end this training
Browse files- .ipynb_checkpoints/README-checkpoint.md +77 -0
- .ipynb_checkpoints/added_tokens-checkpoint.json +1 -0
- .ipynb_checkpoints/config-checkpoint.json +4 -4
- .ipynb_checkpoints/eval-checkpoint.py +2 -2
- .ipynb_checkpoints/eval_results-checkpoint.json +0 -9
- .ipynb_checkpoints/log_speech-recognition-community-v2_dev_data_fr_validation_predictions-checkpoint.txt +0 -0
- .ipynb_checkpoints/log_speech-recognition-community-v2_dev_data_fr_validation_targets-checkpoint.txt +0 -0
- .ipynb_checkpoints/mozilla-foundation_common_voice_8_0_fr_test_eval_results-checkpoint.txt +2 -0
- .ipynb_checkpoints/run-checkpoint.sh +6 -5
- .ipynb_checkpoints/run_speech_recognition_ctc-checkpoint.py +4 -2
- .ipynb_checkpoints/special_tokens_map-checkpoint.json +1 -0
- .ipynb_checkpoints/speech-recognition-community-v2_dev_data_fr_validation_eval_results-checkpoint.txt +2 -0
- .ipynb_checkpoints/tokenizer_config-checkpoint.json +1 -0
- .ipynb_checkpoints/train_results-checkpoint.json +0 -8
- .ipynb_checkpoints/trainer_state-checkpoint.json +0 -1096
- .ipynb_checkpoints/validate-checkpoint.sh +3 -0
- README.md +25 -3
- added_tokens.json +1 -1
- config.json +1 -1
- eval.py +2 -2
- eval_results.json +0 -8
- log_mozilla-foundation_common_voice_8_0_fr_test_predictions.txt +0 -0
- log_mozilla-foundation_common_voice_8_0_fr_test_targets.txt +0 -0
- log_speech-recognition-community-v2_dev_data_fr_validation_predictions.txt +0 -0
- log_speech-recognition-community-v2_dev_data_fr_validation_targets.txt +0 -0
- mozilla-foundation_common_voice_8_0_fr_test_eval_results.txt +2 -0
- run.sh +6 -5
- run_speech_recognition_ctc.py +4 -2
- special_tokens_map.json +1 -1
- speech-recognition-community-v2_dev_data_fr_validation_eval_results.txt +2 -0
- tokenizer_config.json +1 -1
- train_results.json +0 -8
- trainer_state.json +0 -1096
- validate.sh +3 -0
- wandb/run-20220201_142447-3hqii6h6/files/config.yaml +6 -0
- wandb/run-20220201_142447-3hqii6h6/files/output.log +7 -0
- wandb/run-20220201_142447-3hqii6h6/files/wandb-summary.json +1 -1
- wandb/run-20220201_142447-3hqii6h6/logs/debug-internal.log +113 -0
- wandb/run-20220201_142447-3hqii6h6/logs/debug.log +106 -0
- wandb/run-20220201_142447-3hqii6h6/run-3hqii6h6.wandb +0 -0
.ipynb_checkpoints/README-checkpoint.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- fr
|
| 4 |
+
tags:
|
| 5 |
+
- automatic-speech-recognition
|
| 6 |
+
- robust-speech-event
|
| 7 |
+
- mozilla-foundation/common_voice_8_0
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- common_voice
|
| 11 |
+
model-index:
|
| 12 |
+
- name: name: xls-r-300m-fr
|
| 13 |
+
results:
|
| 14 |
+
-task:
|
| 15 |
+
name: Speech Recognition
|
| 16 |
+
type: automatic-speech-recognition
|
| 17 |
+
dataset:
|
| 18 |
+
name: Common Voice 8.0 fr
|
| 19 |
+
type: mozilla-foundation/common_voice_8_0
|
| 20 |
+
args: fr
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Test WER
|
| 23 |
+
type: wer
|
| 24 |
+
value: 21.58
|
| 25 |
+
-task:
|
| 26 |
+
name: Speech Recognition
|
| 27 |
+
type: automatic-speech-recognition
|
| 28 |
+
dataset:
|
| 29 |
+
name: dev_data
|
| 30 |
+
type: speech-recognition-community-v2/dev_data
|
| 31 |
+
args: fr
|
| 32 |
+
metrics:
|
| 33 |
+
- name: Validation WER
|
| 34 |
+
type: wer
|
| 35 |
+
value: 36.03
|
| 36 |
+
---
|
| 37 |
+
|
| 38 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 39 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 40 |
+
|
| 41 |
+
#
|
| 42 |
+
|
| 43 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - FR dataset.
|
| 44 |
+
|
| 45 |
+
## Model description
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Intended uses & limitations
|
| 50 |
+
|
| 51 |
+
More information needed
|
| 52 |
+
|
| 53 |
+
## Training and evaluation data
|
| 54 |
+
|
| 55 |
+
More information needed
|
| 56 |
+
|
| 57 |
+
## Training procedure
|
| 58 |
+
|
| 59 |
+
### Training hyperparameters
|
| 60 |
+
|
| 61 |
+
The following hyperparameters were used during training:
|
| 62 |
+
- learning_rate: 0.0001
|
| 63 |
+
- train_batch_size: 64
|
| 64 |
+
- eval_batch_size: 64
|
| 65 |
+
- seed: 42
|
| 66 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 67 |
+
- lr_scheduler_type: linear
|
| 68 |
+
- lr_scheduler_warmup_steps: 2700
|
| 69 |
+
- num_epochs: 1.0
|
| 70 |
+
- mixed_precision_training: Native AMP
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- Transformers 4.17.0.dev0
|
| 75 |
+
- Pytorch 1.10.2+cu102
|
| 76 |
+
- Datasets 1.18.2.dev0
|
| 77 |
+
- Tokenizers 0.11.0
|
.ipynb_checkpoints/added_tokens-checkpoint.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{}
|
.ipynb_checkpoints/config-checkpoint.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "facebook/wav2vec2-xls-r-300m",
|
| 3 |
-
"activation_dropout": 0.
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
| 6 |
"add_adapter": false,
|
|
@@ -61,7 +61,7 @@
|
|
| 61 |
"layerdrop": 0.0,
|
| 62 |
"mask_feature_length": 10,
|
| 63 |
"mask_feature_min_masks": 0,
|
| 64 |
-
"mask_feature_prob": 0.
|
| 65 |
"mask_time_length": 10,
|
| 66 |
"mask_time_min_masks": 2,
|
| 67 |
"mask_time_prob": 0.75,
|
|
@@ -76,7 +76,7 @@
|
|
| 76 |
"num_hidden_layers": 24,
|
| 77 |
"num_negatives": 100,
|
| 78 |
"output_hidden_size": 1024,
|
| 79 |
-
"pad_token_id":
|
| 80 |
"proj_codevector_dim": 768,
|
| 81 |
"tdnn_dilation": [
|
| 82 |
1,
|
|
@@ -102,6 +102,6 @@
|
|
| 102 |
"torch_dtype": "float32",
|
| 103 |
"transformers_version": "4.17.0.dev0",
|
| 104 |
"use_weighted_layer_sum": false,
|
| 105 |
-
"vocab_size":
|
| 106 |
"xvector_output_dim": 512
|
| 107 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "facebook/wav2vec2-xls-r-300m",
|
| 3 |
+
"activation_dropout": 0.05,
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
| 6 |
"add_adapter": false,
|
|
|
|
| 61 |
"layerdrop": 0.0,
|
| 62 |
"mask_feature_length": 10,
|
| 63 |
"mask_feature_min_masks": 0,
|
| 64 |
+
"mask_feature_prob": 0.25,
|
| 65 |
"mask_time_length": 10,
|
| 66 |
"mask_time_min_masks": 2,
|
| 67 |
"mask_time_prob": 0.75,
|
|
|
|
| 76 |
"num_hidden_layers": 24,
|
| 77 |
"num_negatives": 100,
|
| 78 |
"output_hidden_size": 1024,
|
| 79 |
+
"pad_token_id": 218,
|
| 80 |
"proj_codevector_dim": 768,
|
| 81 |
"tdnn_dilation": [
|
| 82 |
1,
|
|
|
|
| 102 |
"torch_dtype": "float32",
|
| 103 |
"transformers_version": "4.17.0.dev0",
|
| 104 |
"use_weighted_layer_sum": false,
|
| 105 |
+
"vocab_size": 219,
|
| 106 |
"xvector_output_dim": 512
|
| 107 |
}
|
.ipynb_checkpoints/eval-checkpoint.py
CHANGED
|
@@ -85,7 +85,7 @@ def main(args):
|
|
| 85 |
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
|
| 86 |
|
| 87 |
# for testing: only process the first two examples as a test
|
| 88 |
-
dataset = dataset.select(range(20))
|
| 89 |
|
| 90 |
# load processor
|
| 91 |
feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
|
|
@@ -105,7 +105,7 @@ def main(args):
|
|
| 105 |
batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
|
| 106 |
)
|
| 107 |
|
| 108 |
-
batch["prediction"] =
|
| 109 |
batch["target"] = normalize_text(batch["sentence"])
|
| 110 |
return batch
|
| 111 |
|
|
|
|
| 85 |
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
|
| 86 |
|
| 87 |
# for testing: only process the first two examples as a test
|
| 88 |
+
#dataset = dataset.select(range(20))
|
| 89 |
|
| 90 |
# load processor
|
| 91 |
feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
|
|
|
|
| 105 |
batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
|
| 106 |
)
|
| 107 |
|
| 108 |
+
batch["prediction"] = "".join(prediction["text"].split("<s>"))
|
| 109 |
batch["target"] = normalize_text(batch["sentence"])
|
| 110 |
return batch
|
| 111 |
|
.ipynb_checkpoints/eval_results-checkpoint.json
DELETED
|
@@ -1,9 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"epoch": 2.0,
|
| 3 |
-
"eval_loss": 0.23875188827514648,
|
| 4 |
-
"eval_runtime": 294.1776,
|
| 5 |
-
"eval_samples": 5792,
|
| 6 |
-
"eval_samples_per_second": 19.689,
|
| 7 |
-
"eval_steps_per_second": 0.309,
|
| 8 |
-
"eval_wer": 0.3680797679950471
|
| 9 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
.ipynb_checkpoints/log_speech-recognition-community-v2_dev_data_fr_validation_predictions-checkpoint.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
.ipynb_checkpoints/log_speech-recognition-community-v2_dev_data_fr_validation_targets-checkpoint.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
.ipynb_checkpoints/mozilla-foundation_common_voice_8_0_fr_test_eval_results-checkpoint.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
WER: 0.21587470509795875
|
| 2 |
+
CER: 0.06356032070032196
|
.ipynb_checkpoints/run-checkpoint.sh
CHANGED
|
@@ -1,17 +1,17 @@
|
|
| 1 |
WANDB_PROJECT=auto-speech-recognition-french
|
| 2 |
python run_speech_recognition_ctc.py \
|
| 3 |
--dataset_name="mozilla-foundation/common_voice_8_0" \
|
| 4 |
-
--model_name_or_path="./checkpoint-18000" \
|
| 5 |
--dataset_config_name="fr" \
|
| 6 |
--tokenizer_name_or_path="./" \
|
| 7 |
--output_dir="./" \
|
| 8 |
--overwrite_output_dir \
|
| 9 |
-
--num_train_epochs="
|
| 10 |
--per_device_train_batch_size="64" \
|
| 11 |
--per_device_eval_batch_size="64" \
|
| 12 |
--gradient_accumulation_steps="1" \
|
| 13 |
--learning_rate="1e-4" \
|
| 14 |
-
--warmup_steps="
|
| 15 |
--length_column_name="input_length" \
|
| 16 |
--evaluation_strategy="steps" \
|
| 17 |
--text_column_name="sentence" \
|
|
@@ -23,18 +23,19 @@ python run_speech_recognition_ctc.py \
|
|
| 23 |
--save_total_limit="2" \
|
| 24 |
--freeze_feature_encoder \
|
| 25 |
--feat_proj_dropout="0.0" \
|
| 26 |
-
--mask_time_prob="0.
|
| 27 |
--mask_time_length="10" \
|
| 28 |
--mask_feature_prob="0.25" \
|
| 29 |
--mask_feature_length="10" \
|
| 30 |
--gradient_checkpointing \
|
| 31 |
--report_to="wandb" \
|
| 32 |
--run_name="xls-r-300m-fr" \
|
|
|
|
| 33 |
--max_duration_in_seconds="10" \
|
| 34 |
--use_auth_token \
|
| 35 |
--fp16 \
|
| 36 |
--group_by_length \
|
| 37 |
--preprocessing_num_workers="64" \
|
| 38 |
-
--do_train
|
| 39 |
--load_best_model_at_end \
|
| 40 |
--push_to_hub
|
|
|
|
| 1 |
WANDB_PROJECT=auto-speech-recognition-french
|
| 2 |
python run_speech_recognition_ctc.py \
|
| 3 |
--dataset_name="mozilla-foundation/common_voice_8_0" \
|
| 4 |
+
--model_name_or_path="./checkpoint-18000-keep" \
|
| 5 |
--dataset_config_name="fr" \
|
| 6 |
--tokenizer_name_or_path="./" \
|
| 7 |
--output_dir="./" \
|
| 8 |
--overwrite_output_dir \
|
| 9 |
+
--num_train_epochs="4" \
|
| 10 |
--per_device_train_batch_size="64" \
|
| 11 |
--per_device_eval_batch_size="64" \
|
| 12 |
--gradient_accumulation_steps="1" \
|
| 13 |
--learning_rate="1e-4" \
|
| 14 |
+
--warmup_steps="500" \
|
| 15 |
--length_column_name="input_length" \
|
| 16 |
--evaluation_strategy="steps" \
|
| 17 |
--text_column_name="sentence" \
|
|
|
|
| 23 |
--save_total_limit="2" \
|
| 24 |
--freeze_feature_encoder \
|
| 25 |
--feat_proj_dropout="0.0" \
|
| 26 |
+
--mask_time_prob="0.6" \
|
| 27 |
--mask_time_length="10" \
|
| 28 |
--mask_feature_prob="0.25" \
|
| 29 |
--mask_feature_length="10" \
|
| 30 |
--gradient_checkpointing \
|
| 31 |
--report_to="wandb" \
|
| 32 |
--run_name="xls-r-300m-fr" \
|
| 33 |
+
--max_eval_samples="4500" \
|
| 34 |
--max_duration_in_seconds="10" \
|
| 35 |
--use_auth_token \
|
| 36 |
--fp16 \
|
| 37 |
--group_by_length \
|
| 38 |
--preprocessing_num_workers="64" \
|
| 39 |
+
--do_train --do_eval \
|
| 40 |
--load_best_model_at_end \
|
| 41 |
--push_to_hub
|
.ipynb_checkpoints/run_speech_recognition_ctc-checkpoint.py
CHANGED
|
@@ -402,6 +402,7 @@ def main():
|
|
| 402 |
data_args.dataset_config_name,
|
| 403 |
split=data_args.train_split_name,
|
| 404 |
use_auth_token=data_args.use_auth_token,
|
|
|
|
| 405 |
)
|
| 406 |
|
| 407 |
if data_args.audio_column_name not in raw_datasets["train"].column_names:
|
|
@@ -427,6 +428,7 @@ def main():
|
|
| 427 |
data_args.dataset_config_name,
|
| 428 |
split=data_args.eval_split_name,
|
| 429 |
use_auth_token=data_args.use_auth_token,
|
|
|
|
| 430 |
)
|
| 431 |
|
| 432 |
if data_args.max_eval_samples is not None:
|
|
@@ -607,7 +609,7 @@ def main():
|
|
| 607 |
prepare_dataset,
|
| 608 |
remove_columns=next(iter(raw_datasets.values())).column_names,
|
| 609 |
batch_size=-1,
|
| 610 |
-
desc="preprocess datasets",
|
| 611 |
)
|
| 612 |
|
| 613 |
def is_audio_in_length_range(length):
|
|
@@ -686,7 +688,7 @@ def main():
|
|
| 686 |
# 8. Finally, we can start training
|
| 687 |
|
| 688 |
# Training
|
| 689 |
-
if training_args.do_train
|
| 690 |
|
| 691 |
# use last checkpoint if exist
|
| 692 |
if last_checkpoint is not None:
|
|
|
|
| 402 |
data_args.dataset_config_name,
|
| 403 |
split=data_args.train_split_name,
|
| 404 |
use_auth_token=data_args.use_auth_token,
|
| 405 |
+
cache_dir="../downloaded_data/"
|
| 406 |
)
|
| 407 |
|
| 408 |
if data_args.audio_column_name not in raw_datasets["train"].column_names:
|
|
|
|
| 428 |
