
+_wandb:
+  desc: null
+  value:
+    cli_version: 0.12.10
+    framework: huggingface
+    huggingface_version: 4.17.0.dev0
+    is_jupyter_run: false
+    is_kaggle_kernel: false
+    m:
+    - 1: train/global_step
+      6:
+      - 3
+    - 1: train/loss
+      5: 1
+      6:
+      - 1
+    - 1: train/learning_rate
+      5: 1
+      6:
+      - 1
+    - 1: train/epoch
+      5: 1
+      6:
+      - 1
+    python_version: 3.9.5
+    start_time: 1646734887
+    t:
+      1:
+      - 1
+      - 5
+      - 11
+      3:
+      - 13
+      4: 3.9.5
+      5: 0.12.10
+      6: 4.17.0.dev0
+      8:
+      - 5
+adafactor:
+  desc: null
+  value: false
+adam_beta1:
+  desc: null
+  value: 0.9
+adam_beta2:
+  desc: null
+  value: 0.999
+adam_epsilon:
+  desc: null
+  value: 1.0e-08
+add_cross_attention:
+  desc: null
+  value: false
+architectures:
+  desc: null
+  value:
+  - SpeechEncoderDecoderModel
+bad_words_ids:
+  desc: null
+  value: null
+bf16:
+  desc: null
+  value: false
+bf16_full_eval:
+  desc: null
+  value: false
+bos_token_id:
+  desc: null
+  value: null
+chunk_size_feed_forward:
+  desc: null
+  value: 0
+cross_attention_hidden_size:
+  desc: null
+  value: null
+dataloader_drop_last:
+  desc: null
+  value: false
+dataloader_num_workers:
+  desc: null
+  value: 0
+dataloader_pin_memory:
+  desc: null
+  value: true
+ddp_bucket_cap_mb:
+  desc: null
+  value: None
+ddp_find_unused_parameters:
+  desc: null
+  value: None
+debug:
+  desc: null
+  value: '[]'
+decoder:
+  desc: null
+  value:
+    _name_or_path: decoder
+    add_cross_attention: true
+    architectures:
+    - BertLMHeadModel
+    attention_probs_dropout_prob: 0.0
+    bad_words_ids: null
+    bos_token_id: null
+    chunk_size_feed_forward: 0
+    classifier_dropout: null
+    cross_attention_hidden_size: null
+    decoder_start_token_id: null
+    diversity_penalty: 0.0
+    do_sample: false
+    early_stopping: false
+    encoder_no_repeat_ngram_size: 0
+    eos_token_id: null
+    finetuning_task: null
+    forced_bos_token_id: null
+    forced_eos_token_id: null
+    gradient_checkpointing: false
+    hidden_act: gelu
+    hidden_dropout_prob: 0.1
+    hidden_size: 1024
+    id2label:
+      '0': LABEL_0
+      '1': LABEL_1
+    initializer_range: 0.02
+    intermediate_size: 4096
+    is_decoder: true
+    is_encoder_decoder: false
+    label2id:
+      LABEL_0: 0
+      LABEL_1: 1
+    layer_norm_eps: 1.0e-12
+    length_penalty: 1.0
+    max_length: 20
+    max_position_embeddings: 512
+    min_length: 0
+    model_type: bert
+    no_repeat_ngram_size: 0
+    num_attention_heads: 16
+    num_beam_groups: 1
+    num_beams: 1
+    num_hidden_layers: 2
+    num_return_sequences: 1
+    output_attentions: false
+    output_hidden_states: false
+    output_scores: false
+    pad_token_id: 0
+    position_embedding_type: absolute
+    prefix: null
+    problem_type: null
+    pruned_heads: {}
+    remove_invalid_values: false
+    repetition_penalty: 1.0
+    return_dict: true
+    return_dict_in_generate: false
+    sep_token_id: null
+    task_specific_params: null
+    temperature: 1.0
+    tie_encoder_decoder: false
+    tie_word_embeddings: true
+    tokenizer_class: null
+    top_k: 50
+    top_p: 1.0
+    torch_dtype: float32
+    torchscript: false
+    transformers_version: 4.17.0.dev0
+    type_vocab_size: 2
+    use_bfloat16: false
+    use_cache: false
+    vocab_size: 30522
+decoder_start_token_id:
+  desc: null
+  value: 101
+deepspeed:
+  desc: null
+  value: None
+disable_tqdm:
+  desc: null
+  value: false
+diversity_penalty:
+  desc: null
+  value: 0.0
+do_eval:
+  desc: null
+  value: true
+do_predict:
+  desc: null
+  value: false
+do_sample:
+  desc: null
+  value: false
+do_train:
+  desc: null
+  value: true
+early_stopping:
+  desc: null
+  value: false
+encoder:
+  desc: null
+  value:
+    _name_or_path: facebook/wav2vec2-large-lv60
+    activation_dropout: 0.0
+    adapter_kernel_size: 3
+    adapter_stride: 2
+    add_adapter: true
+    add_cross_attention: false
+    apply_spec_augment: false
+    architectures:
+    - Wav2Vec2ForPreTraining
+    attention_dropout: 0.0
+    bad_words_ids: null
+    bos_token_id: 1
+    chunk_size_feed_forward: 0
+    classifier_proj_size: 256
+    codevector_dim: 768
+    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
+    cross_attention_hidden_size: null
+    ctc_loss_reduction: sum
+    ctc_zero_infinity: false
+    decoder_start_token_id: null
+    diversity_loss_weight: 0.1
+    diversity_penalty: 0.0
+    do_sample: false
+    do_stable_layer_norm: true
+    early_stopping: false
+    encoder_no_repeat_ngram_size: 0
+    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
+    finetuning_task: null
+    forced_bos_token_id: null
+    forced_eos_token_id: null
+    gradient_checkpointing: false
+    hidden_act: gelu
+    hidden_dropout: 0.0
+    hidden_dropout_prob: 0.0
+    hidden_size: 1024
+    id2label:
+      '0': LABEL_0
+      '1': LABEL_1
+    initializer_range: 0.02
+    intermediate_size: 4096
+    is_decoder: false
+    is_encoder_decoder: false
+    label2id:
+      LABEL_0: 0
+      LABEL_1: 1
+    layer_norm_eps: 1.0e-05
+    layerdrop: 0.0
+    length_penalty: 1.0
+    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: 0.0
+    max_length: 20
+    min_length: 0
+    model_type: wav2vec2
+    no_repeat_ngram_size: 0
+    num_adapter_layers: 3
+    num_attention_heads: 16
+    num_beam_groups: 1
+    num_beams: 1
+    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: 24
+    num_negatives: 100
+    num_return_sequences: 1
+    output_attentions: false
+    output_hidden_size: 1024
+    output_hidden_states: false
+    output_scores: false
+    pad_token_id: 0
+    prefix: null
+    problem_type: null
+    proj_codevector_dim: 768
+    pruned_heads: {}
+    remove_invalid_values: false
+    repetition_penalty: 1.0
+    return_dict: true
+    return_dict_in_generate: false
+    sep_token_id: null
+    task_specific_params: null
+    tdnn_dilation:
+    - 1
+    - 2
+    - 3
+    - 1
+    - 1
+    tdnn_dim:
