feat: add all check points and runs
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- README.md +83 -0
- added_tokens.json +109 -0
- all_results.json +12 -0
- checkpoint-100/config.json +41 -0
- checkpoint-100/optimizer.pt +3 -0
- checkpoint-100/preprocessor_config.json +0 -0
- checkpoint-100/pytorch_model.bin +3 -0
- checkpoint-100/rng_state.pth +3 -0
- checkpoint-100/scaler.pt +3 -0
- checkpoint-100/scheduler.pt +3 -0
- checkpoint-100/trainer_state.json +145 -0
- checkpoint-100/training_args.bin +3 -0
- checkpoint-200/config.json +41 -0
- checkpoint-200/optimizer.pt +3 -0
- checkpoint-200/preprocessor_config.json +0 -0
- checkpoint-200/pytorch_model.bin +3 -0
- checkpoint-200/rng_state.pth +3 -0
- checkpoint-200/scaler.pt +3 -0
- checkpoint-200/scheduler.pt +3 -0
- checkpoint-200/trainer_state.json +274 -0
- checkpoint-200/training_args.bin +3 -0
- checkpoint-300/config.json +41 -0
- checkpoint-300/optimizer.pt +3 -0
- checkpoint-300/preprocessor_config.json +0 -0
- checkpoint-300/pytorch_model.bin +3 -0
- checkpoint-300/rng_state.pth +3 -0
- checkpoint-300/scaler.pt +3 -0
- checkpoint-300/scheduler.pt +3 -0
- checkpoint-300/trainer_state.json +403 -0
- checkpoint-300/training_args.bin +3 -0
- checkpoint-400/config.json +41 -0
- checkpoint-400/optimizer.pt +3 -0
- checkpoint-400/preprocessor_config.json +0 -0
- checkpoint-400/pytorch_model.bin +3 -0
- checkpoint-400/rng_state.pth +3 -0
- checkpoint-400/scaler.pt +3 -0
- checkpoint-400/scheduler.pt +3 -0
- checkpoint-400/trainer_state.json +532 -0
- checkpoint-400/training_args.bin +3 -0
- checkpoint-500/config.json +41 -0
- checkpoint-500/optimizer.pt +3 -0
- checkpoint-500/preprocessor_config.json +0 -0
- checkpoint-500/pytorch_model.bin +3 -0
- checkpoint-500/rng_state.pth +3 -0
- checkpoint-500/scaler.pt +3 -0
- checkpoint-500/scheduler.pt +3 -0
- checkpoint-500/trainer_state.json +661 -0
- checkpoint-500/training_args.bin +3 -0
- config.json +41 -0
- eval_results.json +8 -0
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- whisper-event
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- google/fleurs
|
| 8 |
+
metrics:
|
| 9 |
+
- wer
|
| 10 |
+
model-index:
|
| 11 |
+
- name: Whisper Medium Pashto
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Automatic Speech Recognition
|
| 15 |
+
type: automatic-speech-recognition
|
| 16 |
+
dataset:
|
| 17 |
+
name: google/fleurs ps_af
|
| 18 |
+
type: google/fleurs
|
| 19 |
+
config: ps_af
|
| 20 |
+
split: test
|
| 21 |
+
args: ps_af
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Wer
|
| 24 |
+
type: wer
|
| 25 |
+
value: 50.6431598062954
|
| 26 |
+
---
|
| 27 |
+
|
| 28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 29 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 30 |
+
|
| 31 |
+
# Whisper Medium Pashto
|
| 32 |
+
|
| 33 |
+
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the google/fleurs ps_af dataset.