data_args.dataset_config_name,
|
| 429 |
split=data_args.eval_split_name,
|
| 430 |
use_auth_token=data_args.use_auth_token,
|
| 431 |
+
cache_dir="../downloaded_data/"
|
| 432 |
)
|
| 433 |
|
| 434 |
if data_args.max_eval_samples is not None:
|
|
|
|
| 609 |
prepare_dataset,
|
| 610 |
remove_columns=next(iter(raw_datasets.values())).column_names,
|
| 611 |
batch_size=-1,
|
| 612 |
+
desc="preprocess datasets",cache_dir="../data_cache/"
|
| 613 |
)
|
| 614 |
|
| 615 |
def is_audio_in_length_range(length):
|
|
|
|
| 688 |
# 8. Finally, we can start training
|
| 689 |
|
| 690 |
# Training
|
| 691 |
+
if training_args.do_train:
|
| 692 |
|
| 693 |
# use last checkpoint if exist
|
| 694 |
if last_checkpoint is not None:
|
.ipynb_checkpoints/special_tokens_map-checkpoint.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": null, "eos_token": null, "unk_token": "<unk>", "pad_token": "<pad>"}
|
.ipynb_checkpoints/speech-recognition-community-v2_dev_data_fr_validation_eval_results-checkpoint.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
WER: 0.3603687452909631
|
| 2 |
+
CER: 0.14097275038988377
|
.ipynb_checkpoints/tokenizer_config-checkpoint.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": null, "eos_token": null, "pad_token": "<pad>", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "padding": true, "truncation": true, "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
.ipynb_checkpoints/train_results-checkpoint.json
DELETED
|
@@ -1,8 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"epoch": 2.0,
|
| 3 |
-
"train_loss": 1.442369053426242,
|
| 4 |
-
"train_runtime": 53680.5392,
|
| 5 |
-
"train_samples": 442265,
|
| 6 |
-
"train_samples_per_second": 16.478,
|
| 7 |
-
"train_steps_per_second": 0.257
|
| 8 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
.ipynb_checkpoints/trainer_state-checkpoint.json
DELETED
|
@@ -1,1096 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": 0.23875188827514648,
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-13500",
|
| 4 |
-
"epoch": 2.0,
|
| 5 |
-
"global_step": 13822,
|
| 6 |
-
"is_hyper_param_search": false,
|
| 7 |
-
"is_local_process_zero": true,
|
| 8 |
-
"is_world_process_zero": true,
|
| 9 |
-
"log_history": [
|
| 10 |
-
{
|
| 11 |
-
"epoch": 0.01,
|
| 12 |
-
"learning_rate": 6.533333333333333e-06,
|
| 13 |
-
"loss": 17.2403,
|
| 14 |
-
"step": 100
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 0.03,
|
| 18 |
-
"learning_rate": 1.32e-05,
|
| 19 |
-
"loss": 10.2311,
|
| 20 |
-
"step": 200
|
| 21 |
-
},
|
| 22 |
-
{
|
| 23 |
-
"epoch": 0.04,
|
| 24 |
-
"learning_rate": 1.9800000000000004e-05,
|
| 25 |
-
"loss": 7.834,
|
| 26 |
-
"step": 300
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 0.06,
|
| 30 |
-
"learning_rate": 2.646666666666667e-05,
|
| 31 |
-
"loss": 6.0656,
|
| 32 |
-
"step": 400
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 0.07,
|
| 36 |
-
"learning_rate": 3.313333333333333e-05,
|
| 37 |
-
"loss": 4.3748,
|
| 38 |
-
"step": 500
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.07,
|
| 42 |
-
"eval_loss": 3.878422975540161,
|
| 43 |
-
"eval_runtime": 285.8223,
|
| 44 |
-
"eval_samples_per_second": 20.264,
|
| 45 |
-
"eval_steps_per_second": 0.318,
|
| 46 |
-
"eval_wer": 1.0,
|
| 47 |
-
"step": 500
|
| 48 |
-
},
|
| 49 |
-
{
|
| 50 |
-
"epoch": 0.09,
|
| 51 |
-
"learning_rate": 3.9800000000000005e-05,
|
| 52 |
-
"loss": 3.2923,
|
| 53 |
-
"step": 600
|
| 54 |
-
},
|
| 55 |
-
{
|
| 56 |
-
"epoch": 0.1,
|
| 57 |
-
"learning_rate": 4.646666666666667e-05,
|
| 58 |
-
"loss": 2.9475,
|
| 59 |
-
"step": 700
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.12,
|
| 63 |
-
"learning_rate": 5.3133333333333335e-05,
|
| 64 |
-
"loss": 2.8639,
|
| 65 |
-
"step": 800
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.13,
|
| 69 |
-
"learning_rate": 5.9800000000000003e-05,
|
| 70 |
-
"loss": 2.8265,
|
| 71 |
-
"step": 900
|
| 72 |
-
},
|
| 73 |
-
{
|
| 74 |
-
"epoch": 0.14,
|
| 75 |
-
"learning_rate": 6.646666666666667e-05,
|
| 76 |
-
"loss": 2.8068,
|
| 77 |
-
"step": 1000
|
| 78 |
-
},
|
| 79 |
-
{
|
| 80 |
-
"epoch": 0.14,
|
| 81 |
-
"eval_loss": 2.828850746154785,
|
| 82 |
-
"eval_runtime": 292.3877,
|
| 83 |
-
"eval_samples_per_second": 19.809,
|
| 84 |
-
"eval_steps_per_second": 0.311,
|
| 85 |
-
"eval_wer": 0.9826485059793412,
|
| 86 |
-
"step": 1000
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.16,
|
| 90 |
-
"learning_rate": 7.306666666666668e-05,
|
| 91 |
-
"loss": 2.779,
|
| 92 |
-
"step": 1100
|
| 93 |
-
},
|
| 94 |
-
{
|
| 95 |
-
"epoch": 0.17,
|
| 96 |
-
"learning_rate": 7.973333333333334e-05,
|
| 97 |
-
"loss": 2.6402,
|
| 98 |
-
"step": 1200
|
| 99 |
-
},
|
| 100 |
-
{
|
| 101 |
-
"epoch": 0.19,
|
| 102 |
-
"learning_rate": 8.64e-05,
|
| 103 |
-
"loss": 2.1119,
|
| 104 |
-
"step": 1300
|
| 105 |
-
},
|
| 106 |
-
{
|
| 107 |
-
"epoch": 0.2,
|
| 108 |
-
"learning_rate": 9.306666666666667e-05,
|
| 109 |
-
"loss": 1.7965,
|
| 110 |
-
"step": 1400
|
| 111 |
-
},
|
| 112 |
-
{
|
| 113 |
-
"epoch": 0.22,
|
| 114 |
-
"learning_rate": 9.973333333333334e-05,
|
| 115 |
-
"loss": 1.6698,
|
| 116 |
-
"step": 1500
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 0.22,
|
| 120 |
-
"eval_loss": 0.881136417388916,
|
| 121 |
-
"eval_runtime": 297.1806,
|
| 122 |
-
"eval_samples_per_second": 19.49,
|
| 123 |
-
"eval_steps_per_second": 0.306,
|
| 124 |
-
"eval_wer": 0.7127472384241911,
|
| 125 |
-
"step": 1500
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 0.23,
|
| 129 |
-
"learning_rate": 9.92209056971271e-05,
|
| 130 |
-
"loss": 1.5882,
|
| 131 |
-
"step": 1600
|
| 132 |
-
},
|
| 133 |
-
{
|
| 134 |
-
"epoch": 0.25,
|
| 135 |
-
"learning_rate": 9.840934913163448e-05,
|
| 136 |
-
"loss": 1.5172,
|
| 137 |
-
"step": 1700
|
| 138 |
-
},
|
| 139 |
-
{
|
| 140 |
-
"epoch": 0.26,
|
| 141 |
-
"learning_rate": 9.759779256614186e-05,
|
| 142 |
-
"loss": 1.4579,
|
| 143 |
-
"step": 1800
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.27,
|
| 147 |
-
"learning_rate": 9.678623600064926e-05,
|
| 148 |
-
"loss": 1.3829,
|
| 149 |
-
"step": 1900
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.29,
|
| 153 |
-
"learning_rate": 9.597467943515663e-05,
|
| 154 |
-
"loss": 1.3488,
|
| 155 |
-
"step": 2000
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 0.29,
|
| 159 |
-
"eval_loss": 0.516592800617218,
|
| 160 |
-
"eval_runtime": 301.2842,
|
| 161 |
-
"eval_samples_per_second": 19.224,
|
| 162 |
-
"eval_steps_per_second": 0.302,
|
| 163 |
-
"eval_wer": 0.5369024731988661,
|
| 164 |
-
"step": 2000
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 0.3,
|
| 168 |
-
"learning_rate": 9.516312286966402e-05,
|
| 169 |
-
"loss": 1.2981,
|
| 170 |
-
"step": 2100
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 0.32,
|
| 174 |
-
"learning_rate": 9.43515663041714e-05,
|
| 175 |
-
"loss": 1.2845,
|
| 176 |
-
"step": 2200
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 0.33,
|
| 180 |
-
"learning_rate": 9.354000973867879e-05,
|
| 181 |
-
"loss": 1.2459,
|
| 182 |
-
"step": 2300
|
| 183 |
-
},
|
| 184 |
-
{
|
| 185 |
-
"epoch": 0.35,
|
| 186 |
-
"learning_rate": 9.272845317318618e-05,
|
| 187 |
-
"loss": 1.2255,
|
| 188 |
-
"step": 2400
|
| 189 |
-
},
|
| 190 |
-
{
|
| 191 |
-
"epoch": 0.36,
|
| 192 |
-
"learning_rate": 9.191689660769356e-05,
|
| 193 |
-
"loss": 1.2239,
|
| 194 |
-
"step": 2500
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 0.36,
|
| 198 |
-
"eval_loss": 0.4104757010936737,
|
| 199 |
-
"eval_runtime": 299.1395,
|
| 200 |
-
"eval_samples_per_second": 19.362,
|
| 201 |
-
"eval_steps_per_second": 0.304,
|
| 202 |
-
"eval_wer": 0.474111245071524,
|
| 203 |
-
"step": 2500
|
| 204 |
-
},
|
| 205 |
-
{
|
| 206 |
-
"epoch": 0.38,
|
| 207 |
-
"learning_rate": 9.110534004220094e-05,
|
| 208 |
-
"loss": 1.2024,
|
| 209 |
-
"step": 2600
|
| 210 |
-
},
|
| 211 |
-
{
|
| 212 |
-
"epoch": 0.39,
|
| 213 |
-
"learning_rate": 9.030189904236326e-05,
|
| 214 |
-
"loss": 1.1851,
|
| 215 |
-
"step": 2700
|
| 216 |
-
},
|
| 217 |
-
{
|
| 218 |
-
"epoch": 0.41,
|
| 219 |
-
"learning_rate": 8.949034247687063e-05,
|
| 220 |
-
"loss": 1.1768,
|
| 221 |
-
"step": 2800
|
| 222 |
-
},
|
| 223 |
-
{
|
| 224 |
-
"epoch": 0.42,
|
| 225 |
-
"learning_rate": 8.867878591137803e-05,
|
| 226 |
-
"loss": 1.1641,
|
| 227 |
-
"step": 2900
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 0.43,
|
| 231 |
-
"learning_rate": 8.786722934588541e-05,
|
| 232 |
-
"loss": 1.1537,
|
| 233 |
-
"step": 3000
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 0.43,
|
| 237 |
-
"eval_loss": 0.35850802063941956,
|
| 238 |
-
"eval_runtime": 299.662,
|
| 239 |
-
"eval_samples_per_second": 19.328,
|
| 240 |
-
"eval_steps_per_second": 0.304,
|
| 241 |
-
"eval_wer": 0.4448499462348073,
|
| 242 |
-
"step": 3000
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 0.45,
|
| 246 |
-
"learning_rate": 8.70556727803928e-05,
|
| 247 |
-
"loss": 1.1449,
|
| 248 |
-
"step": 3100
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 0.46,
|
| 252 |
-
"learning_rate": 8.624411621490018e-05,
|
| 253 |
-
"loss": 1.1379,
|
| 254 |
-
"step": 3200
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 0.48,
|
| 258 |
-
"learning_rate": 8.543255964940758e-05,
|
| 259 |
-
"loss": 1.1331,
|
| 260 |
-
"step": 3300
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 0.49,
|
| 264 |
-
"learning_rate": 8.462100308391495e-05,
|
| 265 |
-
"loss": 1.1205,
|
| 266 |
-
"step": 3400
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 0.51,
|
| 270 |
-
"learning_rate": 8.380944651842234e-05,
|
| 271 |
-
"loss": 1.1184,
|
| 272 |
-
"step": 3500
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 0.51,
|
| 276 |
-
"eval_loss": 0.333638072013855,
|
| 277 |
-
"eval_runtime": 297.0402,
|
| 278 |
-
"eval_samples_per_second": 19.499,
|
| 279 |
-
"eval_steps_per_second": 0.306,
|
| 280 |
-
"eval_wer": 0.42922545537489004,
|
| 281 |
-
"step": 3500
|
| 282 |
-
},
|
| 283 |
-
{
|
| 284 |
-
"epoch": 0.52,
|
| 285 |
-
"learning_rate": 8.299788995292971e-05,
|
| 286 |
-
"loss": 1.1014,
|
| 287 |
-
"step": 3600
|
| 288 |
-
},
|
| 289 |
-
{
|
| 290 |
-
"epoch": 0.54,
|
| 291 |
-
"learning_rate": 8.218633338743711e-05,
|
| 292 |
-
"loss": 1.1114,
|
| 293 |
-
"step": 3700
|
| 294 |
-
},
|
| 295 |
-
{
|
| 296 |
-
"epoch": 0.55,
|
| 297 |
-
"learning_rate": 8.13747768219445e-05,
|
| 298 |
-
"loss": 1.117,
|
| 299 |
-
"step": 3800
|
| 300 |
-
},
|
| 301 |
-
{
|
| 302 |
-
"epoch": 0.56,
|
| 303 |
-
"learning_rate": 8.056322025645188e-05,
|
| 304 |
-
"loss": 1.102,
|
| 305 |
-
"step": 3900
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 0.58,
|
| 309 |
-
"learning_rate": 7.975166369095926e-05,
|
| 310 |
-
"loss": 1.0968,
|
| 311 |
-
"step": 4000
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 0.58,
|
| 315 |
-
"eval_loss": 0.31949570775032043,
|
| 316 |
-
"eval_runtime": 296.0172,
|
| 317 |
-
"eval_samples_per_second": 19.566,
|
| 318 |
-
"eval_steps_per_second": 0.307,
|
| 319 |
-
"eval_wer": 0.4180162273127179,
|
| 320 |
-
"step": 4000
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 0.59,
|
| 324 |
-
"learning_rate": 7.894822269112158e-05,
|
| 325 |
-
"loss": 1.0942,
|
| 326 |
-
"step": 4100
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 0.61,
|
| 330 |
-
"learning_rate": 7.813666612562897e-05,
|
| 331 |
-
"loss": 1.0859,
|
| 332 |
-
"step": 4200
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 0.62,
|
| 336 |
-
"learning_rate": 7.732510956013635e-05,
|
| 337 |
-
"loss": 1.0767,
|
| 338 |
-
"step": 4300
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"epoch": 0.64,
|
| 342 |
-
"learning_rate": 7.652166856029866e-05,
|
| 343 |
-
"loss": 1.0766,
|
| 344 |
-
"step": 4400
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 0.65,
|
| 348 |
-
"learning_rate": 7.571011199480604e-05,
|
| 349 |
-
"loss": 1.0737,
|
| 350 |
-
"step": 4500
|
| 351 |
-
},
|
| 352 |
-
{
|
| 353 |
-
"epoch": 0.65,
|
| 354 |
-
"eval_loss": 0.30754634737968445,
|
| 355 |
-
"eval_runtime": 296.2378,
|
| 356 |
-
"eval_samples_per_second": 19.552,
|
| 357 |
-
"eval_steps_per_second": 0.307,
|
| 358 |
-
"eval_wer": 0.41408973899442797,
|
| 359 |
-
"step": 4500
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 0.67,
|
| 363 |
-
"learning_rate": 7.489855542931342e-05,
|
| 364 |
-
"loss": 1.0807,
|
| 365 |
-
"step": 4600
|
| 366 |
-
},
|
| 367 |
-
{
|
| 368 |
-
"epoch": 0.68,
|
| 369 |
-
"learning_rate": 7.40869988638208e-05,
|
| 370 |
-
"loss": 1.071,
|
| 371 |
-
"step": 4700
|
| 372 |
-
},
|
| 373 |
-
{
|
| 374 |
-
"epoch": 0.69,
|
| 375 |
-
"learning_rate": 7.32754422983282e-05,
|
| 376 |
-
"loss": 1.0613,
|
| 377 |
-
"step": 4800
|
| 378 |
-
},
|
| 379 |
-
{
|
| 380 |
-
"epoch": 0.71,
|
| 381 |
-
"learning_rate": 7.246388573283557e-05,
|
| 382 |
-
"loss": 1.0635,
|
| 383 |
-
"step": 4900
|
| 384 |
-
},
|
| 385 |
-
{
|
| 386 |
-
"epoch": 0.72,
|
| 387 |
-