+    - 512
+    - 512
+    - 512
+    - 512
+    - 1500
+    tdnn_kernel:
+    - 5
+    - 3
+    - 3
+    - 1
+    - 1
+    temperature: 1.0
+    tie_encoder_decoder: false
+    tie_word_embeddings: true
+    tokenizer_class: null
+    top_k: 50
+    top_p: 1.0
+    torch_dtype: null
+    torchscript: false
+    transformers_version: 4.17.0.dev0
+    use_bfloat16: false
+    use_weighted_layer_sum: false
+    vocab_size: 32
+    xvector_output_dim: 512
+encoder_no_repeat_ngram_size:
+  desc: null
+  value: 0
+eos_token_id:
+  desc: null
+  value: 102
+eval_accumulation_steps:
+  desc: null
+  value: None
+eval_batch_size:
+  desc: null
+  value: 8
+eval_steps:
+  desc: null
+  value: 1500
+evaluation_strategy:
+  desc: null
+  value: steps
+finetuning_task:
+  desc: null
+  value: null
+forced_bos_token_id:
+  desc: null
+  value: null
+forced_eos_token_id:
+  desc: null
+  value: null
+fp16:
+  desc: null
+  value: true
+fp16_backend:
+  desc: null
+  value: auto
+fp16_full_eval:
+  desc: null
+  value: false
+fp16_opt_level:
+  desc: null
+  value: O1
+generation_max_length:
+  desc: null
+  value: 40
+generation_num_beams:
+  desc: null
+  value: 1
+gradient_accumulation_steps:
+  desc: null
+  value: 4
+gradient_checkpointing:
+  desc: null
+  value: true
+greater_is_better:
+  desc: null
+  value: None
+group_by_length:
+  desc: null
+  value: true
+half_precision_backend:
+  desc: null
+  value: amp
+hub_model_id:
+  desc: null
+  value: None
+hub_strategy:
+  desc: null
+  value: every_save
+hub_token:
+  desc: null
+  value: <HUB_TOKEN>
+id2label:
+  desc: null
+  value:
+    '0': LABEL_0
+    '1': LABEL_1
+ignore_data_skip:
+  desc: null
+  value: false
+is_decoder:
+  desc: null
+  value: false
+is_encoder_decoder:
+  desc: null
+  value: true
+label2id:
+  desc: null
+  value:
+    LABEL_0: 0
+    LABEL_1: 1
+label_names:
+  desc: null
+  value: None
+label_smoothing_factor:
+  desc: null
+  value: 0.0
+learning_rate:
+  desc: null
+  value: 3.0e-05
+length_column_name:
+  desc: null
+  value: input_length
+length_penalty:
+  desc: null
+  value: 1.0
+load_best_model_at_end:
+  desc: null
+  value: false
+local_rank:
+  desc: null
+  value: -1
+log_level:
+  desc: null
+  value: -1
+log_level_replica:
+  desc: null
+  value: -1
+log_on_each_node:
+  desc: null
+  value: true
+logging_dir:
+  desc: null
+  value: ./runs/Mar08_10-20-58_sanchit--v100
+logging_first_step:
+  desc: null
+  value: false
+logging_nan_inf_filter:
+  desc: null
+  value: true
+logging_steps:
+  desc: null
+  value: 1
+logging_strategy:
+  desc: null
+  value: steps
+lr_scheduler_type:
+  desc: null
+  value: linear
+max_grad_norm:
+  desc: null
+  value: 1.0
+max_length:
+  desc: null
+  value: 50
+max_steps:
+  desc: null
+  value: -1
+metric_for_best_model:
+  desc: null
+  value: None
+min_length:
+  desc: null
+  value: 0
+model_type:
+  desc: null
+  value: speech-encoder-decoder
+mp_parameters:
+  desc: null
+  value: ''
+no_cuda:
+  desc: null
+  value: false
+no_repeat_ngram_size:
+  desc: null
+  value: 0
+num_beam_groups:
+  desc: null
+  value: 1
+num_beams:
+  desc: null
+  value: 1
+num_return_sequences:
+  desc: null
+  value: 1
+num_train_epochs:
+  desc: null
+  value: 20.0
+optim:
+  desc: null
+  value: adamw_hf
+output_attentions:
+  desc: null
+  value: false
+output_dir:
+  desc: null
+  value: ./
+output_hidden_states:
+  desc: null
+  value: false
+output_scores:
+  desc: null
+  value: false
+overwrite_output_dir:
+  desc: null
+  value: true
+pad_token_id:
+  desc: null
+  value: 0
+past_index:
+  desc: null
+  value: -1
+per_device_eval_batch_size:
+  desc: null
+  value: 8
+per_device_train_batch_size:
+  desc: null
+  value: 8
+per_gpu_eval_batch_size:
+  desc: null
+  value: None
+per_gpu_train_batch_size:
+  desc: null
+  value: None
+predict_with_generate:
+  desc: null
+  value: true
+prediction_loss_only:
+  desc: null
+  value: false
+prefix:
+  desc: null
+  value: null
+problem_type:
+  desc: null
+  value: null
+processor_class:
+  desc: null
+  value: Wav2Vec2Processor
+pruned_heads:
+  desc: null
+  value: {}
+push_to_hub:
+  desc: null
+  value: true
+push_to_hub_model_id:
+  desc: null
+  value: None
+push_to_hub_organization:
+  desc: null
+  value: None
+push_to_hub_token:
+  desc: null
+  value: <PUSH_TO_HUB_TOKEN>
+remove_invalid_values:
+  desc: null
+  value: false
+remove_unused_columns:
+  desc: null
+  value: true
+repetition_penalty:
+  desc: null
+  value: 1.0
+report_to:
+  desc: null
+  value: '[''tensorboard'', ''wandb'']'
+resume_from_checkpoint:
+  desc: null
+  value: None
+return_dict:
+  desc: null
+  value: true
+return_dict_in_generate:
+  desc: null
+  value: false
+run_name:
+  desc: null
+  value: ./
+save_on_each_node:
+  desc: null
+  value: false
+save_steps:
+  desc: null
+  value: 1500
+save_strategy:
+  desc: null
+  value: steps
+save_total_limit:
+  desc: null
+  value: 1
+seed:
+  desc: null
+  value: 42
+sep_token_id:
+  desc: null
+  value: null
+sharded_ddp:
+  desc: null
+  value: '[]'
+skip_memory_metrics:
+  desc: null
+  value: true
+sortish_sampler:
+  desc: null
+  value: false
+task_specific_params:
+  desc: null
+  value: null
+temperature:
+  desc: null
+  value: 1.0
+tf32:
+  desc: null
+  value: None
+tie_encoder_decoder:
+  desc: null
+  value: false
+tie_word_embeddings:
+  desc: null
+  value: false
+tokenizer_class:
+  desc: null
+  value: null
+top_k:
+  desc: null
+  value: 50
+top_p:
+  desc: null
+  value: 1.0
+torch_dtype:
+  desc: null
+  value: torch.float32
+torchscript:
+  desc: null
+  value: false
+tpu_metrics_debug:
+  desc: null
+  value: false
+tpu_num_cores:
+  desc: null
+  value: None
+train_batch_size:
+  desc: null
+  value: 8
+transformers_version:
+  desc: null
+  value: null
+use_bfloat16:
+  desc: null
+  value: false
+use_cache:
+  desc: null
+  value: false
+use_legacy_prediction_loop:
+  desc: null
+  value: false
+warmup_ratio:
+  desc: null
+  value: 0.0
+warmup_steps:
+  desc: null
+  value: 1000
+weight_decay:
+  desc: null
+  value: 0.0
+xpu_backend:
+  desc: null
+  value: None