|
| 34 |
+
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 1.2950
|
| 36 |
+
- Wer: 50.6432
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 1e-05
|
| 56 |
+
- train_batch_size: 32
|
| 57 |
+
- eval_batch_size: 16
|
| 58 |
+
- seed: 42
|
| 59 |
+
- gradient_accumulation_steps: 2
|
| 60 |
+
- total_train_batch_size: 64
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- lr_scheduler_warmup_steps: 50
|
| 64 |
+
- training_steps: 500
|
| 65 |
+
- mixed_precision_training: Native AMP
|
| 66 |
+
|
| 67 |
+
### Training results
|
| 68 |
+
|
| 69 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 70 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
| 71 |
+
| 0.0139 | 14.29 | 100 | 1.0302 | 50.1211 |
|
| 72 |
+
| 0.0011 | 28.57 | 200 | 1.2129 | 49.7806 |
|
| 73 |
+
| 0.0008 | 42.86 | 300 | 1.2581 | 50.3178 |
|
| 74 |
+
| 0.0007 | 57.14 | 400 | 1.2850 | 50.5524 |
|
| 75 |
+
| 0.0007 | 71.43 | 500 | 1.2950 | 50.6432 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- Transformers 4.26.0.dev0
|
| 81 |
+
- Pytorch 1.13.0+cu117
|
| 82 |
+
- Datasets 2.7.1.dev0
|
| 83 |
+
- Tokenizers 0.13.2
|
added_tokens.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|af|>": 50327,
|
| 3 |
+
"<|am|>": 50334,
|
| 4 |
+
"<|ar|>": 50272,
|
| 5 |
+
"<|as|>": 50350,
|
| 6 |
+
"<|az|>": 50304,
|
| 7 |
+
"<|ba|>": 50355,
|
| 8 |
+
"<|be|>": 50330,
|
| 9 |
+
"<|bg|>": 50292,
|
| 10 |
+
"<|bn|>": 50302,
|
| 11 |
+
"<|bo|>": 50347,
|
| 12 |
+
"<|br|>": 50309,
|
| 13 |
+
"<|bs|>": 50315,
|
| 14 |
+
"<|ca|>": 50270,
|
| 15 |
+
"<|cs|>": 50283,
|
| 16 |
+
"<|cy|>": 50297,
|
| 17 |
+
"<|da|>": 50285,
|
| 18 |
+
"<|de|>": 50261,
|
| 19 |
+
"<|el|>": 50281,
|
| 20 |
+
"<|endoftext|>": 50257,
|
| 21 |
+
"<|en|>": 50259,
|
| 22 |
+
"<|es|>": 50262,
|
| 23 |
+
"<|et|>": 50307,
|
| 24 |
+
"<|eu|>": 50310,
|
| 25 |
+
"<|fa|>": 50300,
|
| 26 |
+
"<|fi|>": 50277,
|
| 27 |
+
"<|fo|>": 50338,
|
| 28 |
+
"<|fr|>": 50265,
|
| 29 |
+
"<|gl|>": 50319,
|
| 30 |
+
"<|gu|>": 50333,
|
| 31 |
+
"<|haw|>": 50352,
|
| 32 |
+
"<|ha|>": 50354,
|
| 33 |
+
"<|hi|>": 50276,
|
| 34 |
+
"<|hr|>": 50291,
|
| 35 |
+
"<|ht|>": 50339,
|
| 36 |
+
"<|hu|>": 50286,
|
| 37 |
+
"<|hy|>": 50312,
|
| 38 |
+
"<|id|>": 50275,
|
| 39 |
+
"<|is|>": 50311,
|
| 40 |
+
"<|it|>": 50274,
|
| 41 |
+
"<|iw|>": 50279,
|
| 42 |
+
"<|ja|>": 50266,
|
| 43 |
+
"<|jw|>": 50356,
|
| 44 |
+
"<|ka|>": 50329,
|
| 45 |
+
"<|kk|>": 50316,
|
| 46 |
+
"<|km|>": 50323,
|
| 47 |
+
"<|kn|>": 50306,
|
| 48 |
+
"<|ko|>": 50264,
|
| 49 |
+
"<|la|>": 50294,
|
| 50 |
+
"<|lb|>": 50345,
|
| 51 |
+
"<|ln|>": 50353,
|
| 52 |
+
"<|lo|>": 50336,
|
| 53 |
+
"<|lt|>": 50293,
|
| 54 |
+
"<|lv|>": 50301,
|
| 55 |
+
"<|mg|>": 50349,
|
| 56 |
+
"<|mi|>": 50295,
|
| 57 |
+
"<|mk|>": 50308,
|
| 58 |
+
"<|ml|>": 50296,
|
| 59 |
+
"<|mn|>": 50314,
|
| 60 |
+
"<|mr|>": 50320,
|
| 61 |
+
"<|ms|>": 50282,
|
| 62 |
+
"<|mt|>": 50343,
|
| 63 |
+
"<|my|>": 50346,
|
| 64 |
+
"<|ne|>": 50313,
|
| 65 |
+
"<|nl|>": 50271,
|
| 66 |
+
"<|nn|>": 50342,
|
| 67 |
+
"<|nocaptions|>": 50362,
|
| 68 |
+
"<|notimestamps|>": 50363,
|
| 69 |
+
"<|no|>": 50288,
|
| 70 |
+
"<|oc|>": 50328,
|
| 71 |
+
"<|pa|>": 50321,
|
| 72 |
+
"<|pl|>": 50269,
|
| 73 |
+
"<|ps|>": 50340,
|
| 74 |
+
"<|pt|>": 50267,
|
| 75 |
+
"<|ro|>": 50284,
|
| 76 |
+
"<|ru|>": 50263,
|
| 77 |
+
"<|sa|>": 50344,
|
| 78 |
+
"<|sd|>": 50332,
|
| 79 |
+
"<|si|>": 50322,
|
| 80 |
+
"<|sk|>": 50298,
|
| 81 |
+
"<|sl|>": 50305,
|
| 82 |
+
"<|sn|>": 50324,
|
| 83 |
+
"<|so|>": 50326,
|
| 84 |
+
"<|sq|>": 50317,
|
| 85 |
+
"<|sr|>": 50303,
|
| 86 |
+
"<|startoflm|>": 50360,
|
| 87 |
+
"<|startofprev|>": 50361,
|
| 88 |
+
"<|startoftranscript|>": 50258,
|
| 89 |
+
"<|su|>": 50357,
|
| 90 |
+
"<|sv|>": 50273,
|
| 91 |