"learning_rate": 7.165232916734297e-05,
|
| 388 |
-
"loss": 1.0677,
|
| 389 |
-
"step": 5000
|
| 390 |
-
},
|
| 391 |
-
{
|
| 392 |
-
"epoch": 0.72,
|
| 393 |
-
"eval_loss": 0.30150118470191956,
|
| 394 |
-
"eval_runtime": 297.4742,
|
| 395 |
-
"eval_samples_per_second": 19.471,
|
| 396 |
-
"eval_steps_per_second": 0.306,
|
| 397 |
-
"eval_wer": 0.4089250219948516,
|
| 398 |
-
"step": 5000
|
| 399 |
-
},
|
| 400 |
-
{
|
| 401 |
-
"epoch": 0.74,
|
| 402 |
-
"learning_rate": 7.084077260185034e-05,
|
| 403 |
-
"loss": 1.0707,
|
| 404 |
-
"step": 5100
|
| 405 |
-
},
|
| 406 |
-
{
|
| 407 |
-
"epoch": 0.75,
|
| 408 |
-
"learning_rate": 7.002921603635774e-05,
|
| 409 |
-
"loss": 1.0617,
|
| 410 |
-
"step": 5200
|
| 411 |
-
},
|
| 412 |
-
{
|
| 413 |
-
"epoch": 0.77,
|
| 414 |
-
"learning_rate": 6.921765947086512e-05,
|
| 415 |
-
"loss": 1.0566,
|
| 416 |
-
"step": 5300
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 0.78,
|
| 420 |
-
"learning_rate": 6.84061029053725e-05,
|
| 421 |
-
"loss": 1.0518,
|
| 422 |
-
"step": 5400
|
| 423 |
-
},
|
| 424 |
-
{
|
| 425 |
-
"epoch": 0.8,
|
| 426 |
-
"learning_rate": 6.760266190553481e-05,
|
| 427 |
-
"loss": 1.0462,
|
| 428 |
-
"step": 5500
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 0.8,
|
| 432 |
-
"eval_loss": 0.297052800655365,
|
| 433 |
-
"eval_runtime": 296.1592,
|
| 434 |
-
"eval_samples_per_second": 19.557,
|
| 435 |
-
"eval_steps_per_second": 0.307,
|
| 436 |
-
"eval_wer": 0.4077193782788621,
|
| 437 |
-
"step": 5500
|
| 438 |
-
},
|
| 439 |
-
{
|
| 440 |
-
"epoch": 0.81,
|
| 441 |
-
"learning_rate": 6.679110534004221e-05,
|
| 442 |
-
"loss": 1.0514,
|
| 443 |
-
"step": 5600
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 0.82,
|
| 447 |
-
"learning_rate": 6.597954877454959e-05,
|
| 448 |
-
"loss": 1.0446,
|
| 449 |
-
"step": 5700
|
| 450 |
-
},
|
| 451 |
-
{
|
| 452 |
-
"epoch": 0.84,
|
| 453 |
-
"learning_rate": 6.516799220905698e-05,
|
| 454 |
-
"loss": 1.0358,
|
| 455 |
-
"step": 5800
|
| 456 |
-
},
|
| 457 |
-
{
|
| 458 |
-
"epoch": 0.85,
|
| 459 |
-
"learning_rate": 6.435643564356436e-05,
|
| 460 |
-
"loss": 1.0364,
|
| 461 |
-
"step": 5900
|
| 462 |
-
},
|
| 463 |
-
{
|
| 464 |
-
"epoch": 0.87,
|
| 465 |
-
"learning_rate": 6.354487907807174e-05,
|
| 466 |
-
"loss": 1.0392,
|
| 467 |
-
"step": 6000
|
| 468 |
-
},
|
| 469 |
-
{
|
| 470 |
-
"epoch": 0.87,
|
| 471 |
-
"eval_loss": 0.2870033383369446,
|
| 472 |
-
"eval_runtime": 295.9814,
|
| 473 |
-
"eval_samples_per_second": 19.569,
|
| 474 |
-
"eval_steps_per_second": 0.307,
|
| 475 |
-
"eval_wer": 0.3997034768157972,
|
| 476 |
-
"step": 6000
|
| 477 |
-
},
|
| 478 |
-
{
|
| 479 |
-
"epoch": 0.88,
|
| 480 |
-
"learning_rate": 6.273332251257913e-05,
|
| 481 |
-
"loss": 1.0375,
|
| 482 |
-
"step": 6100
|
| 483 |
-
},
|
| 484 |
-
{
|
| 485 |
-
"epoch": 0.9,
|
| 486 |
-
"learning_rate": 6.192176594708652e-05,
|
| 487 |
-
"loss": 1.0408,
|
| 488 |
-
"step": 6200
|
| 489 |
-
},
|
| 490 |
-
{
|
| 491 |
-
"epoch": 0.91,
|
| 492 |
-
"learning_rate": 6.11102093815939e-05,
|
| 493 |
-
"loss": 1.0382,
|
| 494 |
-
"step": 6300
|
| 495 |
-
},
|
| 496 |
-
{
|
| 497 |
-
"epoch": 0.93,
|
| 498 |
-
"learning_rate": 6.0298652816101284e-05,
|
| 499 |
-
"loss": 1.0335,
|
| 500 |
-
"step": 6400
|
| 501 |
-
},
|
| 502 |
-
{
|
| 503 |
-
"epoch": 0.94,
|
| 504 |
-
"learning_rate": 5.948709625060867e-05,
|
| 505 |
-
"loss": 1.0178,
|
| 506 |
-
"step": 6500
|
| 507 |
-
},
|
| 508 |
-
{
|
| 509 |
-
"epoch": 0.94,
|
| 510 |
-
"eval_loss": 0.28046590089797974,
|
| 511 |
-
"eval_runtime": 297.8045,
|
| 512 |
-
"eval_samples_per_second": 19.449,
|
| 513 |
-
"eval_steps_per_second": 0.306,
|
| 514 |
-
"eval_wer": 0.39629834794225943,
|
| 515 |
-
"step": 6500
|
| 516 |
-
},
|
| 517 |
-
{
|
| 518 |
-
"epoch": 0.95,
|
| 519 |
-
"learning_rate": 5.867553968511605e-05,
|
| 520 |
-
"loss": 1.0312,
|
| 521 |
-
"step": 6600
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"epoch": 0.97,
|
| 525 |
-
"learning_rate": 5.786398311962344e-05,
|
| 526 |
-
"loss": 1.033,
|
| 527 |
-
"step": 6700
|
| 528 |
-
},
|
| 529 |
-
{
|
| 530 |
-
"epoch": 0.98,
|
| 531 |
-
"learning_rate": 5.7052426554130825e-05,
|
| 532 |
-
"loss": 1.0289,
|
| 533 |
-
"step": 6800
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 1.0,
|
| 537 |
-
"learning_rate": 5.624086998863821e-05,
|
| 538 |
-
"loss": 1.027,
|
| 539 |
-
"step": 6900
|
| 540 |
-
},
|
| 541 |
-
{
|
| 542 |
-
"epoch": 1.01,
|
| 543 |
-
"learning_rate": 5.54293134231456e-05,
|
| 544 |
-
"loss": 0.992,
|
| 545 |
-
"step": 7000
|
| 546 |
-
},
|
| 547 |
-
{
|
| 548 |
-
"epoch": 1.01,
|
| 549 |
-
"eval_loss": 0.2747785747051239,
|
| 550 |
-
"eval_runtime": 298.4841,
|
| 551 |
-
"eval_samples_per_second": 19.405,
|
| 552 |
-
"eval_steps_per_second": 0.305,
|
| 553 |
-
"eval_wer": 0.39352862589201343,
|
| 554 |
-
"step": 7000
|
| 555 |
-
},
|
| 556 |
-
{
|
| 557 |
-
"epoch": 1.03,
|
| 558 |
-
"learning_rate": 5.461775685765298e-05,
|
| 559 |
-
"loss": 1.0025,
|
| 560 |
-
"step": 7100
|
| 561 |
-
},
|
| 562 |
-
{
|
| 563 |
-
"epoch": 1.04,
|
| 564 |
-
"learning_rate": 5.3806200292160366e-05,
|
| 565 |
-
"loss": 1.0122,
|
| 566 |
-
"step": 7200
|
| 567 |
-
},
|
| 568 |
-
{
|
| 569 |
-
"epoch": 1.06,
|
| 570 |
-
"learning_rate": 5.299464372666775e-05,
|
| 571 |
-
"loss": 1.018,
|
| 572 |
-
"step": 7300
|
| 573 |
-
},
|
| 574 |
-
{
|
| 575 |
-
"epoch": 1.07,
|
| 576 |
-
"learning_rate": 5.218308716117514e-05,
|
| 577 |
-
"loss": 0.9936,
|
| 578 |
-
"step": 7400
|
| 579 |
-
},
|
| 580 |
-
{
|
| 581 |
-
"epoch": 1.09,
|
| 582 |
-
"learning_rate": 5.137153059568252e-05,
|
| 583 |
-
"loss": 1.0197,
|
| 584 |
-
"step": 7500
|
| 585 |
-
},
|
| 586 |
-
{
|
| 587 |
-
"epoch": 1.09,
|
| 588 |
-
"eval_loss": 0.26907604932785034,
|
| 589 |
-
"eval_runtime": 298.796,
|
| 590 |
-
"eval_samples_per_second": 19.384,
|
| 591 |
-
"eval_steps_per_second": 0.305,
|
| 592 |
-
"eval_wer": 0.3884453713056796,
|
| 593 |
-
"step": 7500
|
| 594 |
-
},
|
| 595 |
-
{
|
| 596 |
-
"epoch": 1.1,
|
| 597 |
-
"learning_rate": 5.055997403018991e-05,
|
| 598 |
-
"loss": 1.008,
|
| 599 |
-
"step": 7600
|
| 600 |
-
},
|
| 601 |
-
{
|
| 602 |
-
"epoch": 1.11,
|
| 603 |
-
"learning_rate": 4.97484174646973e-05,
|
| 604 |
-
"loss": 1.0028,
|
| 605 |
-
"step": 7700
|
| 606 |
-
},
|
| 607 |
-
{
|
| 608 |
-
"epoch": 1.13,
|
| 609 |
-
"learning_rate": 4.893686089920468e-05,
|
| 610 |
-
"loss": 0.9929,
|
| 611 |
-
"step": 7800
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 1.14,
|
| 615 |
-
"learning_rate": 4.8125304333712064e-05,
|
| 616 |
-
"loss": 0.995,
|
| 617 |
-
"step": 7900
|
| 618 |
-
},
|
| 619 |
-
{
|
| 620 |
-
"epoch": 1.16,
|
| 621 |
-
"learning_rate": 4.731374776821945e-05,
|
| 622 |
-
"loss": 1.0056,
|
| 623 |
-
"step": 8000
|
| 624 |
-
},
|
| 625 |
-
{
|
| 626 |
-
"epoch": 1.16,
|
| 627 |
-
"eval_loss": 0.26817116141319275,
|
| 628 |
-
"eval_runtime": 298.1504,
|
| 629 |
-
"eval_samples_per_second": 19.426,
|
| 630 |
-
"eval_steps_per_second": 0.305,
|
| 631 |
-
"eval_wer": 0.3888526833718922,
|
| 632 |
-
"step": 8000
|
| 633 |
-
},
|
| 634 |
-
{
|
| 635 |
-
"epoch": 1.17,
|
| 636 |
-
"learning_rate": 4.6510306768381754e-05,
|
| 637 |
-
"loss": 0.9971,
|
| 638 |
-
"step": 8100
|
| 639 |
-
},
|
| 640 |
-
{
|
| 641 |
-
"epoch": 1.19,
|
| 642 |
-
"learning_rate": 4.5698750202889145e-05,
|
| 643 |
-
"loss": 0.9976,
|
| 644 |
-
"step": 8200
|
| 645 |
-
},
|
| 646 |
-
{
|
| 647 |
-
"epoch": 1.2,
|
| 648 |
-
"learning_rate": 4.488719363739653e-05,
|
| 649 |
-
"loss": 1.0014,
|
| 650 |
-
"step": 8300
|
| 651 |
-
},
|
| 652 |
-
{
|
| 653 |
-
"epoch": 1.22,
|
| 654 |
-
"learning_rate": 4.407563707190391e-05,
|
| 655 |
-
"loss": 0.9835,
|
| 656 |
-
"step": 8400
|
| 657 |
-
},
|
| 658 |
-
{
|
| 659 |
-
"epoch": 1.23,
|
| 660 |
-
"learning_rate": 4.3272196072066225e-05,
|
| 661 |
-
"loss": 0.9826,
|
| 662 |
-
"step": 8500
|
| 663 |
-
},
|
| 664 |
-
{
|
| 665 |
-
"epoch": 1.23,
|
| 666 |
-
"eval_loss": 0.26473307609558105,
|
| 667 |
-
"eval_runtime": 299.0161,
|
| 668 |
-
"eval_samples_per_second": 19.37,
|
| 669 |
-
"eval_steps_per_second": 0.304,
|
| 670 |
-
"eval_wer": 0.38675095311023494,
|
| 671 |
-
"step": 8500
|
| 672 |
-
},
|
| 673 |
-
{
|
| 674 |
-
"epoch": 1.24,
|
| 675 |
-
"learning_rate": 4.2460639506573615e-05,
|
| 676 |
-
"loss": 0.9838,
|
| 677 |
-
"step": 8600
|
| 678 |
-
},
|
| 679 |
-
{
|
| 680 |
-
"epoch": 1.26,
|
| 681 |
-
"learning_rate": 4.1649082941081e-05,
|
| 682 |
-
"loss": 0.9836,
|
| 683 |
-
"step": 8700
|
| 684 |
-
},
|
| 685 |
-
{
|
| 686 |
-
"epoch": 1.27,
|
| 687 |
-
"learning_rate": 4.0845641941243305e-05,
|
| 688 |
-
"loss": 0.9824,
|
| 689 |
-
"step": 8800
|
| 690 |
-
},
|
| 691 |
-
{
|
| 692 |
-
"epoch": 1.29,
|
| 693 |
-
"learning_rate": 4.003408537575069e-05,
|
| 694 |
-
"loss": 0.9715,
|
| 695 |
-
"step": 8900
|
| 696 |
-
},
|
| 697 |
-
{
|
| 698 |
-
"epoch": 1.3,
|
| 699 |
-
"learning_rate": 3.922252881025807e-05,
|
| 700 |
-
"loss": 0.9815,
|
| 701 |
-
"step": 9000
|
| 702 |
-
},
|
| 703 |
-
{
|
| 704 |
-
"epoch": 1.3,
|
| 705 |
-
"eval_loss": 0.26034271717071533,
|
| 706 |
-
"eval_runtime": 299.6782,
|
| 707 |
-
"eval_samples_per_second": 19.327,
|
| 708 |
-
"eval_steps_per_second": 0.304,
|
| 709 |
-
"eval_wer": 0.3831828994102121,
|
| 710 |
-
"step": 9000
|
| 711 |
-
},
|
| 712 |
-
{
|
| 713 |
-
"epoch": 1.32,
|
| 714 |
-
"learning_rate": 3.841097224476546e-05,
|
| 715 |
-
"loss": 0.9757,
|
| 716 |
-
"step": 9100
|
| 717 |
-
},
|
| 718 |
-
{
|
| 719 |
-
"epoch": 1.33,
|
| 720 |
-
"learning_rate": 3.7599415679272846e-05,
|
| 721 |
-
"loss": 0.9689,
|
| 722 |
-
"step": 9200
|
| 723 |
-
},
|
| 724 |
-
{
|
| 725 |
-
"epoch": 1.35,
|
| 726 |
-
"learning_rate": 3.678785911378023e-05,
|
| 727 |
-
"loss": 0.9778,
|
| 728 |
-
"step": 9300
|
| 729 |
-
},
|
| 730 |
-
{
|
| 731 |
-
"epoch": 1.36,
|
| 732 |
-
"learning_rate": 3.5976302548287614e-05,
|
| 733 |
-
"loss": 0.9794,
|
| 734 |
-
"step": 9400
|
| 735 |
-
},
|
| 736 |
-
{
|
| 737 |
-
"epoch": 1.37,
|
| 738 |
-
"learning_rate": 3.5164745982795e-05,
|
| 739 |
-
"loss": 0.9717,
|
| 740 |
-
"step": 9500
|
| 741 |
-
},
|
| 742 |
-
{
|
| 743 |
-
"epoch": 1.37,
|
| 744 |
-
"eval_loss": 0.25609487295150757,
|
| 745 |
-
"eval_runtime": 299.6976,
|
| 746 |
-
"eval_samples_per_second": 19.326,
|
| 747 |
-
"eval_steps_per_second": 0.304,
|
| 748 |
-
"eval_wer": 0.3807064420476392,
|
| 749 |
-
"step": 9500
|
| 750 |
-
},
|
| 751 |
-
{
|
| 752 |
-
"epoch": 1.39,
|
| 753 |
-
"learning_rate": 3.435318941730239e-05,
|
| 754 |
-
"loss": 0.9752,
|
| 755 |
-
"step": 9600
|
| 756 |
-
},
|
| 757 |
-
{
|
| 758 |
-
"epoch": 1.4,
|
| 759 |
-
"learning_rate": 3.354163285180977e-05,
|
| 760 |
-
"loss": 0.965,
|
| 761 |
-
"step": 9700
|
| 762 |
-
},
|
| 763 |
-
{
|
| 764 |
-
"epoch": 1.42,
|
| 765 |
-
"learning_rate": 3.2730076286317155e-05,
|
| 766 |
-
"loss": 0.9522,
|
| 767 |
-
"step": 9800
|
| 768 |
-
},
|
| 769 |
-
{
|
| 770 |
-
"epoch": 1.43,
|
| 771 |
-
"learning_rate": 3.191851972082454e-05,
|
| 772 |
-
"loss": 0.9718,
|
| 773 |
-
"step": 9900
|
| 774 |
-
},
|
| 775 |
-
{
|
| 776 |
-
"epoch": 1.45,
|
| 777 |
-
"learning_rate": 3.110696315533193e-05,
|
| 778 |
-
"loss": 0.9605,
|
| 779 |
-
"step": 10000
|
| 780 |
-
},
|
| 781 |
-
{
|
| 782 |
-
"epoch": 1.45,
|
| 783 |
-
"eval_loss": 0.25231894850730896,
|
| 784 |
-
"eval_runtime": 297.5796,
|
| 785 |
-
"eval_samples_per_second": 19.464,
|
| 786 |
-
"eval_steps_per_second": 0.306,
|
| 787 |
-
"eval_wer": 0.3782951546156603,
|
| 788 |
-
"step": 10000
|
| 789 |
-
},
|
| 790 |
-
{
|
| 791 |
-
"epoch": 1.46,
|
| 792 |
-
"learning_rate": 3.0295406589839315e-05,
|
| 793 |
-
"loss": 0.9635,
|
| 794 |
-
"step": 10100
|
| 795 |
-
},
|
| 796 |
-
{
|
| 797 |
-
"epoch": 1.48,
|
| 798 |
-
"learning_rate": 2.94838500243467e-05,
|
| 799 |
-
"loss": 0.9632,
|
| 800 |
-
"step": 10200
|
| 801 |
-
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 1.49,
|
| 804 |
-
"learning_rate": 2.8672293458854082e-05,
|
| 805 |
-
"loss": 0.9548,
|
| 806 |
-
"step": 10300
|
| 807 |
-
},
|
| 808 |
-
{
|
| 809 |
-
"epoch": 1.5,
|
| 810 |
-
"learning_rate": 2.7868852459016392e-05,
|
| 811 |
-
"loss": 0.9554,
|
| 812 |
-
"step": 10400
|
| 813 |
-
},
|
| 814 |
-
{
|
| 815 |
-
"epoch": 1.52,
|
| 816 |
-
"learning_rate": 2.7057295893523783e-05,
|
| 817 |
-
"loss": 0.96,
|
| 818 |
-
"step": 10500
|
| 819 |
-
},
|
| 820 |
-
{
|
| 821 |
-
"epoch": 1.52,
|
| 822 |
-
"eval_loss": 0.24939315021038055,
|
| 823 |
-
"eval_runtime": 300.036,
|
| 824 |
-
"eval_samples_per_second": 19.304,
|
| 825 |
-
"eval_steps_per_second": 0.303,
|
| 826 |
-
"eval_wer": 0.3787513441298185,
|
| 827 |
-
"step": 10500
|
| 828 |
-
},
|
| 829 |
-
{
|
| 830 |
-
"epoch": 1.53,
|
| 831 |
-
"learning_rate": 2.6245739328031166e-05,
|
| 832 |
-
"loss": 0.9611,
|
| 833 |
-
"step": 10600
|
| 834 |
-
},
|
| 835 |
-
{
|
| 836 |
-
"epoch": 1.55,
|
| 837 |
-
"learning_rate": 2.543418276253855e-05,
|