wandb/run-20220308_102127-2todcy4h/files/output.log

@@ -0,0 +1,30 @@
+
+
+  0%|                                                                                                          | 0/17840 [00:00<?, ?it/s][WARNING|modeling_utils.py:388] 2022-03-08 10:21:30,823 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:32,522 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:34,202 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+{'loss': 10.4942, 'learning_rate': 3.0000000000000004e-08, 'epoch': 0.0}
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:35,849 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+
+  0%|                                                                                               | 1/17840 [00:07<36:27:24,  7.36s/it][WARNING|modeling_utils.py:388] 2022-03-08 10:21:37,685 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:39,326 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:40,923 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+{'loss': 10.5135, 'learning_rate': 6.000000000000001e-08, 'epoch': 0.0}
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:42,502 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+  0%|                                                                                               | 2/17840 [00:13<34:15:27,  6.91s/it][WARNING|modeling_utils.py:388] 2022-03-08 10:21:44,157 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:45,762 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:47,334 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+{'loss': 10.5039, 'learning_rate': 9e-08, 'epoch': 0.0}
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:48,878 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+  0%|                                                                                               | 3/17840 [00:20<33:02:24,  6.67s/it][WARNING|modeling_utils.py:388] 2022-03-08 10:21:50,478 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:52,024 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:53,554 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+{'loss': 10.487, 'learning_rate': 1.2000000000000002e-07, 'epoch': 0.0}
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:55,123 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+  0%|                                                                                               | 4/17840 [00:26<32:12:25,  6.50s/it][WARNING|modeling_utils.py:388] 2022-03-08 10:21:56,778 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:58,365 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:21:59,880 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+{'loss': 10.5398, 'learning_rate': 1.5000000000000002e-07, 'epoch': 0.01}
+[WARNING|modeling_utils.py:388] 2022-03-08 10:22:01,455 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+  0%|                                                                                               | 5/17840 [00:32<31:54:14,  6.44s/it][WARNING|modeling_utils.py:388] 2022-03-08 10:22:03,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[WARNING|modeling_utils.py:388] 2022-03-08 10:22:04,655 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed

wandb/run-20220308_102127-2todcy4h/files/requirements.txt

@@ -0,0 +1,184 @@
+absl-py==1.0.0
+aiohttp==3.8.1
+aiosignal==1.2.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+babel==2.9.1
+backcall==0.2.0
+bitsandbytes-cuda113==0.26.0
+black==22.1.0
+bleach==4.1.0
+cachetools==5.0.0
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.11
+chex==0.1.0
+click==8.0.3
+clldutils==3.10.1
+colorlog==6.6.0
+csvw==1.11.0
+cycler==0.11.0
+datasets==1.18.3
+debugpy==1.5.1
+decorator==5.1.1
+defusedxml==0.7.1
+dill==0.3.4
+dlinfo==1.2.1
+dm-tree==0.1.6
+docker-pycreds==0.4.0
+entrypoints==0.4
+executing==0.8.2
+filelock==3.4.2
+flatbuffers==2.0
+flax==0.4.0
+fonttools==4.29.1
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.27
+google-auth-oauthlib==0.4.6
+google-auth==2.6.0
+grpcio==1.43.0
+huggingface-hub==0.4.0
+hypothesis==6.36.1
+idna==3.3
+importlib-metadata==4.10.1
+ipykernel==6.8.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.5
+isodate==0.6.1
+jax==0.2.28
+jaxlib==0.1.76+cuda11.cudnn82
+jedi==0.18.1
+jinja2==3.0.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-console==6.4.0
+jupyter-core==4.9.1
+jupyter-server==1.13.5
+jupyter==1.0.0
+jupyterlab-pygments==0.1.2
+jupyterlab-server==2.10.3
+jupyterlab-widgets==1.0.2
+jupyterlab==3.2.9
+kiwisolver==1.3.2
+librosa==0.8.1
+llvmlite==0.38.0
+markdown==3.3.6
+markupsafe==2.0.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+msgpack==1.0.3
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nbclassic==0.3.5
+nbclient==0.5.10
+nbconvert==6.4.1
+nbformat==5.1.3
+nest-asyncio==1.5.4
+notebook==6.4.8
+numba==0.55.1
+numpy==1.21.5
+oauthlib==3.2.0
+opt-einsum==3.3.0
+optax==0.1.0
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+parso==0.8.3
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+phonemizer==3.0.1
+pickleshare==0.7.5
+pillow==9.0.0
+pip==22.0.2
+pkg-resources==0.0.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.1
+promise==2.3
+prompt-toolkit==3.0.26
+protobuf==3.19.4
+psutil==5.9.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pyctcdecode==0.3.0
+pygments==2.11.2
+pygtrie==2.4.2
+pyparsing==3.0.7
+pyrsistent==0.18.1
+python-dateutil==2.8.2
+python-levenshtein==0.12.2
+pytz==2021.3
+pyyaml==6.0
+pyzmq==22.3.0
+qtconsole==5.2.2
+qtpy==2.0.1
+regex==2022.1.18
+requests-oauthlib==1.3.1
+requests==2.27.1
+resampy==0.2.2
+rfc3986==2.0.0
+rsa==4.8
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segments==2.2.0
+send2trash==1.8.0
+sentry-sdk==1.5.6
+setuptools==44.1.1
+shortuuid==1.0.8
+six==1.16.0
+smmap==5.0.0
+sniffio==1.2.0
+sortedcontainers==2.4.0
+soundfile==0.10.3.post1
+stack-data==0.1.4
+tabulate==0.8.9
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.1.0
+tokenizers==0.11.4
+tomli==2.0.0
+toolz==0.11.2
+torch==1.10.2+cu113
+torchaudio==0.10.2+cu113
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==3.10.0.2
+uritemplate==4.1.1
+urllib3==1.26.8
+wandb==0.12.10
+wcwidth==0.2.5
+webencodings==0.5.1
+websocket-client==1.2.3
+werkzeug==2.0.2
+wheel==0.37.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220308_102127-2todcy4h/files/wandb-metadata.json

@@ -0,0 +1,59 @@
+{
+    "os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
+    "python": "3.9.5",
+    "heartbeatAt": "2022-03-08T10:21:28.475107",
+    "startedAt": "2022-03-08T10:21:27.352899",
+    "docker": null,
+    "gpu": "Tesla V100-SXM2-16GB",
+    "gpu_count": 2,
+    "cpu_count": 16,
+    "cuda": null,
+    "args": [
+        "--dataset_name=librispeech_asr",
+        "--model_name_or_path=./",
+        "--tokenizer_name=./",
+        "--dataset_config_name=clean",
+        "--train_split_name=train.100",
+        "--eval_split_name=validation",
+        "--output_dir=./",
+        "--preprocessing_num_workers=1",
+        "--length_column_name=input_length",
+        "--overwrite_output_dir",
+        "--num_train_epochs=20",
+        "--per_device_train_batch_size=8",
+        "--per_device_eval_batch_size=8",
+        "--gradient_accumulation_steps=4",
+        "--generation_max_length=40",
+        "--generation_num_beams=1",
+        "--learning_rate=3e-5",
+        "--warmup_steps=1000",
+        "--evaluation_strategy=steps",
+        "--text_column_name=text",
+        "--save_steps=1500",
+        "--eval_steps=1500",
+        "--logging_steps=1",
+        "--save_total_limit=1",
+        "--freeze_feature_encoder",
+        "--gradient_checkpointing",
+        "--fp16",
+        "--group_by_length",
+        "--predict_with_generate",
+        "--do_lower_case",
+        "--do_eval",
+        "--do_train",
+        "--push_to_hub",
+        "--use_auth_token"
+    ],
+    "state": "running",
+    "program": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/run_speech_recognition_seq2seq.py",
+    "codePath": "run_speech_recognition_seq2seq.py",
+    "git": {
+        "remote": "https://huggingface.co/sanchit-gandhi/wav2vec2-2-rnd-2-layer",
+        "commit": "dc19a5f957f0a6e0997490e697e088e8f3ce31ea"
+    },
+    "email": "[email protected]",
+    "root": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer",
+    "host": "sanchit--v100",
+    "username": "sanchit_huggingface_co",
+    "executable": "/home/sanchit_huggingface_co/gcp/bin/python"
+}

wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train/loss": 10.4504, "train/learning_rate": 1.8e-07, "train/epoch": 0.01, "train/global_step": 6, "_runtime": 40, "_timestamp": 1646734927, "_step": 5}

wandb/run-20220308_102127-2todcy4h/logs/debug-internal.log

@@ -0,0 +1,92 @@
+2022-03-08 10:21:28,209 INFO    MainThread:1585399 [internal.py:wandb_internal():89] W&B internal server running at pid: 1585399, started at: 2022-03-08 10:21:28.209146
+2022-03-08 10:21:28,212 DEBUG   HandlerThread:1585399 [handler.py:handle_request():131] handle_request: check_version
+2022-03-08 10:21:28,212 INFO    WriterThread:1585399 [datastore.py:open_for_write():77] open: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/run-2todcy4h.wandb
+2022-03-08 10:21:28,213 DEBUG   SenderThread:1585399 [sender.py:send():235] send: header
+2022-03-08 10:21:28,213 DEBUG   SenderThread:1585399 [sender.py:send_request():249] send_request: check_version
+2022-03-08 10:21:28,283 DEBUG   SenderThread:1585399 [sender.py:send():235] send: run
+2022-03-08 10:21:28,384 INFO    SenderThread:1585399 [dir_watcher.py:__init__():169] watching files in: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files
+2022-03-08 10:21:28,384 INFO    SenderThread:1585399 [sender.py:_start_run_threads():809] run started: 2todcy4h with start time 1646734887
+2022-03-08 10:21:28,384 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:21:28,384 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:21:28,385 DEBUG   HandlerThread:1585399 [handler.py:handle_request():131] handle_request: run_start
+2022-03-08 10:21:28,474 DEBUG   HandlerThread:1585399 [meta.py:__init__():36] meta init
+2022-03-08 10:21:28,475 DEBUG   HandlerThread:1585399 [meta.py:__init__():50] meta init done
+2022-03-08 10:21:28,475 DEBUG   HandlerThread:1585399 [meta.py:probe():210] probe
+2022-03-08 10:21:28,481 DEBUG   HandlerThread:1585399 [meta.py:_setup_git():200] setup git
+2022-03-08 10:21:28,495 DEBUG   HandlerThread:1585399 [meta.py:_setup_git():207] setup git done
+2022-03-08 10:21:28,495 DEBUG   HandlerThread:1585399 [meta.py:_save_pip():54] save pip
+2022-03-08 10:21:28,496 DEBUG   HandlerThread:1585399 [meta.py:_save_pip():68] save pip done
+2022-03-08 10:21:28,496 DEBUG   HandlerThread:1585399 [meta.py:probe():248] probe done
+2022-03-08 10:21:28,575 DEBUG   SenderThread:1585399 [sender.py:send():235] send: files
+2022-03-08 10:21:28,575 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-metadata.json with policy now
+2022-03-08 10:21:28,580 DEBUG   HandlerThread:1585399 [handler.py:handle_request():131] handle_request: stop_status
+2022-03-08 10:21:28,580 DEBUG   SenderThread:1585399 [sender.py:send_request():249] send_request: stop_status
+2022-03-08 10:21:28,614 DEBUG   SenderThread:1585399 [sender.py:send():235] send: config
+2022-03-08 10:21:28,615 DEBUG   SenderThread:1585399 [sender.py:send():235] send: metric
+2022-03-08 10:21:28,615 DEBUG   SenderThread:1585399 [sender.py:send():235] send: metric
+2022-03-08 10:21:28,615 WARNING SenderThread:1585399 [sender.py:send_metric():902] Seen metric with glob (shouldnt happen)
+2022-03-08 10:21:28,836 INFO    Thread-11 :1585399 [upload_job.py:push():137] Uploaded file /tmp/tmpipa_cxqiwandb/1a7m0p29-wandb-metadata.json
+2022-03-08 10:21:29,385 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:21:29,385 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:29,386 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-metadata.json
+2022-03-08 10:21:29,386 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_created():217] file/dir created: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/requirements.txt
+2022-03-08 10:21:31,385 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:33,385 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:35,386 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:35,956 DEBUG   SenderThread:1585399 [sender.py:send():235] send: metric
+2022-03-08 10:21:35,957 DEBUG   SenderThread:1585399 [sender.py:send():235] send: metric
+2022-03-08 10:21:35,957 DEBUG   SenderThread:1585399 [sender.py:send():235] send: metric
+2022-03-08 10:21:35,957 DEBUG   SenderThread:1585399 [sender.py:send():235] send: history
+2022-03-08 10:21:35,957 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:21:35,957 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:21:36,386 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:21:37,387 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:39,387 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:41,388 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:42,560 DEBUG   SenderThread:1585399 [sender.py:send():235] send: history
+2022-03-08 10:21:42,560 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:21:42,561 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:21:43,389 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:21:43,389 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:43,811 DEBUG   HandlerThread:1585399 [handler.py:handle_request():131] handle_request: stop_status
+2022-03-08 10:21:43,812 DEBUG   SenderThread:1585399 [sender.py:send_request():249] send_request: stop_status
+2022-03-08 10:21:45,390 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:47,390 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:49,020 DEBUG   SenderThread:1585399 [sender.py:send():235] send: history
+2022-03-08 10:21:49,021 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:21:49,021 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:21:49,391 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:21:49,391 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:51,392 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:53,392 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:55,180 DEBUG   SenderThread:1585399 [sender.py:send():235] send: history
+2022-03-08 10:21:55,181 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:21:55,181 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:21:55,393 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:21:55,393 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:57,322 DEBUG   SenderThread:1585399 [sender.py:send():235] send: stats
+2022-03-08 10:21:57,394 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:21:58,861 DEBUG   HandlerThread:1585399 [handler.py:handle_request():131] handle_request: stop_status
+2022-03-08 10:21:58,861 DEBUG   SenderThread:1585399 [sender.py:send_request():249] send_request: stop_status
+2022-03-08 10:21:59,394 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/config.yaml
+2022-03-08 10:21:59,395 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:22:01,395 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:22:01,511 DEBUG   SenderThread:1585399 [sender.py:send():235] send: history
+2022-03-08 10:22:01,512 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:22:01,512 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:22:02,395 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:22:03,396 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:22:04,396 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:22:06,397 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:22:07,752 DEBUG   SenderThread:1585399 [sender.py:send():235] send: history
+2022-03-08 10:22:07,752 DEBUG   SenderThread:1585399 [sender.py:send():235] send: summary
+2022-03-08 10:22:07,753 INFO    SenderThread:1585399 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-08 10:22:08,255 WARNING MainThread:1585399 [internal.py:wandb_internal():152] Internal process interrupt: 1
+2022-03-08 10:22:08,397 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/wandb-summary.json
+2022-03-08 10:22:08,397 INFO    Thread-8  :1585399 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/files/output.log
+2022-03-08 10:22:08,403 WARNING MainThread:1585399 [internal.py:wandb_internal():152] Internal process interrupt: 2
+2022-03-08 10:22:08,403 ERROR   MainThread:1585399 [internal.py:wandb_internal():155] Internal process interrupted.
+2022-03-08 10:22:08,753 INFO    WriterThread:1585399 [datastore.py:close():281] close: /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/run-2todcy4h.wandb
+2022-03-08 10:22:08,753 INFO    SenderThread:1585399 [sender.py:finish():1075] shutting down sender
+2022-03-08 10:22:08,753 INFO    SenderThread:1585399 [dir_watcher.py:finish():283] shutting down directory watcher
+2022-03-08 10:22:08,766 INFO    MainThread:1585399 [internal.py:handle_exit():79] Internal process exited