+
"<|sw|>": 50318,
|
| 92 |
+
"<|ta|>": 50287,
|
| 93 |
+
"<|te|>": 50299,
|
| 94 |
+
"<|tg|>": 50331,
|
| 95 |
+
"<|th|>": 50289,
|
| 96 |
+
"<|tk|>": 50341,
|
| 97 |
+
"<|tl|>": 50348,
|
| 98 |
+
"<|transcribe|>": 50359,
|
| 99 |
+
"<|translate|>": 50358,
|
| 100 |
+
"<|tr|>": 50268,
|
| 101 |
+
"<|tt|>": 50351,
|
| 102 |
+
"<|uk|>": 50280,
|
| 103 |
+
"<|ur|>": 50290,
|
| 104 |
+
"<|uz|>": 50337,
|
| 105 |
+
"<|vi|>": 50278,
|
| 106 |
+
"<|yi|>": 50335,
|
| 107 |
+
"<|yo|>": 50325,
|
| 108 |
+
"<|zh|>": 50260
|
| 109 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 71.43,
|
| 3 |
+
"eval_loss": 1.2950359582901,
|
| 4 |
+
"eval_runtime": 1406.7383,
|
| 5 |
+
"eval_samples_per_second": 0.364,
|
| 6 |
+
"eval_steps_per_second": 0.023,
|
| 7 |
+
"eval_wer": 50.6431598062954,
|
| 8 |
+
"train_loss": 0.15911377191729845,
|
| 9 |
+
"train_runtime": 10886.8048,
|
| 10 |
+
"train_samples_per_second": 2.939,
|
| 11 |
+
"train_steps_per_second": 0.046
|
| 12 |
+
}
|
checkpoint-100/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
checkpoint-100/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0fb9e51e60d79519fd5171098988e8560e9d36c85ee41acd1cc2918faa49d2e
|
| 3 |
+
size 6111427735
|
checkpoint-100/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-100/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3c6b9f5ed51b592b1d4df94d83892687710f94c0908b1141b8d4c9cd6c0524d
|
| 3 |
+
size 3055754841
|
checkpoint-100/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3709c416ef836e922c3858f4994f9cac2a7416be0deeff25b125cf99d5cdbaff
|
| 3 |
+
size 14575
|
checkpoint-100/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b105b45da1b1eeaa2580b2040606e7e6bb5bf197448a40b52314adb128fc0fe
|
| 3 |
+
size 557
|
checkpoint-100/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:825d34697209381e23160a364b6758cbaa60e05da184658f891484adc408cfa4
|
| 3 |
+
size 627
|
checkpoint-100/trainer_state.json
ADDED
|
@@ -0,0 +1,145 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.285714285714286,
|
| 5 |
+
"global_step": 100,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.71,
|
| 12 |
+
"learning_rate": 6.000000000000001e-07,
|
| 13 |
+
"loss": 2.8734,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.43,
|
| 18 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 19 |
+
"loss": 2.7517,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.14,
|
| 24 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 25 |
+
"loss": 2.2084,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.86,
|
| 30 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 31 |
+
"loss": 1.7218,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.57,
|
| 36 |
+
"learning_rate": 4.4e-06,
|
| 37 |
+
"loss": 1.3608,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.29,
|
| 42 |
+
"learning_rate": 5.400000000000001e-06,
|
| 43 |
+
"loss": 1.046,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 5.0,
|
| 48 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 49 |
+
"loss": 0.8672,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.71,
|
| 54 |
+
"learning_rate": 7.4e-06,
|
| 55 |
+
"loss": 0.6926,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 6.43,
|
| 60 |
+
"learning_rate": 8.400000000000001e-06,
|
| 61 |
+
"loss": 0.5528,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 7.14,
|
| 66 |
+
"learning_rate": 9.4e-06,
|
| 67 |
+
"loss": 0.4595,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 7.86,
|
| 72 |
+
"learning_rate": 9.955555555555556e-06,
|
| 73 |
+
"loss": 0.3677,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 8.57,
|
| 78 |
+
"learning_rate": 9.844444444444446e-06,
|
| 79 |
+
"loss": 0.2566,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 9.29,
|
| 84 |