| 838 |
-
"loss": 0.9594,
|
| 839 |
-
"step": 10700
|
| 840 |
-
},
|
| 841 |
-
{
|
| 842 |
-
"epoch": 1.56,
|
| 843 |
-
"learning_rate": 2.4622626197045937e-05,
|
| 844 |
-
"loss": 0.9589,
|
| 845 |
-
"step": 10800
|
| 846 |
-
},
|
| 847 |
-
{
|
| 848 |
-
"epoch": 1.58,
|
| 849 |
-
"learning_rate": 2.381106963155332e-05,
|
| 850 |
-
"loss": 0.9441,
|
| 851 |
-
"step": 10900
|
| 852 |
-
},
|
| 853 |
-
{
|
| 854 |
-
"epoch": 1.59,
|
| 855 |
-
"learning_rate": 2.2999513066060704e-05,
|
| 856 |
-
"loss": 0.9442,
|
| 857 |
-
"step": 11000
|
| 858 |
-
},
|
| 859 |
-
{
|
| 860 |
-
"epoch": 1.59,
|
| 861 |
-
"eval_loss": 0.24783751368522644,
|
| 862 |
-
"eval_runtime": 297.7741,
|
| 863 |
-
"eval_samples_per_second": 19.451,
|
| 864 |
-
"eval_steps_per_second": 0.306,
|
| 865 |
-
"eval_wer": 0.3760142070448695,
|
| 866 |
-
"step": 11000
|
| 867 |
-
},
|
| 868 |
-
{
|
| 869 |
-
"epoch": 1.61,
|
| 870 |
-
"learning_rate": 2.218795650056809e-05,
|
| 871 |
-
"loss": 0.9496,
|
| 872 |
-
"step": 11100
|
| 873 |
-
},
|
| 874 |
-
{
|
| 875 |
-
"epoch": 1.62,
|
| 876 |
-
"learning_rate": 2.1376399935075474e-05,
|
| 877 |
-
"loss": 0.9486,
|
| 878 |
-
"step": 11200
|
| 879 |
-
},
|
| 880 |
-
{
|
| 881 |
-
"epoch": 1.64,
|
| 882 |
-
"learning_rate": 2.056484336958286e-05,
|
| 883 |
-
"loss": 0.9558,
|
| 884 |
-
"step": 11300
|
| 885 |
-
},
|
| 886 |
-
{
|
| 887 |
-
"epoch": 1.65,
|
| 888 |
-
"learning_rate": 1.9753286804090245e-05,
|
| 889 |
-
"loss": 0.9486,
|
| 890 |
-
"step": 11400
|
| 891 |
-
},
|
| 892 |
-
{
|
| 893 |
-
"epoch": 1.66,
|
| 894 |
-
"learning_rate": 1.8941730238597632e-05,
|
| 895 |
-
"loss": 0.9564,
|
| 896 |
-
"step": 11500
|
| 897 |
-
},
|
| 898 |
-
{
|
| 899 |
-
"epoch": 1.66,
|
| 900 |
-
"eval_loss": 0.2454409897327423,
|
| 901 |
-
"eval_runtime": 296.0265,
|
| 902 |
-
"eval_samples_per_second": 19.566,
|
| 903 |
-
"eval_steps_per_second": 0.307,
|
| 904 |
-
"eval_wer": 0.3733096549252175,
|
| 905 |
-
"step": 11500
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"epoch": 1.68,
|
| 909 |
-
"learning_rate": 1.8130173673105015e-05,
|
| 910 |
-
"loss": 0.9427,
|
| 911 |
-
"step": 11600
|
| 912 |
-
},
|
| 913 |
-
{
|
| 914 |
-
"epoch": 1.69,
|
| 915 |
-
"learning_rate": 1.73186171076124e-05,
|
| 916 |
-
"loss": 0.9423,
|
| 917 |
-
"step": 11700
|
| 918 |
-
},
|
| 919 |
-
{
|
| 920 |
-
"epoch": 1.71,
|
| 921 |
-
"learning_rate": 1.6507060542119786e-05,
|
| 922 |
-
"loss": 0.9503,
|
| 923 |
-
"step": 11800
|
| 924 |
-
},
|
| 925 |
-
{
|
| 926 |
-
"epoch": 1.72,
|
| 927 |
-
"learning_rate": 1.5695503976627173e-05,
|
| 928 |
-
"loss": 0.9383,
|
| 929 |
-
"step": 11900
|
| 930 |
-
},
|
| 931 |
-
{
|
| 932 |
-
"epoch": 1.74,
|
| 933 |
-
"learning_rate": 1.4883947411134558e-05,
|
| 934 |
-
"loss": 0.9436,
|
| 935 |
-
"step": 12000
|
| 936 |
-
},
|
| 937 |
-
{
|
| 938 |
-
"epoch": 1.74,
|
| 939 |
-
"eval_loss": 0.24390804767608643,
|
| 940 |
-
"eval_runtime": 295.4584,
|
| 941 |
-
"eval_samples_per_second": 19.603,
|
| 942 |
-
"eval_steps_per_second": 0.308,
|
| 943 |
-
"eval_wer": 0.37467822346769203,
|
| 944 |
-
"step": 12000
|
| 945 |
-
},
|
| 946 |
-
{
|
| 947 |
-
"epoch": 1.75,
|
| 948 |
-
"learning_rate": 1.4072390845641942e-05,
|
| 949 |
-
"loss": 0.9491,
|
| 950 |
-
"step": 12100
|
| 951 |
-
},
|
| 952 |
-
{
|
| 953 |
-
"epoch": 1.77,
|
| 954 |
-
"learning_rate": 1.3260834280149325e-05,
|
| 955 |
-
"loss": 0.9419,
|
| 956 |
-
"step": 12200
|
| 957 |
-
},
|
| 958 |
-
{
|
| 959 |
-
"epoch": 1.78,
|
| 960 |
-
"learning_rate": 1.2449277714656712e-05,
|
| 961 |
-
"loss": 0.9517,
|
| 962 |
-
"step": 12300
|
| 963 |
-
},
|
| 964 |
-
{
|
| 965 |
-
"epoch": 1.79,
|
| 966 |
-
"learning_rate": 1.1637721149164097e-05,
|
| 967 |
-
"loss": 0.9367,
|
| 968 |
-
"step": 12400
|
| 969 |
-
},
|
| 970 |
-
{
|
| 971 |
-
"epoch": 1.81,
|
| 972 |
-
"learning_rate": 1.0826164583671483e-05,
|
| 973 |
-
"loss": 0.938,
|
| 974 |
-
"step": 12500
|
| 975 |
-
},
|
| 976 |
-
{
|
| 977 |
-
"epoch": 1.81,
|
| 978 |
-
"eval_loss": 0.24111612141132355,
|
| 979 |
-
"eval_runtime": 296.9314,
|
| 980 |
-
"eval_samples_per_second": 19.506,
|
| 981 |
-
"eval_steps_per_second": 0.306,
|
| 982 |
-
"eval_wer": 0.37159894424712436,
|
| 983 |
-
"step": 12500
|
| 984 |
-
},
|
| 985 |
-
{
|
| 986 |
-
"epoch": 1.82,
|
| 987 |
-
"learning_rate": 1.0014608018178868e-05,
|
| 988 |
-
"loss": 0.9337,
|
| 989 |
-
"step": 12600
|
| 990 |
-
},
|
| 991 |
-
{
|
| 992 |
-
"epoch": 1.84,
|
| 993 |
-
"learning_rate": 9.203051452686253e-06,
|
| 994 |
-
"loss": 0.9284,
|
| 995 |
-
"step": 12700
|
| 996 |
-
},
|
| 997 |
-
{
|
| 998 |
-
"epoch": 1.85,
|
| 999 |
-
"learning_rate": 8.391494887193638e-06,
|
| 1000 |
-
"loss": 0.938,
|
| 1001 |
-
"step": 12800
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"epoch": 1.87,
|
| 1005 |
-
"learning_rate": 7.579938321701023e-06,
|
| 1006 |
-
"loss": 0.9365,
|
| 1007 |
-
"step": 12900
|
| 1008 |
-
},
|
| 1009 |
-
{
|
| 1010 |
-
"epoch": 1.88,
|
| 1011 |
-
"learning_rate": 6.768381756208409e-06,
|
| 1012 |
-
"loss": 0.9353,
|
| 1013 |
-
"step": 13000
|
| 1014 |
-
},
|
| 1015 |
-
{
|
| 1016 |
-
"epoch": 1.88,
|
| 1017 |
-
"eval_loss": 0.23965783417224884,
|
| 1018 |
-
"eval_runtime": 296.6078,
|
| 1019 |
-
"eval_samples_per_second": 19.527,
|
| 1020 |
-
"eval_steps_per_second": 0.307,
|
| 1021 |
-
"eval_wer": 0.3697904786731402,
|
| 1022 |
-
"step": 13000
|
| 1023 |
-
},
|
| 1024 |
-
{
|
| 1025 |
-
"epoch": 1.9,
|
| 1026 |
-
"learning_rate": 5.956825190715793e-06,
|
| 1027 |
-
"loss": 0.9413,
|
| 1028 |
-
"step": 13100
|
| 1029 |
-
},
|
| 1030 |
-
{
|
| 1031 |
-
"epoch": 1.91,
|
| 1032 |
-
"learning_rate": 5.153384190878104e-06,
|
| 1033 |
-
"loss": 0.9356,
|
| 1034 |
-
"step": 13200
|
| 1035 |
-
},
|
| 1036 |
-
{
|
| 1037 |
-
"epoch": 1.92,
|
| 1038 |
-
"learning_rate": 4.34182762538549e-06,
|
| 1039 |
-
"loss": 0.9209,
|
| 1040 |
-
"step": 13300
|
| 1041 |
-
},
|
| 1042 |
-
{
|
| 1043 |
-
"epoch": 1.94,
|
| 1044 |
-
"learning_rate": 3.530271059892875e-06,
|
| 1045 |
-
"loss": 0.9362,
|
| 1046 |
-
"step": 13400
|
| 1047 |
-
},
|
| 1048 |
-
{
|
| 1049 |
-
"epoch": 1.95,
|
| 1050 |
-
"learning_rate": 2.71871449440026e-06,
|
| 1051 |
-
"loss": 0.9271,
|
| 1052 |
-
"step": 13500
|
| 1053 |
-
},
|
| 1054 |
-
{
|
| 1055 |
-
"epoch": 1.95,
|
| 1056 |
-
"eval_loss": 0.23875188827514648,
|
| 1057 |
-
"eval_runtime": 296.9414,
|
| 1058 |
-
"eval_samples_per_second": 19.506,
|
| 1059 |
-
"eval_steps_per_second": 0.306,
|
| 1060 |
-
"eval_wer": 0.3680797679950471,
|
| 1061 |
-
"step": 13500
|
| 1062 |
-
},
|
| 1063 |
-
{
|
| 1064 |
-
"epoch": 1.97,
|
| 1065 |
-
"learning_rate": 1.907157928907645e-06,
|
| 1066 |
-
"loss": 0.9288,
|
| 1067 |
-
"step": 13600
|
| 1068 |
-
},
|
| 1069 |
-
{
|
| 1070 |
-
"epoch": 1.98,
|
| 1071 |
-
"learning_rate": 1.0956013634150302e-06,
|
| 1072 |
-
"loss": 0.9345,
|
| 1073 |
-
"step": 13700
|
| 1074 |
-
},
|
| 1075 |
-
{
|
| 1076 |
-
"epoch": 2.0,
|
| 1077 |
-
"learning_rate": 2.840447979224152e-07,
|
| 1078 |
-
"loss": 0.9326,
|
| 1079 |
-
"step": 13800
|
| 1080 |
-
},
|
| 1081 |
-
{
|
| 1082 |
-
"epoch": 2.0,
|
| 1083 |
-
"step": 13822,
|
| 1084 |
-
"total_flos": 1.2600843645735263e+20,
|
| 1085 |
-
"train_loss": 1.442369053426242,
|
| 1086 |
-
"train_runtime": 53680.5392,
|
| 1087 |
-
"train_samples_per_second": 16.478,
|
| 1088 |
-
"train_steps_per_second": 0.257
|
| 1089 |
-
}
|
| 1090 |
-
],
|
| 1091 |
-
"max_steps": 13822,
|
| 1092 |
-
"num_train_epochs": 2,
|
| 1093 |
-
"total_flos": 1.2600843645735263e+20,
|
| 1094 |
-
"trial_name": null,
|
| 1095 |
-
"trial_params": null
|
| 1096 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
.ipynb_checkpoints/validate-checkpoint.sh
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
python eval.py --model_id="./" --dataset="mozilla-foundation/common_voice_8_0" --config="fr" --split="test" --log_outputs
|
| 2 |
+
python eval.py --model_id="./" --dataset="speech-recognition-community-v2/dev_data" --config="fr" \
|
| 3 |
+
--split="validation" --chunk_length_s="5.0" --stride_length_s="1.0" --log_outputs
|
README.md
CHANGED
|
@@ -9,8 +9,30 @@ tags:
|
|
| 9 |
datasets:
|
| 10 |
- common_voice
|
| 11 |
model-index:
|
| 12 |
-
- name:
|
| 13 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
---
|
| 15 |
|
| 16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -18,7 +40,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
#
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 9 |
datasets:
|
| 10 |
- common_voice
|
| 11 |
model-index:
|
| 12 |
+
- name: name: xls-r-300m-fr
|
| 13 |
+
results:
|
| 14 |
+
-task:
|
| 15 |
+
name: Speech Recognition
|
| 16 |
+
type: automatic-speech-recognition
|
| 17 |
+
dataset:
|
| 18 |
+
name: Common Voice 8.0 fr
|
| 19 |
+
type: mozilla-foundation/common_voice_8_0
|
| 20 |
+
args: fr
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Test WER
|
| 23 |
+
type: wer
|
| 24 |
+
value: 21.58
|
| 25 |
+
-task:
|
| 26 |
+
name: Speech Recognition
|
| 27 |
+
type: automatic-speech-recognition
|
| 28 |
+
dataset:
|
| 29 |
+
name: dev_data
|
| 30 |
+
type: speech-recognition-community-v2/dev_data
|
| 31 |
+
args: fr
|
| 32 |
+
metrics:
|
| 33 |
+
- name: Validation WER
|
| 34 |
+
type: wer
|
| 35 |
+
value: 36.03
|
| 36 |
---
|
| 37 |
|
| 38 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 40 |
|
| 41 |
#
|
| 42 |
|
| 43 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - FR dataset.
|
| 44 |
|
| 45 |
## Model description
|
| 46 |
|
added_tokens.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{
|
|
|
|
| 1 |
+
{}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_dropout": 0.05,
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-xls-r-300m",
|
| 3 |
"activation_dropout": 0.05,
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
eval.py
CHANGED
|
@@ -85,7 +85,7 @@ def main(args):
|
|
| 85 |
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
|
| 86 |
|
| 87 |
# for testing: only process the first two examples as a test
|
| 88 |
-
dataset = dataset.select(range(20))
|
| 89 |
|
| 90 |
# load processor
|
| 91 |
feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
|
|
@@ -105,7 +105,7 @@ def main(args):
|
|
| 105 |
batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
|
| 106 |
)
|
| 107 |
|
| 108 |
-
batch["prediction"] =
|
| 109 |
batch["target"] = normalize_text(batch["sentence"])
|
| 110 |
return batch
|
| 111 |
|
|
|
|
| 85 |
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
|
| 86 |
|
| 87 |
# for testing: only process the first two examples as a test
|
| 88 |
+
#dataset = dataset.select(range(20))
|
| 89 |
|
| 90 |
# load processor
|
| 91 |
feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
|
|
|
|
| 105 |
batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
|
| 106 |
)
|
| 107 |
|
| 108 |
+
batch["prediction"] = "".join(prediction["text"].split("<s>"))
|
| 109 |
batch["target"] = normalize_text(batch["sentence"])
|
| 110 |
return batch
|
| 111 |
|
eval_results.json
DELETED
|
@@ -1,8 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"eval_loss": 0.22597630321979523,
|
| 3 |
-
"eval_runtime": 276.6037,
|
| 4 |
-
"eval_samples": 4469,
|
| 5 |
-
"eval_samples_per_second": 16.157,
|
| 6 |
-
"eval_steps_per_second": 0.253,
|
| 7 |
-
"eval_wer": 0.2948882012145466
|
| 8 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
log_mozilla-foundation_common_voice_8_0_fr_test_predictions.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
log_mozilla-foundation_common_voice_8_0_fr_test_targets.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
log_speech-recognition-community-v2_dev_data_fr_validation_predictions.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
log_speech-recognition-community-v2_dev_data_fr_validation_targets.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
mozilla-foundation_common_voice_8_0_fr_test_eval_results.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
WER: 0.21587470509795875
|
| 2 |
+
CER: 0.06356032070032196
|
run.sh
CHANGED
|
@@ -1,17 +1,17 @@
|
|
| 1 |
WANDB_PROJECT=auto-speech-recognition-french
|
| 2 |
python run_speech_recognition_ctc.py \
|
| 3 |
--dataset_name="mozilla-foundation/common_voice_8_0" \
|
| 4 |
-
--model_name_or_path="./checkpoint-18000" \
|
| 5 |
--dataset_config_name="fr" \
|
| 6 |
--tokenizer_name_or_path="./" \
|
| 7 |
--output_dir="./" \
|
| 8 |
--overwrite_output_dir \
|
| 9 |
-
--num_train_epochs="
|
| 10 |
--per_device_train_batch_size="64" \
|
| 11 |
--per_device_eval_batch_size="64" \
|
| 12 |
--gradient_accumulation_steps="1" \
|
| 13 |
--learning_rate="1e-4" \
|
| 14 |
-
--warmup_steps="
|
| 15 |
--length_column_name="input_length" \
|
| 16 |
--evaluation_strategy="steps" \
|
| 17 |
--text_column_name="sentence" \
|
|
@@ -23,18 +23,19 @@ python run_speech_recognition_ctc.py \
|
|
| 23 |
--save_total_limit="2" \
|
| 24 |
--freeze_feature_encoder \
|
| 25 |
--feat_proj_dropout="0.0" \
|
| 26 |
-
--mask_time_prob="0.