wandb/run-20220308_102127-2todcy4h/logs/debug.log

@@ -0,0 +1,29 @@
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/settings
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_setup.py:_flush():75] Loading settings from environment variables: {}
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq.py', 'program': '/home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/run_speech_recognition_seq2seq.py'}
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/logs/debug.log
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102127-2todcy4h/logs/debug-internal.log
+2022-03-08 10:21:27,354 INFO    MainThread:1584958 [wandb_init.py:init():420] calling init triggers
+2022-03-08 10:21:27,355 INFO    MainThread:1584958 [wandb_init.py:init():425] wandb.init called with sweep_config: {}
+config: {}
+2022-03-08 10:21:27,355 INFO    MainThread:1584958 [wandb_init.py:init():471] starting backend
+2022-03-08 10:21:27,355 INFO    MainThread:1584958 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-03-08 10:21:27,393 INFO    MainThread:1584958 [backend.py:ensure_launched():219] starting backend process...
+2022-03-08 10:21:27,429 INFO    MainThread:1584958 [backend.py:ensure_launched():224] started backend process with pid: 1585399
+2022-03-08 10:21:27,431 INFO    MainThread:1584958 [wandb_init.py:init():480] backend started and connected
+2022-03-08 10:21:27,440 INFO    MainThread:1584958 [wandb_init.py:init():550] updated telemetry
+2022-03-08 10:21:27,524 INFO    MainThread:1584958 [wandb_init.py:init():581] communicating current version
+2022-03-08 10:21:28,282 INFO    MainThread:1584958 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.11 is available!  To upgrade, please run:\n $ pip install wandb --upgrade"
+
+2022-03-08 10:21:28,282 INFO    MainThread:1584958 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
+2022-03-08 10:21:28,384 INFO    MainThread:1584958 [wandb_init.py:init():624] starting run threads in backend
+2022-03-08 10:21:28,580 INFO    MainThread:1584958 [wandb_run.py:_console_start():1827] atexit reg
+2022-03-08 10:21:28,580 INFO    MainThread:1584958 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
+2022-03-08 10:21:28,581 INFO    MainThread:1584958 [wandb_run.py:_redirect():1706] Redirecting console.
+2022-03-08 10:21:28,582 INFO    MainThread:1584958 [wandb_run.py:_redirect():1762] Redirects installed.
+2022-03-08 10:21:28,582 INFO    MainThread:1584958 [wandb_init.py:init():651] run started, returning control to user process
+2022-03-08 10:21:28,584 INFO    MainThread:1584958 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['SpeechEncoderDecoderModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 0, 'eos_token_id': 102, 'sep_token_id': None, 'decoder_start_token_id': 101, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['BertLMHeadModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 0, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'decoder', 'transformers_version': '4.17.0.dev0', 'gradient_checkpointing': False, 'vocab_size': 30522, 'hidden_size': 1024, 'num_hidden_layers': 2, 'num_attention_heads': 16, 'hidden_act': 'gelu', 'intermediate_size': 4096, 'hidden_dropout_prob': 0.1, 'attention_probs_dropout_prob': 0.0, 'max_position_embeddings': 512, 'type_vocab_size': 2, 'initializer_range': 0.02, 'layer_norm_eps': 1e-12, 'position_embedding_type': 'absolute', 'use_cache': False, 'classifier_dropout': None, 'model_type': 'bert'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 0, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.0, 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.0, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'sum', 'ctc_zero_infinity': False, 'add_adapter': True, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 8, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': 'None', 'learning_rate': 3e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 20.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 1000, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar08_10-20-58_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1500, 'save_total_limit': 1, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'eval_steps': 1500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'input_length', 'report_to': "['tensorboard', 'wandb']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'gradient_checkpointing': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 8, 'eval_batch_size': 8}
+2022-03-08 10:21:28,587 INFO    MainThread:1584958 [wandb_watch.py:watch():43] Watching
+2022-03-08 10:22:08,362 INFO    MainThread:1584958 [wandb_run.py:_atexit_cleanup():1797] got exitcode: 255
+2022-03-08 10:22:08,363 INFO    MainThread:1584958 [wandb_run.py:_restore():1769] restore