+
"learning_rate": 9.733333333333334e-06,
|
| 85 |
+
"loss": 0.2176,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 10.0,
|
| 90 |
+
"learning_rate": 9.622222222222222e-06,
|
| 91 |
+
"loss": 0.1514,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.71,
|
| 96 |
+
"learning_rate": 9.511111111111112e-06,
|
| 97 |
+
"loss": 0.0989,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.43,
|
| 102 |
+
"learning_rate": 9.4e-06,
|
| 103 |
+
"loss": 0.0644,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 12.14,
|
| 108 |
+
"learning_rate": 9.28888888888889e-06,
|
| 109 |
+
"loss": 0.0445,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.86,
|
| 114 |
+
"learning_rate": 9.17777777777778e-06,
|
| 115 |
+
"loss": 0.0279,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 13.57,
|
| 120 |
+
"learning_rate": 9.066666666666667e-06,
|
| 121 |
+
"loss": 0.0191,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 14.29,
|
| 126 |
+
"learning_rate": 8.955555555555555e-06,
|
| 127 |
+
"loss": 0.0139,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 14.29,
|
| 132 |
+
"eval_loss": 1.0302417278289795,
|
| 133 |
+
"eval_runtime": 1389.5619,
|
| 134 |
+
"eval_samples_per_second": 0.368,
|
| 135 |
+
"eval_steps_per_second": 0.023,
|
| 136 |
+
"eval_wer": 50.12106537530266,
|
| 137 |
+
"step": 100
|
| 138 |
+
}
|
| 139 |
+
],
|
| 140 |
+
"max_steps": 500,
|
| 141 |
+
"num_train_epochs": 72,
|
| 142 |
+
"total_flos": 6.28896744013824e+18,
|
| 143 |
+
"trial_name": null,
|
| 144 |
+
"trial_params": null
|
| 145 |
+
}
|
checkpoint-100/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb79c84ed9cdcdf9027db44a0b593b1eebe5ee0791d07229cec597018c5ec2ca
|
| 3 |
+
size 3515
|
checkpoint-200/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
checkpoint-200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99186b9e3c457c5c66823b532ccc6fe65f760e1655feb6eb9e38de421c636771
|
| 3 |
+
size 6111427735
|
checkpoint-200/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-200/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e2e99b5c8e1b2c8fd910a6255954e543689d0899047183ed6ec4c6ccae0eeb02
|
| 3 |
+
size 3055754841
|
checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:059c2818547187b20bdfb682888aaa741b90af75bee8ac217b2bd048f8283e10
|
| 3 |
+
size 14575
|
checkpoint-200/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a06a53451305e7676837abc472465f3eeec15043b9cef01cfcc02862f53efbcb
|
| 3 |
+
size 557
|
checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a4b13dcd02297706644d474717d32685ea7b3a491901f271c0e11f08c8779ef
|
| 3 |
+
size 627
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,274 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 28.571428571428573,
|
| 5 |
+
"global_step": 200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.71,
|
| 12 |
+
"learning_rate": 6.000000000000001e-07,
|
| 13 |
+
"loss": 2.8734,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.43,
|
| 18 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 19 |
+
"loss": 2.7517,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.14,
|
| 24 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 25 |
+
"loss": 2.2084,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.86,
|
| 30 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 31 |
+
"loss": 1.7218,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.57,
|
| 36 |
+
"learning_rate": 4.4e-06,
|
| 37 |
+
"loss": 1.3608,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.29,
|
| 42 |
+
"learning_rate": 5.400000000000001e-06,
|
| 43 |
+
"loss": 1.046,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 5.0,
|
| 48 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 49 |
+
"loss": 0.8672,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.71,