|
| 27 |
--mask_time_length="10" \
|
| 28 |
--mask_feature_prob="0.25" \
|
| 29 |
--mask_feature_length="10" \
|
| 30 |
--gradient_checkpointing \
|
| 31 |
--report_to="wandb" \
|
| 32 |
--run_name="xls-r-300m-fr" \
|
|
|
|
| 33 |
--max_duration_in_seconds="10" \
|
| 34 |
--use_auth_token \
|
| 35 |
--fp16 \
|
| 36 |
--group_by_length \
|
| 37 |
--preprocessing_num_workers="64" \
|
| 38 |
-
--do_train
|
| 39 |
--load_best_model_at_end \
|
| 40 |
--push_to_hub
|
|
|
|
| 1 |
WANDB_PROJECT=auto-speech-recognition-french
|
| 2 |
python run_speech_recognition_ctc.py \
|
| 3 |
--dataset_name="mozilla-foundation/common_voice_8_0" \
|
| 4 |
+
--model_name_or_path="./checkpoint-18000-keep" \
|
| 5 |
--dataset_config_name="fr" \
|
| 6 |
--tokenizer_name_or_path="./" \
|
| 7 |
--output_dir="./" \
|
| 8 |
--overwrite_output_dir \
|
| 9 |
+
--num_train_epochs="4" \
|
| 10 |
--per_device_train_batch_size="64" \
|
| 11 |
--per_device_eval_batch_size="64" \
|
| 12 |
--gradient_accumulation_steps="1" \
|
| 13 |
--learning_rate="1e-4" \
|
| 14 |
+
--warmup_steps="500" \
|
| 15 |
--length_column_name="input_length" \
|
| 16 |
--evaluation_strategy="steps" \
|
| 17 |
--text_column_name="sentence" \
|
|
|
|
| 23 |
--save_total_limit="2" \
|
| 24 |
--freeze_feature_encoder \
|
| 25 |
--feat_proj_dropout="0.0" \
|
| 26 |
+
--mask_time_prob="0.6" \
|
| 27 |
--mask_time_length="10" \
|
| 28 |
--mask_feature_prob="0.25" \
|
| 29 |
--mask_feature_length="10" \
|
| 30 |
--gradient_checkpointing \
|
| 31 |
--report_to="wandb" \
|
| 32 |
--run_name="xls-r-300m-fr" \
|
| 33 |
+
--max_eval_samples="4500" \
|
| 34 |
--max_duration_in_seconds="10" \
|
| 35 |
--use_auth_token \
|
| 36 |
--fp16 \
|
| 37 |
--group_by_length \
|
| 38 |
--preprocessing_num_workers="64" \
|
| 39 |
+
--do_train --do_eval \
|
| 40 |
--load_best_model_at_end \
|
| 41 |
--push_to_hub
|
run_speech_recognition_ctc.py
CHANGED
|
@@ -402,6 +402,7 @@ def main():
|
|
| 402 |
data_args.dataset_config_name,
|
| 403 |
split=data_args.train_split_name,
|
| 404 |
use_auth_token=data_args.use_auth_token,
|
|
|
|
| 405 |
)
|
| 406 |
|
| 407 |
if data_args.audio_column_name not in raw_datasets["train"].column_names:
|
|
@@ -427,6 +428,7 @@ def main():
|
|
| 427 |
data_args.dataset_config_name,
|
| 428 |
split=data_args.eval_split_name,
|
| 429 |
use_auth_token=data_args.use_auth_token,
|
|
|
|
| 430 |
)
|
| 431 |
|
| 432 |
if data_args.max_eval_samples is not None:
|
|
@@ -607,7 +609,7 @@ def main():
|
|
| 607 |
prepare_dataset,
|
| 608 |
remove_columns=next(iter(raw_datasets.values())).column_names,
|
| 609 |
batch_size=-1,
|
| 610 |
-
desc="preprocess datasets",
|
| 611 |
)
|
| 612 |
|
| 613 |
def is_audio_in_length_range(length):
|
|
@@ -686,7 +688,7 @@ def main():
|
|
| 686 |
# 8. Finally, we can start training
|
| 687 |
|
| 688 |
# Training
|
| 689 |
-
if training_args.do_train
|
| 690 |
|
| 691 |
# use last checkpoint if exist
|
| 692 |
if last_checkpoint is not None:
|
|
|
|
| 402 |
data_args.dataset_config_name,
|
| 403 |
split=data_args.train_split_name,
|
| 404 |
use_auth_token=data_args.use_auth_token,
|
| 405 |
+
cache_dir="../downloaded_data/"
|
| 406 |
)
|
| 407 |
|
| 408 |
if data_args.audio_column_name not in raw_datasets["train"].column_names:
|
|
|
|
| 428 |
data_args.dataset_config_name,
|
| 429 |
split=data_args.eval_split_name,
|
| 430 |
use_auth_token=data_args.use_auth_token,
|
| 431 |
+
cache_dir="../downloaded_data/"
|
| 432 |
)
|
| 433 |
|
| 434 |
if data_args.max_eval_samples is not None:
|
|
|
|
| 609 |
prepare_dataset,
|
| 610 |
remove_columns=next(iter(raw_datasets.values())).column_names,
|
| 611 |
batch_size=-1,
|
| 612 |
+
desc="preprocess datasets",cache_dir="../data_cache/"
|
| 613 |
)
|
| 614 |
|
| 615 |
def is_audio_in_length_range(length):
|
|
|
|
| 688 |
# 8. Finally, we can start training
|
| 689 |
|
| 690 |
# Training
|
| 691 |
+
if training_args.do_train:
|
| 692 |
|
| 693 |
# use last checkpoint if exist
|
| 694 |
if last_checkpoint is not None:
|
special_tokens_map.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"bos_token":
|
|
|
|
| 1 |
+
{"bos_token": null, "eos_token": null, "unk_token": "<unk>", "pad_token": "<pad>"}
|
speech-recognition-community-v2_dev_data_fr_validation_eval_results.txt
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
WER: 0.3603687452909631
|
| 2 |
+
CER: 0.14097275038988377
|
tokenizer_config.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"unk_token": "<unk>", "bos_token":
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": null, "eos_token": null, "pad_token": "<pad>", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "padding": true, "truncation": true, "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
train_results.json
DELETED
|
@@ -1,8 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"epoch": 2.0,
|
| 3 |
-
"train_loss": 1.442369053426242,
|
| 4 |
-
"train_runtime": 53680.5392,
|
| 5 |
-
"train_samples": 442265,
|
| 6 |
-
"train_samples_per_second": 16.478,
|
| 7 |
-
"train_steps_per_second": 0.257
|
| 8 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
trainer_state.json
DELETED
|
@@ -1,1096 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": 0.23875188827514648,
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-13500",
|
| 4 |
-
"epoch": 2.0,
|
| 5 |
-
"global_step": 13822,
|
| 6 |
-
"is_hyper_param_search": false,
|
| 7 |
-
"is_local_process_zero": true,
|
| 8 |
-
"is_world_process_zero": true,
|
| 9 |
-
"log_history": [
|
| 10 |
-
{
|
| 11 |
-
"epoch": 0.01,
|
| 12 |
-
"learning_rate": 6.533333333333333e-06,
|
| 13 |
-
"loss": 17.2403,
|
| 14 |
-
"step": 100
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 0.03,
|
| 18 |
-
"learning_rate": 1.32e-05,
|
| 19 |
-
"loss": 10.2311,
|
| 20 |
-
"step": 200
|
| 21 |
-
},
|
| 22 |
-
{
|
| 23 |
-
"epoch": 0.04,
|
| 24 |
-
"learning_rate": 1.9800000000000004e-05,
|
| 25 |
-
"loss": 7.834,
|
| 26 |
-
"step": 300
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 0.06,
|
| 30 |
-
"learning_rate": 2.646666666666667e-05,
|
| 31 |
-
"loss": 6.0656,
|
| 32 |
-
"step": 400
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 0.07,
|
| 36 |
-
"learning_rate": 3.313333333333333e-05,
|
| 37 |
-
"loss": 4.3748,
|
| 38 |
-
"step": 500
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.07,
|
| 42 |
-
"eval_loss": 3.878422975540161,
|
| 43 |
-
"eval_runtime": 285.8223,
|
| 44 |
-
"eval_samples_per_second": 20.264,
|
| 45 |
-
"eval_steps_per_second": 0.318,
|
| 46 |
-
"eval_wer": 1.0,
|
| 47 |
-
"step": 500
|
| 48 |
-
},
|
| 49 |
-
{
|
| 50 |
-
"epoch": 0.09,
|
| 51 |
-
"learning_rate": 3.9800000000000005e-05,
|
| 52 |
-
"loss": 3.2923,
|
| 53 |
-
"step": 600
|
| 54 |
-
},
|
| 55 |
-
{
|
| 56 |
-
"epoch": 0.1,
|
| 57 |
-
"learning_rate": 4.646666666666667e-05,
|
| 58 |
-
"loss": 2.9475,
|
| 59 |
-
"step": 700
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.12,
|
| 63 |
-
"learning_rate": 5.3133333333333335e-05,
|
| 64 |
-
"loss": 2.8639,
|
| 65 |
-
"step": 800
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.13,
|
| 69 |
-
"learning_rate": 5.9800000000000003e-05,
|
| 70 |
-
"loss": 2.8265,
|
| 71 |
-
"step": 900
|
| 72 |
-
},
|
| 73 |
-
{
|
| 74 |
-
"epoch": 0.14,
|
| 75 |
-
"learning_rate": 6.646666666666667e-05,
|
| 76 |
-
"loss": 2.8068,
|
| 77 |
-
"step": 1000
|
| 78 |
-
},
|
| 79 |
-
{
|
| 80 |
-
"epoch": 0.14,
|
| 81 |
-
"eval_loss": 2.828850746154785,
|
| 82 |
-
"eval_runtime": 292.3877,
|
| 83 |
-
"eval_samples_per_second": 19.809,
|
| 84 |
-
"eval_steps_per_second": 0.311,
|
| 85 |
-
"eval_wer": 0.9826485059793412,
|
| 86 |
-
"step": 1000
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.16,
|
| 90 |
-
"learning_rate": 7.306666666666668e-05,
|
| 91 |
-
"loss": 2.779,
|
| 92 |
-
"step": 1100
|
| 93 |
-
},
|
| 94 |
-
{
|
| 95 |
-
"epoch": 0.17,
|
| 96 |
-
"learning_rate": 7.973333333333334e-05,
|
| 97 |
-
"loss": 2.6402,
|
| 98 |
-
"step": 1200
|
| 99 |
-
},
|
| 100 |
-
{
|
| 101 |
-
"epoch": 0.19,
|
| 102 |
-
"learning_rate": 8.64e-05,
|
| 103 |
-
"loss": 2.1119,
|
| 104 |
-
"step": 1300
|
| 105 |
-
},
|
| 106 |
-
{
|
| 107 |
-
"epoch": 0.2,
|
| 108 |
-
"learning_rate": 9.306666666666667e-05,
|
| 109 |
-
"loss": 1.7965,
|
| 110 |
-
"step": 1400
|
| 111 |
-
},
|
| 112 |
-
{
|
| 113 |
-
"epoch": 0.22,
|
| 114 |
-
"learning_rate": 9.973333333333334e-05,
|
| 115 |
-
"loss": 1.6698,
|
| 116 |
-
"step": 1500
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 0.22,
|
| 120 |
-
"eval_loss": 0.881136417388916,
|
| 121 |
-
"eval_runtime": 297.1806,
|
| 122 |
-
"eval_samples_per_second": 19.49,
|
| 123 |
-
"eval_steps_per_second": 0.306,
|
| 124 |
-
"eval_wer": 0.7127472384241911,
|
| 125 |
-
"step": 1500
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 0.23,
|
| 129 |
-
"learning_rate": 9.92209056971271e-05,
|
| 130 |
-
"loss": 1.5882,
|
| 131 |
-
"step": 1600
|
| 132 |
-
},
|
| 133 |
-
{
|
| 134 |
-
"epoch": 0.25,
|
| 135 |
-
"learning_rate": 9.840934913163448e-05,
|
| 136 |
-
"loss": 1.5172,
|
| 137 |
-
"step": 1700
|
| 138 |
-
},
|
| 139 |
-
{
|
| 140 |
-
"epoch": 0.26,
|
| 141 |
-
"learning_rate": 9.759779256614186e-05,
|
| 142 |
-
"loss": 1.4579,
|
| 143 |
-
"step": 1800
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.27,
|
| 147 |
-
"learning_rate": 9.678623600064926e-05,
|
| 148 |
-
"loss": 1.3829,
|
| 149 |
-
"step": 1900
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.29,
|
| 153 |
-
"learning_rate": 9.597467943515663e-05,
|
| 154 |
-
"loss": 1.3488,
|
| 155 |
-
"step": 2000
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 0.29,
|
| 159 |
-
"eval_loss": 0.516592800617218,
|
| 160 |
-
"eval_runtime": 301.2842,
|
| 161 |
-
"eval_samples_per_second": 19.224,
|
| 162 |
-
"eval_steps_per_second": 0.302,
|
| 163 |
-
"eval_wer": 0.5369024731988661,
|
| 164 |
-
"step": 2000
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 0.3,
|
| 168 |
-
"learning_rate": 9.516312286966402e-05,
|
| 169 |
-
"loss": 1.2981,
|
| 170 |
-
"step": 2100
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 0.32,
|
| 174 |
-
"learning_rate": 9.43515663041714e-05,
|
| 175 |
-
"loss": 1.2845,
|
| 176 |
-
"step": 2200
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 0.33,
|
| 180 |
-
"learning_rate": 9.354000973867879e-05,
|
| 181 |
-
"loss": 1.2459,
|
| 182 |
-
"step": 2300
|
| 183 |
-
},
|
| 184 |
-
{
|
| 185 |
-
"epoch": 0.35,
|
| 186 |
-
"learning_rate": 9.272845317318618e-05,
|
| 187 |
-
"loss": 1.2255,
|
| 188 |
-
"step": 2400
|
| 189 |
-
},
|
| 190 |
-
{
|
| 191 |
-
"epoch": 0.36,
|
| 192 |
-
"learning_rate": 9.191689660769356e-05,
|
| 193 |
-
"loss": 1.2239,
|
| 194 |
-
"step": 2500
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 0.36,
|
| 198 |
-
"eval_loss": 0.4104757010936737,
|
| 199 |
-
"eval_runtime": 299.1395,
|
| 200 |
-
"eval_samples_per_second": 19.362,
|
| 201 |
-
"eval_steps_per_second": 0.304,
|
| 202 |
-
"eval_wer": 0.474111245071524,
|
| 203 |
-
"step": 2500
|
| 204 |
-
},
|
| 205 |
-
{
|
| 206 |
-
"epoch": 0.38,
|
| 207 |
-
"learning_rate": 9.110534004220094e-05,
|
| 208 |
-
"loss": 1.2024,
|
| 209 |
-
"step": 2600
|
| 210 |
-
},
|
| 211 |
-
{
|
| 212 |
-
"epoch": 0.39,
|
| 213 |
-
"learning_rate": 9.030189904236326e-05,
|
| 214 |
-
"loss": 1.1851,
|
| 215 |
-
"step": 2700
|
| 216 |
-
},
|
| 217 |
-
{
|
| 218 |
-
"epoch": 0.41,
|
| 219 |
-
"learning_rate": 8.949034247687063e-05,
|
| 220 |
-
"loss": 1.1768,
|
| 221 |
-
"step": 2800
|
| 222 |
-
},
|
| 223 |
-
{
|
| 224 |
-
"epoch": 0.42,
|
| 225 |
-
"learning_rate": 8.867878591137803e-05,
|
| 226 |
-
"loss": 1.1641,
|
| 227 |
-
"step": 2900
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 0.43,
|
| 231 |
-
"learning_rate": 8.786722934588541e-05,
|
| 232 |
-
"loss": 1.1537,
|
| 233 |
-
"step": 3000
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 0.43,
|
| 237 |
-
"eval_loss": 0.35850802063941956,
|
| 238 |
-
"eval_runtime": 299.662,
|
| 239 |
-
"eval_samples_per_second": 19.328,
|
| 240 |
-
"eval_steps_per_second": 0.304,
|
| 241 |
-
"eval_wer": 0.4448499462348073,
|
| 242 |
-
"step": 3000
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 0.45,
|
| 246 |
-
"learning_rate": 8.70556727803928e-05,
|
| 247 |
-
"loss": 1.1449,
|
| 248 |
-
"step": 3100
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 0.46,
|
| 252 |
-
"learning_rate": 8.624411621490018e-05,
|
| 253 |
-
"loss": 1.1379,
|
| 254 |
-
"step": 3200
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 0.48,
|
| 258 |
-
"learning_rate": 8.543255964940758e-05,
|
| 259 |
-
"loss": 1.1331,
|
| 260 |
-
"step": 3300
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 0.49,
|
| 264 |
-
"learning_rate": 8.462100308391495e-05,
|
| 265 |
-
"loss": 1.1205,
|
| 266 |
-
"step": 3400
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 0.51,
|
| 270 |
-
"learning_rate": 8.380944651842234e-05,
|
| 271 |
-
"loss": 1.1184,
|
| 272 |
-
"step": 3500
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 0.51,
|
| 276 |
-
"eval_loss": 0.333638072013855,
|
| 277 |
-
"eval_runtime": 297.0402,
|
| 278 |
-
"eval_samples_per_second": 19.499,
|
| 279 |
-
"eval_steps_per_second": 0.306,
|
| 280 |
-
"eval_wer": 0.42922545537489004,
|
| 281 |
-
"step": 3500
|
| 282 |
-
},
|
| 283 |
-
{
|
| 284 |
-
"epoch": 0.52,
|
| 285 |
-
"learning_rate": 8.299788995292971e-05,
|
| 286 |
-
"loss": 1.1014,
|
| 287 |
-
"step": 3600
|
| 288 |
-
},
|
| 289 |
-
{
|
| 290 |
-
"epoch": 0.54,
|
| 291 |
-
"learning_rate": 8.218633338743711e-05,
|
| 292 |
-
"loss": 1.1114,
|
| 293 |
-
"step": 3700
|
| 294 |
-
},
|
| 295 |
-
{
|
| 296 |
-
"epoch": 0.55,
|
| 297 |
-
"learning_rate": 8.13747768219445e-05,
|
| 298 |
-
"loss": 1.117,
|
| 299 |
-
"step": 3800
|
| 300 |
-
},
|
| 301 |
-
{
|
| 302 |
-
"epoch": 0.56,
|
| 303 |
-
"learning_rate": 8.056322025645188e-05,
|
| 304 |
-
"loss": 1.102,
|
| 305 |
-
"step": 3900
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 0.58,
|
| 309 |
-
"learning_rate": 7.975166369095926e-05,
|
| 310 |
-
"loss": 1.0968,
|
| 311 |
-
"step": 4000
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 0.58,
|
| 315 |
-
"eval_loss": 0.31949570775032043,
|
| 316 |
-
"eval_runtime": 296.0172,
|
| 317 |
-
"eval_samples_per_second": 19.566,
|
| 318 |
-
"eval_steps_per_second": 0.307,
|
| 319 |
-
"eval_wer": 0.4180162273127179,
|
| 320 |
-
"step": 4000
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 0.59,
|
| 324 |
-
"learning_rate": 7.894822269112158e-05,
|
| 325 |
-
"loss": 1.0942,
|
| 326 |
-
"step": 4100
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 0.61,
|
| 330 |
-
"learning_rate": 7.813666612562897e-05,
|
| 331 |
-
"loss": 1.0859,
|
| 332 |
-
"step": 4200
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 0.62,
|
| 336 |
-
"learning_rate": 7.732510956013635e-05,
|
| 337 |
-
"loss": 1.0767,
|
| 338 |
-
"step": 4300
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"epoch": 0.64,
|
| 342 |
-
"learning_rate": 7.652166856029866e-05,
|
| 343 |
-
"loss": 1.0766,
|
| 344 |
-
"step": 4400