wandb/run-20220308_102248-3h7e1p8o/files/requirements.txt

@@ -0,0 +1,184 @@
+absl-py==1.0.0
+aiohttp==3.8.1
+aiosignal==1.2.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+babel==2.9.1
+backcall==0.2.0
+bitsandbytes-cuda113==0.26.0
+black==22.1.0
+bleach==4.1.0
+cachetools==5.0.0
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.11
+chex==0.1.0
+click==8.0.3
+clldutils==3.10.1
+colorlog==6.6.0
+csvw==1.11.0
+cycler==0.11.0
+datasets==1.18.3
+debugpy==1.5.1
+decorator==5.1.1
+defusedxml==0.7.1
+dill==0.3.4
+dlinfo==1.2.1
+dm-tree==0.1.6
+docker-pycreds==0.4.0
+entrypoints==0.4
+executing==0.8.2
+filelock==3.4.2
+flatbuffers==2.0
+flax==0.4.0
+fonttools==4.29.1
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.27
+google-auth-oauthlib==0.4.6
+google-auth==2.6.0
+grpcio==1.43.0
+huggingface-hub==0.4.0
+hypothesis==6.36.1
+idna==3.3
+importlib-metadata==4.10.1
+ipykernel==6.8.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.5
+isodate==0.6.1
+jax==0.2.28
+jaxlib==0.1.76+cuda11.cudnn82
+jedi==0.18.1
+jinja2==3.0.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-console==6.4.0
+jupyter-core==4.9.1
+jupyter-server==1.13.5
+jupyter==1.0.0
+jupyterlab-pygments==0.1.2
+jupyterlab-server==2.10.3
+jupyterlab-widgets==1.0.2
+jupyterlab==3.2.9
+kiwisolver==1.3.2
+librosa==0.8.1
+llvmlite==0.38.0
+markdown==3.3.6
+markupsafe==2.0.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+msgpack==1.0.3
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nbclassic==0.3.5
+nbclient==0.5.10
+nbconvert==6.4.1
+nbformat==5.1.3
+nest-asyncio==1.5.4
+notebook==6.4.8
+numba==0.55.1
+numpy==1.21.5
+oauthlib==3.2.0
+opt-einsum==3.3.0
+optax==0.1.0
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+parso==0.8.3
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+phonemizer==3.0.1
+pickleshare==0.7.5
+pillow==9.0.0
+pip==22.0.2
+pkg-resources==0.0.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.1
+promise==2.3
+prompt-toolkit==3.0.26
+protobuf==3.19.4
+psutil==5.9.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pyctcdecode==0.3.0
+pygments==2.11.2
+pygtrie==2.4.2
+pyparsing==3.0.7
+pyrsistent==0.18.1
+python-dateutil==2.8.2
+python-levenshtein==0.12.2
+pytz==2021.3
+pyyaml==6.0
+pyzmq==22.3.0
+qtconsole==5.2.2
+qtpy==2.0.1
+regex==2022.1.18
+requests-oauthlib==1.3.1
+requests==2.27.1
+resampy==0.2.2
+rfc3986==2.0.0
+rsa==4.8
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segments==2.2.0
+send2trash==1.8.0
+sentry-sdk==1.5.6
+setuptools==44.1.1
+shortuuid==1.0.8
+six==1.16.0
+smmap==5.0.0
+sniffio==1.2.0
+sortedcontainers==2.4.0
+soundfile==0.10.3.post1
+stack-data==0.1.4
+tabulate==0.8.9
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.1.0
+tokenizers==0.11.4
+tomli==2.0.0
+toolz==0.11.2
+torch==1.10.2+cu113
+torchaudio==0.10.2+cu113
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==3.10.0.2
+uritemplate==4.1.1
+urllib3==1.26.8
+wandb==0.12.10
+wcwidth==0.2.5
+webencodings==0.5.1
+websocket-client==1.2.3
+werkzeug==2.0.2
+wheel==0.37.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220308_102248-3h7e1p8o/files/wandb-metadata.json

@@ -0,0 +1,59 @@
+{
+    "os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
+    "python": "3.9.5",
+    "heartbeatAt": "2022-03-08T10:22:49.070528",
+    "startedAt": "2022-03-08T10:22:48.070919",
+    "docker": null,
+    "gpu": "Tesla V100-SXM2-16GB",
+    "gpu_count": 2,
+    "cpu_count": 16,
+    "cuda": null,
+    "args": [
+        "--dataset_name=librispeech_asr",
+        "--model_name_or_path=./",
+        "--tokenizer_name=./",
+        "--dataset_config_name=clean",
+        "--train_split_name=train.100",
+        "--eval_split_name=validation",
+        "--output_dir=./",
+        "--preprocessing_num_workers=1",
+        "--length_column_name=input_length",
+        "--overwrite_output_dir",
+        "--num_train_epochs=20",
+        "--per_device_train_batch_size=16",
+        "--per_device_eval_batch_size=16",
+        "--gradient_accumulation_steps=8",
+        "--generation_max_length=40",
+        "--generation_num_beams=1",
+        "--learning_rate=3e-5",
+        "--warmup_steps=1000",
+        "--evaluation_strategy=steps",
+        "--text_column_name=text",
+        "--save_steps=1500",
+        "--eval_steps=1500",
+        "--logging_steps=1",
+        "--save_total_limit=1",
+        "--freeze_feature_encoder",
+        "--gradient_checkpointing",
+        "--fp16",
+        "--group_by_length",
+        "--predict_with_generate",
+        "--do_lower_case",
+        "--do_eval",
+        "--do_train",
+        "--push_to_hub",
+        "--use_auth_token"
+    ],
+    "state": "running",
+    "program": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/run_speech_recognition_seq2seq.py",
+    "codePath": "run_speech_recognition_seq2seq.py",
+    "git": {
+        "remote": "https://huggingface.co/sanchit-gandhi/wav2vec2-2-rnd-2-layer",
+        "commit": "dc19a5f957f0a6e0997490e697e088e8f3ce31ea"
+    },
+    "email": "[email protected]",
+    "root": "/home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer",
+    "host": "sanchit--v100",
+    "username": "sanchit_huggingface_co",
+    "executable": "/home/sanchit_huggingface_co/gcp/bin/python"
+}