|
| 54 |
+
"learning_rate": 7.4e-06,
|
| 55 |
+
"loss": 0.6926,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 6.43,
|
| 60 |
+
"learning_rate": 8.400000000000001e-06,
|
| 61 |
+
"loss": 0.5528,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 7.14,
|
| 66 |
+
"learning_rate": 9.4e-06,
|
| 67 |
+
"loss": 0.4595,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 7.86,
|
| 72 |
+
"learning_rate": 9.955555555555556e-06,
|
| 73 |
+
"loss": 0.3677,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 8.57,
|
| 78 |
+
"learning_rate": 9.844444444444446e-06,
|
| 79 |
+
"loss": 0.2566,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 9.29,
|
| 84 |
+
"learning_rate": 9.733333333333334e-06,
|
| 85 |
+
"loss": 0.2176,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 10.0,
|
| 90 |
+
"learning_rate": 9.622222222222222e-06,
|
| 91 |
+
"loss": 0.1514,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.71,
|
| 96 |
+
"learning_rate": 9.511111111111112e-06,
|
| 97 |
+
"loss": 0.0989,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.43,
|
| 102 |
+
"learning_rate": 9.4e-06,
|
| 103 |
+
"loss": 0.0644,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 12.14,
|
| 108 |
+
"learning_rate": 9.28888888888889e-06,
|
| 109 |
+
"loss": 0.0445,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.86,
|
| 114 |
+
"learning_rate": 9.17777777777778e-06,
|
| 115 |
+
"loss": 0.0279,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 13.57,
|
| 120 |
+
"learning_rate": 9.066666666666667e-06,
|
| 121 |
+
"loss": 0.0191,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 14.29,
|
| 126 |
+
"learning_rate": 8.955555555555555e-06,
|
| 127 |
+
"loss": 0.0139,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 14.29,
|
| 132 |
+
"eval_loss": 1.0302417278289795,
|
| 133 |
+
"eval_runtime": 1389.5619,
|
| 134 |
+
"eval_samples_per_second": 0.368,
|
| 135 |
+
"eval_steps_per_second": 0.023,
|
| 136 |
+
"eval_wer": 50.12106537530266,
|
| 137 |
+
"step": 100
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 15.0,
|
| 141 |
+
"learning_rate": 8.844444444444445e-06,
|
| 142 |
+
"loss": 0.0104,
|
| 143 |
+
"step": 105
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 15.71,
|
| 147 |
+
"learning_rate": 8.733333333333333e-06,
|
| 148 |
+
"loss": 0.0091,
|
| 149 |
+
"step": 110
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.43,
|
| 153 |
+
"learning_rate": 8.622222222222223e-06,
|
| 154 |
+
"loss": 0.0062,
|
| 155 |
+
"step": 115
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 17.14,
|
| 159 |
+
"learning_rate": 8.511111111111113e-06,
|
| 160 |
+
"loss": 0.0056,
|
| 161 |
+
"step": 120
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 17.86,
|
| 165 |
+
"learning_rate": 8.400000000000001e-06,
|
| 166 |
+
"loss": 0.0052,
|
| 167 |
+
"step": 125
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 18.57,
|
| 171 |
+
"learning_rate": 8.288888888888889e-06,
|
| 172 |
+
"loss": 0.0047,
|
| 173 |
+
"step": 130
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 19.29,
|
| 177 |
+
"learning_rate": 8.177777777777779e-06,
|
| 178 |
+
"loss": 0.0049,
|
| 179 |
+
"step": 135
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 20.0,
|
| 183 |
+
"learning_rate": 8.066666666666667e-06,
|
| 184 |
+
"loss": 0.0042,
|
| 185 |
+
"step": 140
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 20.71,
|
| 189 |
+
"learning_rate": 7.955555555555557e-06,
|
| 190 |
+
"loss": 0.0033,
|
| 191 |
+
"step": 145
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 21.43,
|
| 195 |
+
"learning_rate": 7.844444444444446e-06,
|
| 196 |
+
"loss": 0.0028,
|
| 197 |
+
"step": 150
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 22.14,
|