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 0.65,
|
| 348 |
-
"learning_rate": 7.571011199480604e-05,
|
| 349 |
-
"loss": 1.0737,
|
| 350 |
-
"step": 4500
|
| 351 |
-
},
|
| 352 |
-
{
|
| 353 |
-
"epoch": 0.65,
|
| 354 |
-
"eval_loss": 0.30754634737968445,
|
| 355 |
-
"eval_runtime": 296.2378,
|
| 356 |
-
"eval_samples_per_second": 19.552,
|
| 357 |
-
"eval_steps_per_second": 0.307,
|
| 358 |
-
"eval_wer": 0.41408973899442797,
|
| 359 |
-
"step": 4500
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 0.67,
|
| 363 |
-
"learning_rate": 7.489855542931342e-05,
|
| 364 |
-
"loss": 1.0807,
|
| 365 |
-
"step": 4600
|
| 366 |
-
},
|
| 367 |
-
{
|
| 368 |
-
"epoch": 0.68,
|
| 369 |
-
"learning_rate": 7.40869988638208e-05,
|
| 370 |
-
"loss": 1.071,
|
| 371 |
-
"step": 4700
|
| 372 |
-
},
|
| 373 |
-
{
|
| 374 |
-
"epoch": 0.69,
|
| 375 |
-
"learning_rate": 7.32754422983282e-05,
|
| 376 |
-
"loss": 1.0613,
|
| 377 |
-
"step": 4800
|
| 378 |
-
},
|
| 379 |
-
{
|
| 380 |
-
"epoch": 0.71,
|
| 381 |
-
"learning_rate": 7.246388573283557e-05,
|
| 382 |
-
"loss": 1.0635,
|
| 383 |
-
"step": 4900
|
| 384 |
-
},
|
| 385 |
-
{
|
| 386 |
-
"epoch": 0.72,
|
| 387 |
-
"learning_rate": 7.165232916734297e-05,
|
| 388 |
-
"loss": 1.0677,
|
| 389 |
-
"step": 5000
|
| 390 |
-
},
|
| 391 |
-
{
|
| 392 |
-
"epoch": 0.72,
|
| 393 |
-
"eval_loss": 0.30150118470191956,
|
| 394 |
-
"eval_runtime": 297.4742,
|
| 395 |
-
"eval_samples_per_second": 19.471,
|
| 396 |
-
"eval_steps_per_second": 0.306,
|
| 397 |
-
"eval_wer": 0.4089250219948516,
|
| 398 |
-
"step": 5000
|
| 399 |
-
},
|
| 400 |
-
{
|
| 401 |
-
"epoch": 0.74,
|
| 402 |
-
"learning_rate": 7.084077260185034e-05,
|
| 403 |
-
"loss": 1.0707,
|
| 404 |
-
"step": 5100
|
| 405 |
-
},
|
| 406 |
-
{
|
| 407 |
-
"epoch": 0.75,
|
| 408 |
-
"learning_rate": 7.002921603635774e-05,
|
| 409 |
-
"loss": 1.0617,
|
| 410 |
-
"step": 5200
|
| 411 |
-
},
|
| 412 |
-
{
|
| 413 |
-
"epoch": 0.77,
|
| 414 |
-
"learning_rate": 6.921765947086512e-05,
|
| 415 |
-
"loss": 1.0566,
|
| 416 |
-
"step": 5300
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 0.78,
|
| 420 |
-
"learning_rate": 6.84061029053725e-05,
|
| 421 |
-
"loss": 1.0518,
|
| 422 |
-
"step": 5400
|
| 423 |
-
},
|
| 424 |
-
{
|
| 425 |
-
"epoch": 0.8,
|
| 426 |
-
"learning_rate": 6.760266190553481e-05,
|
| 427 |
-
"loss": 1.0462,
|
| 428 |
-
"step": 5500
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 0.8,
|
| 432 |
-
"eval_loss": 0.297052800655365,
|
| 433 |
-
"eval_runtime": 296.1592,
|
| 434 |
-
"eval_samples_per_second": 19.557,
|
| 435 |
-
"eval_steps_per_second": 0.307,
|
| 436 |
-
"eval_wer": 0.4077193782788621,
|
| 437 |
-
"step": 5500
|
| 438 |
-
},
|
| 439 |
-
{
|
| 440 |
-
"epoch": 0.81,
|
| 441 |
-
"learning_rate": 6.679110534004221e-05,
|
| 442 |
-
"loss": 1.0514,
|
| 443 |
-
"step": 5600
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 0.82,
|
| 447 |
-
"learning_rate": 6.597954877454959e-05,
|
| 448 |
-
"loss": 1.0446,
|
| 449 |
-
"step": 5700
|
| 450 |
-
},
|
| 451 |
-
{
|
| 452 |
-
"epoch": 0.84,
|
| 453 |
-
"learning_rate": 6.516799220905698e-05,
|
| 454 |
-
"loss": 1.0358,
|
| 455 |
-
"step": 5800
|
| 456 |
-
},
|
| 457 |
-
{
|
| 458 |
-
"epoch": 0.85,
|
| 459 |
-
"learning_rate": 6.435643564356436e-05,
|
| 460 |
-
"loss": 1.0364,
|
| 461 |
-
"step": 5900
|
| 462 |
-
},
|
| 463 |
-
{
|
| 464 |
-
"epoch": 0.87,
|
| 465 |
-
"learning_rate": 6.354487907807174e-05,
|
| 466 |
-
"loss": 1.0392,
|
| 467 |
-
"step": 6000
|
| 468 |
-
},
|
| 469 |
-
{
|
| 470 |
-
"epoch": 0.87,
|
| 471 |
-
"eval_loss": 0.2870033383369446,
|
| 472 |
-
"eval_runtime": 295.9814,
|
| 473 |
-
"eval_samples_per_second": 19.569,
|
| 474 |
-
"eval_steps_per_second": 0.307,
|
| 475 |
-
"eval_wer": 0.3997034768157972,
|
| 476 |
-
"step": 6000
|
| 477 |
-
},
|
| 478 |
-
{
|
| 479 |
-
"epoch": 0.88,
|
| 480 |
-
"learning_rate": 6.273332251257913e-05,
|
| 481 |
-
"loss": 1.0375,
|
| 482 |
-
"step": 6100
|
| 483 |
-
},
|
| 484 |
-
{
|
| 485 |
-
"epoch": 0.9,
|
| 486 |
-
"learning_rate": 6.192176594708652e-05,
|
| 487 |
-
"loss": 1.0408,
|
| 488 |
-
"step": 6200
|
| 489 |
-
},
|
| 490 |
-
{
|
| 491 |
-
"epoch": 0.91,
|
| 492 |
-
"learning_rate": 6.11102093815939e-05,
|
| 493 |
-
"loss": 1.0382,
|
| 494 |
-
"step": 6300
|
| 495 |
-
},
|
| 496 |
-
{
|
| 497 |
-
"epoch": 0.93,
|
| 498 |
-
"learning_rate": 6.0298652816101284e-05,
|
| 499 |
-
"loss": 1.0335,
|
| 500 |
-
"step": 6400
|
| 501 |
-
},
|
| 502 |
-
{
|
| 503 |
-
"epoch": 0.94,
|
| 504 |
-
"learning_rate": 5.948709625060867e-05,
|
| 505 |
-
"loss": 1.0178,
|
| 506 |
-
"step": 6500
|
| 507 |
-
},
|
| 508 |
-
{
|
| 509 |
-
"epoch": 0.94,
|
| 510 |
-
"eval_loss": 0.28046590089797974,
|
| 511 |
-
"eval_runtime": 297.8045,
|
| 512 |
-
"eval_samples_per_second": 19.449,
|
| 513 |
-
"eval_steps_per_second": 0.306,
|
| 514 |
-
"eval_wer": 0.39629834794225943,
|
| 515 |
-
"step": 6500
|
| 516 |
-
},
|
| 517 |
-
{
|
| 518 |
-
"epoch": 0.95,
|
| 519 |
-
"learning_rate": 5.867553968511605e-05,
|
| 520 |
-
"loss": 1.0312,
|
| 521 |
-
"step": 6600
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"epoch": 0.97,
|
| 525 |
-
"learning_rate": 5.786398311962344e-05,
|
| 526 |
-
"loss": 1.033,
|
| 527 |
-
"step": 6700
|
| 528 |
-
},
|
| 529 |
-
{
|
| 530 |
-
"epoch": 0.98,
|
| 531 |
-
"learning_rate": 5.7052426554130825e-05,
|
| 532 |
-
"loss": 1.0289,
|
| 533 |
-
"step": 6800
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 1.0,
|
| 537 |
-
"learning_rate": 5.624086998863821e-05,
|
| 538 |
-
"loss": 1.027,
|
| 539 |
-
"step": 6900
|
| 540 |
-
},
|
| 541 |
-
{
|
| 542 |
-
"epoch": 1.01,
|
| 543 |
-
"learning_rate": 5.54293134231456e-05,
|
| 544 |
-
"loss": 0.992,
|
| 545 |
-
"step": 7000
|
| 546 |
-
},
|
| 547 |
-
{
|
| 548 |
-
"epoch": 1.01,
|
| 549 |
-
"eval_loss": 0.2747785747051239,
|
| 550 |
-
"eval_runtime": 298.4841,
|
| 551 |
-
"eval_samples_per_second": 19.405,
|
| 552 |
-
"eval_steps_per_second": 0.305,
|
| 553 |
-
"eval_wer": 0.39352862589201343,
|
| 554 |
-
"step": 7000
|
| 555 |
-
},
|
| 556 |
-
{
|
| 557 |
-
"epoch": 1.03,
|
| 558 |
-
"learning_rate": 5.461775685765298e-05,
|
| 559 |
-
"loss": 1.0025,
|
| 560 |
-
"step": 7100
|
| 561 |
-
},
|
| 562 |
-
{
|
| 563 |
-
"epoch": 1.04,
|
| 564 |
-
"learning_rate": 5.3806200292160366e-05,
|
| 565 |
-
"loss": 1.0122,
|
| 566 |
-
"step": 7200
|
| 567 |
-
},
|
| 568 |
-
{
|
| 569 |
-
"epoch": 1.06,
|
| 570 |
-
"learning_rate": 5.299464372666775e-05,
|
| 571 |
-
"loss": 1.018,
|
| 572 |
-
"step": 7300
|
| 573 |
-
},
|
| 574 |
-
{
|
| 575 |
-
"epoch": 1.07,
|
| 576 |
-
"learning_rate": 5.218308716117514e-05,
|
| 577 |
-
"loss": 0.9936,
|
| 578 |
-
"step": 7400
|
| 579 |
-
},
|
| 580 |
-
{
|
| 581 |
-
"epoch": 1.09,
|
| 582 |
-
"learning_rate": 5.137153059568252e-05,
|
| 583 |
-
"loss": 1.0197,
|
| 584 |
-
"step": 7500
|
| 585 |
-
},
|
| 586 |
-
{
|
| 587 |
-
"epoch": 1.09,
|
| 588 |
-
"eval_loss": 0.26907604932785034,
|
| 589 |
-
"eval_runtime": 298.796,
|
| 590 |
-
"eval_samples_per_second": 19.384,
|
| 591 |
-
"eval_steps_per_second": 0.305,
|
| 592 |
-
"eval_wer": 0.3884453713056796,
|
| 593 |
-
"step": 7500
|
| 594 |
-
},
|
| 595 |
-
{
|
| 596 |
-
"epoch": 1.1,
|
| 597 |
-
"learning_rate": 5.055997403018991e-05,
|
| 598 |
-
"loss": 1.008,
|
| 599 |
-
"step": 7600
|
| 600 |
-
},
|
| 601 |
-
{
|
| 602 |
-
"epoch": 1.11,
|
| 603 |
-
"learning_rate": 4.97484174646973e-05,
|
| 604 |
-
"loss": 1.0028,
|
| 605 |
-
"step": 7700
|
| 606 |
-
},
|
| 607 |
-
{
|
| 608 |
-
"epoch": 1.13,
|
| 609 |
-
"learning_rate": 4.893686089920468e-05,
|
| 610 |
-
"loss": 0.9929,
|
| 611 |
-
"step": 7800
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 1.14,
|
| 615 |
-
"learning_rate": 4.8125304333712064e-05,
|
| 616 |
-
"loss": 0.995,
|
| 617 |
-
"step": 7900
|
| 618 |
-
},
|
| 619 |
-
{
|
| 620 |
-
"epoch": 1.16,
|
| 621 |
-
"learning_rate": 4.731374776821945e-05,
|
| 622 |
-
"loss": 1.0056,
|
| 623 |
-
"step": 8000
|
| 624 |
-
},
|
| 625 |
-
{
|
| 626 |
-
"epoch": 1.16,
|
| 627 |
-
"eval_loss": 0.26817116141319275,
|
| 628 |
-
"eval_runtime": 298.1504,
|
| 629 |
-
"eval_samples_per_second": 19.426,
|
| 630 |
-
"eval_steps_per_second": 0.305,
|
| 631 |
-
"eval_wer": 0.3888526833718922,
|
| 632 |
-
"step": 8000
|
| 633 |
-
},
|
| 634 |
-
{
|
| 635 |
-
"epoch": 1.17,
|
| 636 |
-
"learning_rate": 4.6510306768381754e-05,
|
| 637 |
-
"loss": 0.9971,
|
| 638 |
-
"step": 8100
|
| 639 |
-
},
|
| 640 |
-
{
|
| 641 |
-
"epoch": 1.19,
|
| 642 |
-
"learning_rate": 4.5698750202889145e-05,
|
| 643 |
-
"loss": 0.9976,
|
| 644 |
-
"step": 8200
|
| 645 |
-
},
|
| 646 |
-
{
|
| 647 |
-
"epoch": 1.2,
|
| 648 |
-
"learning_rate": 4.488719363739653e-05,
|
| 649 |
-
"loss": 1.0014,
|
| 650 |
-
"step": 8300
|
| 651 |
-
},
|
| 652 |
-
{
|
| 653 |
-
"epoch": 1.22,
|
| 654 |
-
"learning_rate": 4.407563707190391e-05,
|
| 655 |
-
"loss": 0.9835,
|
| 656 |
-
"step": 8400
|
| 657 |
-
},
|
| 658 |
-
{
|
| 659 |
-
"epoch": 1.23,
|
| 660 |
-
"learning_rate": 4.3272196072066225e-05,
|
| 661 |
-
"loss": 0.9826,
|
| 662 |
-
"step": 8500
|
| 663 |
-
},
|
| 664 |
-
{
|
| 665 |
-
"epoch": 1.23,
|
| 666 |
-
"eval_loss": 0.26473307609558105,
|
| 667 |
-
"eval_runtime": 299.0161,
|
| 668 |
-
"eval_samples_per_second": 19.37,
|
| 669 |
-
"eval_steps_per_second": 0.304,
|
| 670 |
-
"eval_wer": 0.38675095311023494,
|
| 671 |
-
"step": 8500
|
| 672 |
-
},
|
| 673 |
-
{
|
| 674 |
-
"epoch": 1.24,
|
| 675 |
-
"learning_rate": 4.2460639506573615e-05,
|
| 676 |
-
"loss": 0.9838,
|
| 677 |
-
"step": 8600
|
| 678 |
-
},
|
| 679 |
-
{
|
| 680 |
-
"epoch": 1.26,
|
| 681 |
-
"learning_rate": 4.1649082941081e-05,
|
| 682 |
-
"loss": 0.9836,
|
| 683 |
-
"step": 8700
|
| 684 |
-
},
|
| 685 |
-
{
|
| 686 |
-
"epoch": 1.27,
|
| 687 |
-
"learning_rate": 4.0845641941243305e-05,
|
| 688 |
-
"loss": 0.9824,
|
| 689 |
-
"step": 8800
|
| 690 |
-
},
|
| 691 |
-
{
|
| 692 |
-
"epoch": 1.29,
|
| 693 |
-
"learning_rate": 4.003408537575069e-05,
|
| 694 |
-
"loss": 0.9715,
|
| 695 |
-
"step": 8900
|
| 696 |
-
},
|
| 697 |
-
{
|
| 698 |
-
"epoch": 1.3,
|
| 699 |
-
"learning_rate": 3.922252881025807e-05,
|
| 700 |
-
"loss": 0.9815,
|
| 701 |
-
"step": 9000
|
| 702 |
-
},
|
| 703 |
-
{
|
| 704 |
-
"epoch": 1.3,
|
| 705 |
-
"eval_loss": 0.26034271717071533,
|
| 706 |
-
"eval_runtime": 299.6782,
|
| 707 |
-
"eval_samples_per_second": 19.327,
|
| 708 |
-
"eval_steps_per_second": 0.304,
|
| 709 |
-
"eval_wer": 0.3831828994102121,
|
| 710 |
-
"step": 9000
|
| 711 |
-
},
|
| 712 |
-
{
|
| 713 |
-
"epoch": 1.32,
|
| 714 |
-
"learning_rate": 3.841097224476546e-05,
|
| 715 |
-
"loss": 0.9757,
|
| 716 |
-
"step": 9100
|
| 717 |
-
},
|
| 718 |
-
{
|
| 719 |
-
"epoch": 1.33,
|
| 720 |
-
"learning_rate": 3.7599415679272846e-05,
|
| 721 |
-
"loss": 0.9689,
|
| 722 |
-
"step": 9200
|
| 723 |
-
},
|
| 724 |
-
{
|
| 725 |
-
"epoch": 1.35,
|
| 726 |
-
"learning_rate": 3.678785911378023e-05,
|
| 727 |
-
"loss": 0.9778,
|
| 728 |
-
"step": 9300
|
| 729 |
-
},
|
| 730 |
-
{
|
| 731 |
-
"epoch": 1.36,
|
| 732 |
-
"learning_rate": 3.5976302548287614e-05,
|
| 733 |
-
"loss": 0.9794,
|
| 734 |
-
"step": 9400
|
| 735 |
-
},
|
| 736 |
-
{
|
| 737 |
-
"epoch": 1.37,
|
| 738 |
-
"learning_rate": 3.5164745982795e-05,
|
| 739 |
-
"loss": 0.9717,
|
| 740 |
-
"step": 9500
|
| 741 |
-
},
|
| 742 |
-
{
|
| 743 |
-
"epoch": 1.37,
|
| 744 |
-
"eval_loss": 0.25609487295150757,
|
| 745 |
-
"eval_runtime": 299.6976,
|
| 746 |
-
"eval_samples_per_second": 19.326,
|
| 747 |
-
"eval_steps_per_second": 0.304,
|
| 748 |
-
"eval_wer": 0.3807064420476392,
|
| 749 |
-
"step": 9500
|
| 750 |
-
},
|
| 751 |
-
{
|
| 752 |
-
"epoch": 1.39,
|
| 753 |
-
"learning_rate": 3.435318941730239e-05,
|
| 754 |
-
"loss": 0.9752,
|
| 755 |
-
"step": 9600
|
| 756 |
-
},
|
| 757 |
-
{
|
| 758 |
-
"epoch": 1.4,
|
| 759 |
-
"learning_rate": 3.354163285180977e-05,
|
| 760 |
-
"loss": 0.965,
|
| 761 |
-
"step": 9700
|
| 762 |
-
},
|
| 763 |
-
{
|
| 764 |
-
"epoch": 1.42,
|
| 765 |
-
"learning_rate": 3.2730076286317155e-05,
|
| 766 |
-
"loss": 0.9522,
|
| 767 |
-
"step": 9800
|
| 768 |
-
},
|
| 769 |
-
{
|
| 770 |
-
"epoch": 1.43,
|
| 771 |
-
"learning_rate": 3.191851972082454e-05,
|
| 772 |
-
"loss": 0.9718,
|
| 773 |
-
"step": 9900
|
| 774 |
-
},
|
| 775 |
-
{
|
| 776 |
-
"epoch": 1.45,
|
| 777 |
-
"learning_rate": 3.110696315533193e-05,
|
| 778 |
-
"loss": 0.9605,
|
| 779 |
-
"step": 10000
|
| 780 |
-
},
|
| 781 |
-
{
|
| 782 |
-
"epoch": 1.45,
|
| 783 |
-
"eval_loss": 0.25231894850730896,
|
| 784 |
-
"eval_runtime": 297.5796,
|
| 785 |
-
"eval_samples_per_second": 19.464,
|
| 786 |
-
"eval_steps_per_second": 0.306,
|
| 787 |
-
"eval_wer": 0.3782951546156603,
|
| 788 |
-
"step": 10000
|
| 789 |
-
},
|
| 790 |
-
{
|
| 791 |
-
"epoch": 1.46,
|
| 792 |
-
"learning_rate": 3.0295406589839315e-05,
|
| 793 |
-
"loss": 0.9635,
|
| 794 |
-
"step": 10100
|
| 795 |
-
},
|
| 796 |
-
{
|
| 797 |
-
"epoch": 1.48,
|
| 798 |
-
"learning_rate": 2.94838500243467e-05,
|
| 799 |
-
"loss": 0.9632,
|
| 800 |
-
"step": 10200
|
| 801 |
-
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 1.49,
|
| 804 |
-
"learning_rate": 2.8672293458854082e-05,
|
| 805 |
-
"loss": 0.9548,
|
| 806 |
-
"step": 10300
|
| 807 |
-
},
|
| 808 |
-
{
|
| 809 |
-
"epoch": 1.5,
|
| 810 |
-
"learning_rate": 2.7868852459016392e-05,
|
| 811 |
-
"loss": 0.9554,
|
| 812 |
-
"step": 10400
|
| 813 |
-
},
|
| 814 |
-
{
|
| 815 |
-
"epoch": 1.52,
|
| 816 |
-
"learning_rate": 2.7057295893523783e-05,
|
| 817 |
-
"loss": 0.96,
|
| 818 |
-
"step": 10500
|
| 819 |
-
},
|
| 820 |
-
{
|
| 821 |
-
"epoch": 1.52,
|
| 822 |
-
"eval_loss": 0.24939315021038055,
|
| 823 |
-
"eval_runtime": 300.036,
|
| 824 |
-
"eval_samples_per_second": 19.304,
|
| 825 |
-
"eval_steps_per_second": 0.303,
|
| 826 |
-
"eval_wer": 0.3787513441298185,
|
| 827 |
-
"step": 10500
|
| 828 |
-
},
|
| 829 |
-
{
|
| 830 |
-
"epoch": 1.53,
|
| 831 |
-
"learning_rate": 2.6245739328031166e-05,
|
| 832 |
-
"loss": 0.9611,
|
| 833 |
-
"step": 10600
|
| 834 |
-
},
|
| 835 |
-
{
|
| 836 |
-
"epoch": 1.55,
|
| 837 |
-
"learning_rate": 2.543418276253855e-05,
|
| 838 |
-
"loss": 0.9594,
|
| 839 |
-
"step": 10700
|
| 840 |
-
},
|
| 841 |
-
{
|
| 842 |
-
"epoch": 1.56,
|
| 843 |
-
"learning_rate": 2.4622626197045937e-05,
|
| 844 |
-
"loss": 0.9589,
|
| 845 |
-
"step": 10800
|
| 846 |
-
},
|
| 847 |
-
{
|
| 848 |
-
"epoch": 1.58,
|
| 849 |
-
"learning_rate": 2.381106963155332e-05,
|
| 850 |
-
"loss": 0.9441,
|
| 851 |
-
"step": 10900
|
| 852 |
-
},
|
| 853 |
-
{
|
| 854 |
-
"epoch": 1.59,
|
| 855 |
-
"learning_rate": 2.2999513066060704e-05,
|
| 856 |
-
"loss": 0.9442,
|
| 857 |
-
"step": 11000
|
| 858 |
-
},
|
| 859 |
-
{
|
| 860 |
-
"epoch": 1.59,
|
| 861 |
-
"eval_loss": 0.24783751368522644,
|
| 862 |
-
"eval_runtime": 297.7741,
|
| 863 |
-
"eval_samples_per_second": 19.451,
|
| 864 |
-
"eval_steps_per_second": 0.306,
|
| 865 |
-
"eval_wer": 0.3760142070448695,
|
| 866 |
-
"step": 11000
|
| 867 |
-
},
|
| 868 |
-
{
|
| 869 |
-
"epoch": 1.61,
|
| 870 |
-