wandb/run-20220308_102248-3h7e1p8o/logs/debug.log

@@ -0,0 +1,27 @@
+2022-03-08 10:22:48,072 INFO    MainThread:1586051 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
+2022-03-08 10:22:48,072 INFO    MainThread:1586051 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/settings
+2022-03-08 10:22:48,072 INFO    MainThread:1586051 [wandb_setup.py:_flush():75] Loading settings from environment variables: {}
+2022-03-08 10:22:48,072 INFO    MainThread:1586051 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq.py', 'program': '/home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/run_speech_recognition_seq2seq.py'}
+2022-03-08 10:22:48,072 INFO    MainThread:1586051 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102248-3h7e1p8o/logs/debug.log
+2022-03-08 10:22:48,072 INFO    MainThread:1586051 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/wav2vec2-2-rnd-2-layer/wandb/run-20220308_102248-3h7e1p8o/logs/debug-internal.log
+2022-03-08 10:22:48,073 INFO    MainThread:1586051 [wandb_init.py:init():420] calling init triggers
+2022-03-08 10:22:48,073 INFO    MainThread:1586051 [wandb_init.py:init():425] wandb.init called with sweep_config: {}
+config: {}
+2022-03-08 10:22:48,073 INFO    MainThread:1586051 [wandb_init.py:init():471] starting backend
+2022-03-08 10:22:48,073 INFO    MainThread:1586051 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-03-08 10:22:48,108 INFO    MainThread:1586051 [backend.py:ensure_launched():219] starting backend process...
+2022-03-08 10:22:48,142 INFO    MainThread:1586051 [backend.py:ensure_launched():224] started backend process with pid: 1586410
+2022-03-08 10:22:48,144 INFO    MainThread:1586051 [wandb_init.py:init():480] backend started and connected
+2022-03-08 10:22:48,153 INFO    MainThread:1586051 [wandb_init.py:init():550] updated telemetry
+2022-03-08 10:22:48,236 INFO    MainThread:1586051 [wandb_init.py:init():581] communicating current version
+2022-03-08 10:22:48,932 INFO    MainThread:1586051 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.11 is available!  To upgrade, please run:\n $ pip install wandb --upgrade"
+
+2022-03-08 10:22:48,932 INFO    MainThread:1586051 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
+2022-03-08 10:22:49,066 INFO    MainThread:1586051 [wandb_init.py:init():624] starting run threads in backend
+2022-03-08 10:22:49,165 INFO    MainThread:1586051 [wandb_run.py:_console_start():1827] atexit reg
+2022-03-08 10:22:49,165 INFO    MainThread:1586051 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
+2022-03-08 10:22:49,166 INFO    MainThread:1586051 [wandb_run.py:_redirect():1706] Redirecting console.
+2022-03-08 10:22:49,167 INFO    MainThread:1586051 [wandb_run.py:_redirect():1762] Redirects installed.
+2022-03-08 10:22:49,167 INFO    MainThread:1586051 [wandb_init.py:init():651] run started, returning control to user process
+2022-03-08 10:22:49,170 INFO    MainThread:1586051 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['SpeechEncoderDecoderModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 0, 'eos_token_id': 102, 'sep_token_id': None, 'decoder_start_token_id': 101, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['BertLMHeadModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 0, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'decoder', 'transformers_version': '4.17.0.dev0', 'gradient_checkpointing': False, 'vocab_size': 30522, 'hidden_size': 1024, 'num_hidden_layers': 2, 'num_attention_heads': 16, 'hidden_act': 'gelu', 'intermediate_size': 4096, 'hidden_dropout_prob': 0.1, 'attention_probs_dropout_prob': 0.0, 'max_position_embeddings': 512, 'type_vocab_size': 2, 'initializer_range': 0.02, 'layer_norm_eps': 1e-12, 'position_embedding_type': 'absolute', 'use_cache': False, 'classifier_dropout': None, 'model_type': 'bert'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 0, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.0, 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.0, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'sum', 'ctc_zero_infinity': False, 'add_adapter': True, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 16, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': 'None', 'learning_rate': 3e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 20.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 1000, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar08_10-22-25_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1500, 'save_total_limit': 1, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'eval_steps': 1500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'input_length', 'report_to': "['tensorboard', 'wandb']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'gradient_checkpointing': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 16, 'eval_batch_size': 16}
+2022-03-08 10:22:49,172 INFO    MainThread:1586051 [wandb_watch.py:watch():43] Watching

wandb/run-20220308_102248-3h7e1p8o/run-3h7e1p8o.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42cfbbba05fc1729a2b265f966535d7f92d0b6e55b0316fb0ceddd61158947db
+size 103631861