"learning_rate": 2.218795650056809e-05,
|
| 871 |
-
"loss": 0.9496,
|
| 872 |
-
"step": 11100
|
| 873 |
-
},
|
| 874 |
-
{
|
| 875 |
-
"epoch": 1.62,
|
| 876 |
-
"learning_rate": 2.1376399935075474e-05,
|
| 877 |
-
"loss": 0.9486,
|
| 878 |
-
"step": 11200
|
| 879 |
-
},
|
| 880 |
-
{
|
| 881 |
-
"epoch": 1.64,
|
| 882 |
-
"learning_rate": 2.056484336958286e-05,
|
| 883 |
-
"loss": 0.9558,
|
| 884 |
-
"step": 11300
|
| 885 |
-
},
|
| 886 |
-
{
|
| 887 |
-
"epoch": 1.65,
|
| 888 |
-
"learning_rate": 1.9753286804090245e-05,
|
| 889 |
-
"loss": 0.9486,
|
| 890 |
-
"step": 11400
|
| 891 |
-
},
|
| 892 |
-
{
|
| 893 |
-
"epoch": 1.66,
|
| 894 |
-
"learning_rate": 1.8941730238597632e-05,
|
| 895 |
-
"loss": 0.9564,
|
| 896 |
-
"step": 11500
|
| 897 |
-
},
|
| 898 |
-
{
|
| 899 |
-
"epoch": 1.66,
|
| 900 |
-
"eval_loss": 0.2454409897327423,
|
| 901 |
-
"eval_runtime": 296.0265,
|
| 902 |
-
"eval_samples_per_second": 19.566,
|
| 903 |
-
"eval_steps_per_second": 0.307,
|
| 904 |
-
"eval_wer": 0.3733096549252175,
|
| 905 |
-
"step": 11500
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"epoch": 1.68,
|
| 909 |
-
"learning_rate": 1.8130173673105015e-05,
|
| 910 |
-
"loss": 0.9427,
|
| 911 |
-
"step": 11600
|
| 912 |
-
},
|
| 913 |
-
{
|
| 914 |
-
"epoch": 1.69,
|
| 915 |
-
"learning_rate": 1.73186171076124e-05,
|
| 916 |
-
"loss": 0.9423,
|
| 917 |
-
"step": 11700
|
| 918 |
-
},
|
| 919 |
-
{
|
| 920 |
-
"epoch": 1.71,
|
| 921 |
-
"learning_rate": 1.6507060542119786e-05,
|
| 922 |
-
"loss": 0.9503,
|
| 923 |
-
"step": 11800
|
| 924 |
-
},
|
| 925 |
-
{
|
| 926 |
-
"epoch": 1.72,
|
| 927 |
-
"learning_rate": 1.5695503976627173e-05,
|
| 928 |
-
"loss": 0.9383,
|
| 929 |
-
"step": 11900
|
| 930 |
-
},
|
| 931 |
-
{
|
| 932 |
-
"epoch": 1.74,
|
| 933 |
-
"learning_rate": 1.4883947411134558e-05,
|
| 934 |
-
"loss": 0.9436,
|
| 935 |
-
"step": 12000
|
| 936 |
-
},
|
| 937 |
-
{
|
| 938 |
-
"epoch": 1.74,
|
| 939 |
-
"eval_loss": 0.24390804767608643,
|
| 940 |
-
"eval_runtime": 295.4584,
|
| 941 |
-
"eval_samples_per_second": 19.603,
|
| 942 |
-
"eval_steps_per_second": 0.308,
|
| 943 |
-
"eval_wer": 0.37467822346769203,
|
| 944 |
-
"step": 12000
|
| 945 |
-
},
|
| 946 |
-
{
|
| 947 |
-
"epoch": 1.75,
|
| 948 |
-
"learning_rate": 1.4072390845641942e-05,
|
| 949 |
-
"loss": 0.9491,
|
| 950 |
-
"step": 12100
|
| 951 |
-
},
|
| 952 |
-
{
|
| 953 |
-
"epoch": 1.77,
|
| 954 |
-
"learning_rate": 1.3260834280149325e-05,
|
| 955 |
-
"loss": 0.9419,
|
| 956 |
-
"step": 12200
|
| 957 |
-
},
|
| 958 |
-
{
|
| 959 |
-
"epoch": 1.78,
|
| 960 |
-
"learning_rate": 1.2449277714656712e-05,
|
| 961 |
-
"loss": 0.9517,
|
| 962 |
-
"step": 12300
|
| 963 |
-
},
|
| 964 |
-
{
|
| 965 |
-
"epoch": 1.79,
|
| 966 |
-
"learning_rate": 1.1637721149164097e-05,
|
| 967 |
-
"loss": 0.9367,
|
| 968 |
-
"step": 12400
|
| 969 |
-
},
|
| 970 |
-
{
|
| 971 |
-
"epoch": 1.81,
|
| 972 |
-
"learning_rate": 1.0826164583671483e-05,
|
| 973 |
-
"loss": 0.938,
|
| 974 |
-
"step": 12500
|
| 975 |
-
},
|
| 976 |
-
{
|
| 977 |
-
"epoch": 1.81,
|
| 978 |
-
"eval_loss": 0.24111612141132355,
|
| 979 |
-
"eval_runtime": 296.9314,
|
| 980 |
-
"eval_samples_per_second": 19.506,
|
| 981 |
-
"eval_steps_per_second": 0.306,
|
| 982 |
-
"eval_wer": 0.37159894424712436,
|
| 983 |
-
"step": 12500
|
| 984 |
-
},
|
| 985 |
-
{
|
| 986 |
-
"epoch": 1.82,
|
| 987 |
-
"learning_rate": 1.0014608018178868e-05,
|
| 988 |
-
"loss": 0.9337,
|
| 989 |
-
"step": 12600
|
| 990 |
-
},
|
| 991 |
-
{
|
| 992 |
-
"epoch": 1.84,
|
| 993 |
-
"learning_rate": 9.203051452686253e-06,
|
| 994 |
-
"loss": 0.9284,
|
| 995 |
-
"step": 12700
|
| 996 |
-
},
|
| 997 |
-
{
|
| 998 |
-
"epoch": 1.85,
|
| 999 |
-
"learning_rate": 8.391494887193638e-06,
|
| 1000 |
-
"loss": 0.938,
|
| 1001 |
-
"step": 12800
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"epoch": 1.87,
|
| 1005 |
-
"learning_rate": 7.579938321701023e-06,
|
| 1006 |
-
"loss": 0.9365,
|
| 1007 |
-
"step": 12900
|
| 1008 |
-
},
|
| 1009 |
-
{
|
| 1010 |
-
"epoch": 1.88,
|
| 1011 |
-
"learning_rate": 6.768381756208409e-06,
|
| 1012 |
-
"loss": 0.9353,
|
| 1013 |
-
"step": 13000
|
| 1014 |
-
},
|
| 1015 |
-
{
|
| 1016 |
-
"epoch": 1.88,
|
| 1017 |
-
"eval_loss": 0.23965783417224884,
|
| 1018 |
-
"eval_runtime": 296.6078,
|
| 1019 |
-
"eval_samples_per_second": 19.527,
|
| 1020 |
-
"eval_steps_per_second": 0.307,
|
| 1021 |
-
"eval_wer": 0.3697904786731402,
|
| 1022 |
-
"step": 13000
|
| 1023 |
-
},
|
| 1024 |
-
{
|
| 1025 |
-
"epoch": 1.9,
|
| 1026 |
-
"learning_rate": 5.956825190715793e-06,
|
| 1027 |
-
"loss": 0.9413,
|
| 1028 |
-
"step": 13100
|
| 1029 |
-
},
|
| 1030 |
-
{
|
| 1031 |
-
"epoch": 1.91,
|
| 1032 |
-
"learning_rate": 5.153384190878104e-06,
|
| 1033 |
-
"loss": 0.9356,
|
| 1034 |
-
"step": 13200
|
| 1035 |
-
},
|
| 1036 |
-
{
|
| 1037 |
-
"epoch": 1.92,
|
| 1038 |
-
"learning_rate": 4.34182762538549e-06,
|
| 1039 |
-
"loss": 0.9209,
|
| 1040 |
-
"step": 13300
|
| 1041 |
-
},
|
| 1042 |
-
{
|
| 1043 |
-
"epoch": 1.94,
|
| 1044 |
-
"learning_rate": 3.530271059892875e-06,
|
| 1045 |
-
"loss": 0.9362,
|
| 1046 |
-
"step": 13400
|
| 1047 |
-
},
|
| 1048 |
-
{
|
| 1049 |
-
"epoch": 1.95,
|
| 1050 |
-
"learning_rate": 2.71871449440026e-06,
|
| 1051 |
-
"loss": 0.9271,
|
| 1052 |
-
"step": 13500
|
| 1053 |
-
},
|
| 1054 |
-
{
|
| 1055 |
-
"epoch": 1.95,
|
| 1056 |
-
"eval_loss": 0.23875188827514648,
|
| 1057 |
-
"eval_runtime": 296.9414,
|
| 1058 |
-
"eval_samples_per_second": 19.506,
|
| 1059 |
-
"eval_steps_per_second": 0.306,
|
| 1060 |
-
"eval_wer": 0.3680797679950471,
|
| 1061 |
-
"step": 13500
|
| 1062 |
-
},
|
| 1063 |
-
{
|
| 1064 |
-
"epoch": 1.97,
|
| 1065 |
-
"learning_rate": 1.907157928907645e-06,
|
| 1066 |
-
"loss": 0.9288,
|
| 1067 |
-
"step": 13600
|
| 1068 |
-
},
|
| 1069 |
-
{
|
| 1070 |
-
"epoch": 1.98,
|
| 1071 |
-
"learning_rate": 1.0956013634150302e-06,
|
| 1072 |
-
"loss": 0.9345,
|
| 1073 |
-
"step": 13700
|
| 1074 |
-
},
|
| 1075 |
-
{
|
| 1076 |
-
"epoch": 2.0,
|
| 1077 |
-
"learning_rate": 2.840447979224152e-07,
|
| 1078 |
-
"loss": 0.9326,
|
| 1079 |
-
"step": 13800
|
| 1080 |
-
},
|
| 1081 |
-
{
|
| 1082 |
-
"epoch": 2.0,
|
| 1083 |
-
"step": 13822,
|
| 1084 |
-
"total_flos": 1.2600843645735263e+20,
|
| 1085 |
-
"train_loss": 1.442369053426242,
|
| 1086 |
-
"train_runtime": 53680.5392,
|
| 1087 |
-
"train_samples_per_second": 16.478,
|
| 1088 |
-
"train_steps_per_second": 0.257
|
| 1089 |
-
}
|
| 1090 |
-
],
|
| 1091 |
-
"max_steps": 13822,
|
| 1092 |
-
"num_train_epochs": 2,
|
| 1093 |
-
"total_flos": 1.2600843645735263e+20,
|
| 1094 |
-
"trial_name": null,
|
| 1095 |
-
"trial_params": null
|
| 1096 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
validate.sh
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
python eval.py --model_id="./" --dataset="mozilla-foundation/common_voice_8_0" --config="fr" --split="test" --log_outputs
|
| 2 |
+
python eval.py --model_id="./" --dataset="speech-recognition-community-v2/dev_data" --config="fr" \
|
| 3 |
+
--split="validation" --chunk_length_s="5.0" --stride_length_s="1.0" --log_outputs
|
wandb/run-20220201_142447-3hqii6h6/files/config.yaml
CHANGED
|
@@ -45,7 +45,13 @@ _wandb:
|
|
| 45 |
- 1
|
| 46 |
- 5
|
| 47 |
- 11
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
3:
|
|
|
|
|
|
|
| 49 |
- 13
|
| 50 |
4: 3.8.8
|
| 51 |
5: 0.12.9
|
|
|
|
| 45 |
- 1
|
| 46 |
- 5
|
| 47 |
- 11
|
| 48 |
+
2:
|
| 49 |
+
- 1
|
| 50 |
+
- 5
|
| 51 |
+
- 11
|
| 52 |
3:
|
| 53 |
+
- 1
|
| 54 |
+
- 7
|
| 55 |
- 13
|
| 56 |
4: 3.8.8
|
| 57 |
5: 0.12.9
|
wandb/run-20220201_142447-3hqii6h6/files/output.log
CHANGED
|
@@ -14,3 +14,10 @@ Configuration saved in ./preprocessor_config.json
|
|
| 14 |
a37375f..4641f67 main -> main
|
| 15 |
Upload file training_args.bin: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.98k/2.98k [00:00<?, ?B/s]To https://huggingface.co/AlexN/xls-r-300m-fr
|
| 16 |
a37375f..4641f67 main -> main
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
a37375f..4641f67 main -> main
|
| 15 |
Upload file training_args.bin: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.98k/2.98k [00:00<?, ?B/s]To https://huggingface.co/AlexN/xls-r-300m-fr
|
| 16 |
a37375f..4641f67 main -> main
|
| 17 |
+
Upload file training_args.bin: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.98k/2.98k [00:01<?, ?B/s]
|
| 18 |
+
Dropping the following result as it does not have all the necessary fields:
|
| 19 |
+
{'dataset': {'name': 'MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - FR', 'type': 'common_voice', 'args': 'Config: fr, Training split: train+validation, Eval split: test'}}
|
| 20 |
+
02/01/2022 14:26:17 - WARNING - huggingface_hub.repository - To https://huggingface.co/AlexN/xls-r-300m-fr
|
| 21 |
+
4641f67..cbeda52 main -> main
|
| 22 |
+
To https://huggingface.co/AlexN/xls-r-300m-fr
|
| 23 |
+
4641f67..cbeda52 main -> main
|
wandb/run-20220201_142447-3hqii6h6/files/wandb-summary.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"eval/loss": 0.22597630321979523, "eval/wer": 0.2948882012145466, "eval/runtime": 276.6037, "eval/samples_per_second": 16.157, "eval/steps_per_second": 0.253, "train/global_step": 0, "_runtime": 2, "_timestamp": 1643725490, "_step": 0}
|
|
|
|
| 1 |
+
{"eval/loss": 0.22597630321979523, "eval/wer": 0.2948882012145466, "eval/runtime": 276.6037, "eval/samples_per_second": 16.157, "eval/steps_per_second": 0.253, "train/global_step": 0, "_runtime": 2, "_timestamp": 1643725490, "_step": 0, "_wandb": {"runtime": 94}}
|
wandb/run-20220201_142447-3hqii6h6/logs/debug-internal.log
CHANGED
|
@@ -57,3 +57,116 @@
|
|
| 57 |
2022-02-01 14:26:05,830 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: stop_status
|
| 58 |
2022-02-01 14:26:05,830 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: stop_status
|
| 59 |
2022-02-01 14:26:12,326 INFO Thread-8 :29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
2022-02-01 14:26:05,830 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: stop_status
|
| 58 |
2022-02-01 14:26:05,830 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: stop_status
|
| 59 |
2022-02-01 14:26:12,326 INFO Thread-8 :29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log
|
| 60 |
+
2022-02-01 14:26:16,328 INFO Thread-8 :29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log
|
| 61 |
+
2022-02-01 14:26:18,207 DEBUG SenderThread:29947 [sender.py:send():234] send: stats
|
| 62 |
+
2022-02-01 14:26:18,330 INFO Thread-8 :29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log
|
| 63 |
+
2022-02-01 14:26:23,406 DEBUG SenderThread:29947 [sender.py:send():234] send: telemetry
|
| 64 |
+
2022-02-01 14:26:23,407 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 65 |
+
2022-02-01 14:26:23,407 DEBUG SenderThread:29947 [sender.py:send():234] send: exit
|
| 66 |
+
2022-02-01 14:26:23,407 INFO SenderThread:29947 [sender.py:send_exit():366] handling exit code: 0
|
| 67 |
+
2022-02-01 14:26:23,407 INFO SenderThread:29947 [sender.py:send_exit():368] handling runtime: 94
|
| 68 |
+
2022-02-01 14:26:23,411 INFO SenderThread:29947 [sender.py:_save_file():939] saving file wandb-summary.json with policy end
|
| 69 |
+
2022-02-01 14:26:23,411 INFO SenderThread:29947 [sender.py:send_exit():374] send defer
|
| 70 |
+
2022-02-01 14:26:23,412 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 71 |
+
2022-02-01 14:26:23,412 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 72 |
+
2022-02-01 14:26:23,413 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 0
|
| 73 |
+
2022-02-01 14:26:23,413 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 74 |
+
2022-02-01 14:26:23,413 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 0
|
| 75 |
+
2022-02-01 14:26:23,413 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 1
|
| 76 |
+
2022-02-01 14:26:23,414 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 77 |
+
2022-02-01 14:26:23,414 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 1
|
| 78 |
+
2022-02-01 14:26:23,486 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 79 |
+
2022-02-01 14:26:23,486 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 1
|
| 80 |
+
2022-02-01 14:26:23,486 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 2
|
| 81 |
+
2022-02-01 14:26:23,486 DEBUG SenderThread:29947 [sender.py:send():234] send: stats
|
| 82 |
+
2022-02-01 14:26:23,487 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 83 |
+
2022-02-01 14:26:23,487 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 2
|
| 84 |
+
2022-02-01 14:26:23,488 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 85 |
+
2022-02-01 14:26:23,488 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 2
|
| 86 |
+
2022-02-01 14:26:23,488 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 3
|
| 87 |
+
2022-02-01 14:26:23,488 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 88 |
+
2022-02-01 14:26:23,488 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 3
|
| 89 |
+
2022-02-01 14:26:23,489 DEBUG SenderThread:29947 [sender.py:send():234] send: summary
|
| 90 |
+
2022-02-01 14:26:23,490 INFO SenderThread:29947 [sender.py:_save_file():939] saving file wandb-summary.json with policy end
|
| 91 |
+
2022-02-01 14:26:23,490 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 92 |
+
2022-02-01 14:26:23,490 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 3
|
| 93 |
+
2022-02-01 14:26:23,490 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 4
|
| 94 |
+
2022-02-01 14:26:23,491 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 95 |
+
2022-02-01 14:26:23,491 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 4
|
| 96 |
+
2022-02-01 14:26:23,491 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 97 |
+
2022-02-01 14:26:23,491 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 4
|
| 98 |
+
2022-02-01 14:26:23,515 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 99 |
+
2022-02-01 14:26:23,724 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 5
|
| 100 |
+
2022-02-01 14:26:23,724 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 101 |
+
2022-02-01 14:26:23,726 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 102 |
+
2022-02-01 14:26:23,726 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 5
|
| 103 |
+
2022-02-01 14:26:23,726 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 104 |
+
2022-02-01 14:26:23,726 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 5
|
| 105 |
+
2022-02-01 14:26:23,726 INFO SenderThread:29947 [dir_watcher.py:finish():283] shutting down directory watcher
|
| 106 |
+
2022-02-01 14:26:23,827 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 107 |
+
2022-02-01 14:26:24,332 INFO SenderThread:29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log
|
| 108 |
+
2022-02-01 14:26:24,333 INFO SenderThread:29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/wandb-summary.json
|
| 109 |
+
2022-02-01 14:26:24,334 INFO SenderThread:29947 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/config.yaml
|
| 110 |
+
2022-02-01 14:26:24,334 INFO SenderThread:29947 [dir_watcher.py:finish():313] scan: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files
|
| 111 |
+
2022-02-01 14:26:24,334 INFO SenderThread:29947 [dir_watcher.py:finish():327] scan save: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log output.log
|
| 112 |
+
2022-02-01 14:26:24,335 INFO SenderThread:29947 [dir_watcher.py:finish():327] scan save: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/wandb-metadata.json wandb-metadata.json
|
| 113 |
+
2022-02-01 14:26:24,335 INFO SenderThread:29947 [dir_watcher.py:finish():327] scan save: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/config.yaml config.yaml
|
| 114 |
+
2022-02-01 14:26:24,341 INFO SenderThread:29947 [dir_watcher.py:finish():327] scan save: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/conda-environment.yaml conda-environment.yaml
|
| 115 |
+
2022-02-01 14:26:24,348 INFO SenderThread:29947 [dir_watcher.py:finish():327] scan save: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/wandb-summary.json wandb-summary.json
|
| 116 |
+
2022-02-01 14:26:24,348 INFO SenderThread:29947 [dir_watcher.py:finish():327] scan save: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/requirements.txt requirements.txt
|
| 117 |
+
2022-02-01 14:26:24,349 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 6
|
| 118 |
+
2022-02-01 14:26:24,350 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 119 |
+
2022-02-01 14:26:24,357 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 120 |
+
2022-02-01 14:26:24,357 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 6
|
| 121 |
+
2022-02-01 14:26:24,357 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 122 |
+
2022-02-01 14:26:24,358 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 6
|
| 123 |
+
2022-02-01 14:26:24,358 INFO SenderThread:29947 [file_pusher.py:finish():177] shutting down file pusher
|
| 124 |
+
2022-02-01 14:26:24,452 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 125 |
+
2022-02-01 14:26:24,453 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 126 |
+
2022-02-01 14:26:24,555 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 127 |
+
2022-02-01 14:26:24,560 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 128 |
+
2022-02-01 14:26:24,667 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 129 |
+
2022-02-01 14:26:24,668 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 130 |
+
2022-02-01 14:26:24,770 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 131 |
+
2022-02-01 14:26:24,771 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 132 |
+
2022-02-01 14:26:24,873 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 133 |
+
2022-02-01 14:26:24,874 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 134 |
+
2022-02-01 14:26:24,888 INFO Thread-12 :29947 [upload_job.py:push():137] Uploaded file /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/output.log
|
| 135 |
+
2022-02-01 14:26:24,897 INFO Thread-13 :29947 [upload_job.py:push():137] Uploaded file /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/config.yaml
|
| 136 |
+
2022-02-01 14:26:24,913 INFO Thread-15 :29947 [upload_job.py:push():137] Uploaded file /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/requirements.txt
|
| 137 |
+
2022-02-01 14:26:24,919 INFO Thread-14 :29947 [upload_job.py:push():137] Uploaded file /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/files/wandb-summary.json
|
| 138 |
+
2022-02-01 14:26:24,976 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 139 |
+
2022-02-01 14:26:24,976 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 140 |
+
2022-02-01 14:26:25,079 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 141 |
+
2022-02-01 14:26:25,079 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 142 |
+
2022-02-01 14:26:25,120 INFO Thread-7 :29947 [sender.py:transition_state():387] send defer: 7
|
| 143 |
+
2022-02-01 14:26:25,122 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 144 |
+
2022-02-01 14:26:25,122 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 7
|
| 145 |
+
2022-02-01 14:26:25,122 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 146 |
+
2022-02-01 14:26:25,122 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 7
|
| 147 |
+
2022-02-01 14:26:25,182 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 148 |
+
2022-02-01 14:26:25,552 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 8
|
| 149 |
+
2022-02-01 14:26:25,553 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 150 |
+
2022-02-01 14:26:25,554 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 151 |
+
2022-02-01 14:26:25,554 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 8
|
| 152 |
+
2022-02-01 14:26:25,555 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 153 |
+
2022-02-01 14:26:25,555 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 8
|
| 154 |
+
2022-02-01 14:26:25,555 INFO SenderThread:29947 [sender.py:transition_state():387] send defer: 9
|
| 155 |
+
2022-02-01 14:26:25,556 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: defer
|
| 156 |
+
2022-02-01 14:26:25,557 INFO HandlerThread:29947 [handler.py:handle_request_defer():147] handle defer: 9
|
| 157 |
+
2022-02-01 14:26:25,557 DEBUG SenderThread:29947 [sender.py:send():234] send: final
|
| 158 |
+
2022-02-01 14:26:25,557 DEBUG SenderThread:29947 [sender.py:send():234] send: footer
|
| 159 |
+
2022-02-01 14:26:25,557 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: defer
|
| 160 |
+
2022-02-01 14:26:25,557 INFO SenderThread:29947 [sender.py:send_request_defer():383] handle sender defer: 9
|
| 161 |
+
2022-02-01 14:26:25,656 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: poll_exit
|
| 162 |
+
2022-02-01 14:26:25,656 DEBUG SenderThread:29947 [sender.py:send_request():248] send_request: poll_exit
|
| 163 |
+
2022-02-01 14:26:25,657 INFO SenderThread:29947 [file_pusher.py:join():182] waiting for file pusher
|
| 164 |
+
2022-02-01 14:26:25,966 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: get_summary
|
| 165 |
+
2022-02-01 14:26:25,968 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: sampled_history
|
| 166 |
+
2022-02-01 14:26:25,970 DEBUG HandlerThread:29947 [handler.py:handle_request():130] handle_request: shutdown
|
| 167 |
+
2022-02-01 14:26:25,970 INFO HandlerThread:29947 [handler.py:finish():731] shutting down handler
|
| 168 |
+
2022-02-01 14:26:26,557 INFO WriterThread:29947 [datastore.py:close():281] close: /workspace/xls-r-300m-fr/wandb/run-20220201_142447-3hqii6h6/run-3hqii6h6.wandb
|
| 169 |
+
2022-02-01 14:26:26,963 INFO SenderThread:29947 [sender.py:finish():1070] shutting down sender
|
| 170 |
+
2022-02-01 14:26:26,964 INFO SenderThread:29947 [file_pusher.py:finish():177] shutting down file pusher
|
| 171 |
+
2022-02-01 14:26:26,964 INFO SenderThread:29947 [file_pusher.py:join():182] waiting for file pusher
|
| 172 |
+
2022-02-01 14:26:26,967 INFO MainThread:29947 [internal.py:handle_exit():77] Internal process exited
|
wandb/run-20220201_142447-3hqii6h6/logs/debug.log
CHANGED
|
@@ -22,3 +22,109 @@ config: {}
|
|
| 22 |
2022-02-01 14:24:50,018 INFO MainThread:27520 [wandb_init.py:init():633] run started, returning control to user process
|
| 23 |
2022-02-01 14:24:50,021 INFO MainThread:27520 [wandb_run.py:_config_callback():956] config_cb None None {'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': 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': ['Wav2Vec2ForCTC'], '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': 218, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './checkpoint-18000', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'model_type': 'wav2vec2', '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': 0.05, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 219, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.75, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.25, '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': 'mean', 'ctc_zero_infinity': False, 'add_adapter': False, '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, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 64, 'per_device_eval_batch_size': 64, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': 'None', 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 1.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 2700, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Feb01_14-18-51_job-1abccd0a-3293-4ffe-8274-9e8f841f653f', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 100, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 2, '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': 500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': 'xls-r-300m-fr', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': True, 'metric_for_best_model': 'loss', 'greater_is_better': False, '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': "['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': '', 'train_batch_size': 64, 'eval_batch_size': 64}
|
| 24 |
2022-02-01 14:24:50,026 INFO MainThread:27520 [wandb_watch.py:watch():43] Watching
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
2022-02-01 14:24:50,018 INFO MainThread:27520 [wandb_init.py:init():633] run started, returning control to user process
|
| 23 |
2022-02-01 14:24:50,021 INFO MainThread:27520 [wandb_run.py:_config_callback():956] config_cb None None {'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': 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': ['Wav2Vec2ForCTC'], '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': 218, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './checkpoint-18000', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'model_type': 'wav2vec2', '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': 0.05, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 219, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.75, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.25, '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': 'mean', 'ctc_zero_infinity': False, 'add_adapter': False, '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, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 64, 'per_device_eval_batch_size': 64, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': 'None', 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 1.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 2700, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Feb01_14-18-51_job-1abccd0a-3293-4ffe-8274-9e8f841f653f', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 100, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 2, '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': 500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': 'xls-r-300m-fr', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': True, 'metric_for_best_model': 'loss', 'greater_is_better': False, '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': "['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': '', 'train_batch_size': 64, 'eval_batch_size': 64}
|
| 24 |
2022-02-01 14:24:50,026 INFO MainThread:27520 [wandb_watch.py:watch():43] Watching
|
| 25 |
+
2022-02-01 14:26:20,525 INFO MainThread:27520 [wandb_run.py:_atexit_cleanup():1780] got exitcode: 0
|
| 26 |
+
2022-02-01 14:26:20,528 INFO MainThread:27520 [wandb_run.py:_restore():1752] restore
|
| 27 |
+
2022-02-01 14:26:23,413 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 28 |
+
wandb_count: 1
|
| 29 |
+
}
|
| 30 |
+
pusher_stats {
|
| 31 |
+
uploaded_bytes: 2177
|
| 32 |
+
total_bytes: 2177
|
| 33 |
+
}
|
| 34 |
+
|
| 35 |
+
2022-02-01 14:26:23,726 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 36 |
+
wandb_count: 1
|
| 37 |
+
}
|
| 38 |
+
pusher_stats {
|
| 39 |
+
uploaded_bytes: 2177
|
| 40 |
+
total_bytes: 2177
|
| 41 |
+
}
|
| 42 |
+
|
| 43 |
+
2022-02-01 14:26:24,351 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 44 |
+
wandb_count: 4
|
| 45 |
+
}
|
| 46 |
+
pusher_stats {
|
| 47 |
+
uploaded_bytes: 2177
|
| 48 |
+
total_bytes: 14447
|
| 49 |
+
}
|
| 50 |
+
|
| 51 |
+
2022-02-01 14:26:24,454 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 52 |
+
wandb_count: 5
|
| 53 |
+
}
|
| 54 |
+
pusher_stats {
|
| 55 |
+
uploaded_bytes: 2177
|
| 56 |
+
total_bytes: 17580
|
| 57 |
+
}
|
| 58 |
+
|
| 59 |
+
2022-02-01 14:26:24,566 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 60 |
+
wandb_count: 5
|
| 61 |
+
}
|
| 62 |
+
pusher_stats {
|
| 63 |
+
uploaded_bytes: 2177
|
| 64 |
+
total_bytes: 17580
|
| 65 |
+
}
|
| 66 |
+
|
| 67 |
+
2022-02-01 14:26:24,669 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 68 |
+
wandb_count: 5
|
| 69 |
+
}
|
| 70 |
+
pusher_stats {
|
| 71 |
+
uploaded_bytes: 17580
|
| 72 |
+
total_bytes: 17580
|
| 73 |
+
}
|
| 74 |
+
|
| 75 |
+
2022-02-01 14:26:24,772 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 76 |
+
wandb_count: 5
|
| 77 |
+
}
|
| 78 |
+
pusher_stats {
|
| 79 |
+
uploaded_bytes: 17580
|
| 80 |
+
total_bytes: 17580
|
| 81 |
+
}
|
| 82 |
+
|
| 83 |
+
2022-02-01 14:26:24,875 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 84 |
+
wandb_count: 5
|
| 85 |
+
}
|
| 86 |
+
pusher_stats {
|
| 87 |
+
uploaded_bytes: 17580
|
| 88 |
+
total_bytes: 17580
|
| 89 |
+
}
|
| 90 |
+
|
| 91 |
+
2022-02-01 14:26:24,978 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 92 |
+
wandb_count: 5
|
| 93 |
+
}
|
| 94 |
+
pusher_stats {
|
| 95 |
+
uploaded_bytes: 17580
|
| 96 |
+
total_bytes: 17580
|
| 97 |
+
}
|
| 98 |
+
|
| 99 |
+
2022-02-01 14:26:25,080 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 100 |
+
wandb_count: 5
|
| 101 |
+
}
|
| 102 |
+
pusher_stats {
|
| 103 |
+
uploaded_bytes: 17580
|
| 104 |
+
total_bytes: 17580
|
| 105 |
+
}
|
| 106 |
+
|
| 107 |
+
2022-02-01 14:26:25,555 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: file_counts {
|
| 108 |
+
wandb_count: 5
|
| 109 |
+
}
|
| 110 |
+
pusher_stats {
|
| 111 |
+
uploaded_bytes: 17580
|
| 112 |
+
total_bytes: 17580
|
| 113 |
+
}
|
| 114 |
+
|
| 115 |
+
2022-02-01 14:26:25,964 INFO MainThread:27520 [wandb_run.py:_wait_for_finish():1912] got exit ret: done: true
|
| 116 |
+
exit_result {
|
| 117 |
+
}
|
| 118 |
+
file_counts {
|
| 119 |
+
wandb_count: 5
|
| 120 |
+
}
|
| 121 |
+
pusher_stats {
|
| 122 |
+
uploaded_bytes: 17580
|
| 123 |
+
total_bytes: 17580
|
| 124 |
+
}
|
| 125 |
+
local_info {
|
| 126 |
+
}
|
| 127 |
+
|
| 128 |
+
2022-02-01 14:26:27,100 INFO MainThread:27520 [wandb_run.py:_append_history():2130] rendering history
|
| 129 |
+
2022-02-01 14:26:27,101 INFO MainThread:27520 [wandb_run.py:_append_summary():2085] rendering summary
|
| 130 |
+
2022-02-01 14:26:27,101 INFO MainThread:27520 [wandb_run.py:_append_files():2180] logging synced files
|
wandb/run-20220201_142447-3hqii6h6/run-3hqii6h6.wandb
CHANGED
|
Binary files a/wandb/run-20220201_142447-3hqii6h6/run-3hqii6h6.wandb and b/wandb/run-20220201_142447-3hqii6h6/run-3hqii6h6.wandb differ
|
|
|