| 201 |
+
"learning_rate": 7.733333333333334e-06,
|
| 202 |
+
"loss": 0.002,
|
| 203 |
+
"step": 155
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 22.86,
|
| 207 |
+
"learning_rate": 7.622222222222223e-06,
|
| 208 |
+
"loss": 0.0018,
|
| 209 |
+
"step": 160
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 23.57,
|
| 213 |
+
"learning_rate": 7.511111111111111e-06,
|
| 214 |
+
"loss": 0.0016,
|
| 215 |
+
"step": 165
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 24.29,
|
| 219 |
+
"learning_rate": 7.4e-06,
|
| 220 |
+
"loss": 0.0014,
|
| 221 |
+
"step": 170
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 25.0,
|
| 225 |
+
"learning_rate": 7.28888888888889e-06,
|
| 226 |
+
"loss": 0.0013,
|
| 227 |
+
"step": 175
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 25.71,
|
| 231 |
+
"learning_rate": 7.177777777777778e-06,
|
| 232 |
+
"loss": 0.0013,
|
| 233 |
+
"step": 180
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 26.43,
|
| 237 |
+
"learning_rate": 7.066666666666667e-06,
|
| 238 |
+
"loss": 0.0012,
|
| 239 |
+
"step": 185
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 27.14,
|
| 243 |
+
"learning_rate": 6.955555555555557e-06,
|
| 244 |
+
"loss": 0.0011,
|
| 245 |
+
"step": 190
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 27.86,
|
| 249 |
+
"learning_rate": 6.844444444444445e-06,
|
| 250 |
+
"loss": 0.0011,
|
| 251 |
+
"step": 195
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 28.57,
|
| 255 |
+
"learning_rate": 6.733333333333334e-06,
|
| 256 |
+
"loss": 0.0011,
|
| 257 |
+
"step": 200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 28.57,
|
| 261 |
+
"eval_loss": 1.2129117250442505,
|
| 262 |
+
"eval_runtime": 1370.5005,
|
| 263 |
+
"eval_samples_per_second": 0.374,
|
| 264 |
+
"eval_steps_per_second": 0.023,
|
| 265 |
+
"eval_wer": 49.78056900726393,
|
| 266 |
+
"step": 200
|
| 267 |
+
}
|
| 268 |
+
],
|
| 269 |
+
"max_steps": 500,
|
| 270 |
+
"num_train_epochs": 72,
|
| 271 |
+
"total_flos": 1.257793488027648e+19,
|
| 272 |
+
"trial_name": null,
|
| 273 |
+
"trial_params": null
|
| 274 |
+
}
|
checkpoint-200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb79c84ed9cdcdf9027db44a0b593b1eebe5ee0791d07229cec597018c5ec2ca
|
| 3 |
+
size 3515
|
checkpoint-300/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
checkpoint-300/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3febe8568f84b912912648c22dc86dd5ff62befc139771f780d8b6aa3c764126
|
| 3 |
+
size 6111428695
|
checkpoint-300/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-300/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd98c9ab1ae5be0a92949bd972ecd00d15b8bde350107182e1a3f04fb6c7688f
|
| 3 |
+
size 3055754841
|
checkpoint-300/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d565104da7782a07976346107bc03b93e32987893b5d75d00a81b2084e2dcd6
|
| 3 |
+
size 14511
|
checkpoint-300/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f957d37d058cecdb04122df1b56b6bf6f5057fc9d616138235fc0d83a0d50ce2
|
| 3 |
+
size 557
|
checkpoint-300/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17b4e04312a30fafac8ff85d6f7114d460411b8e78f5915892b4c6e3d6f6d4f4
|
| 3 |
+
size 627
|
checkpoint-300/trainer_state.json
ADDED
|
@@ -0,0 +1,403 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 42.857142857142854,
|
| 5 |
+
"global_step": 300,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.71,
|
| 12 |
+
"learning_rate": 6.000000000000001e-07,
|
| 13 |
+
"loss": 2.8734,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.43,
|
| 18 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 19 |
+
"loss": 2.7517,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.14,
|
| 24 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 25 |
+
"loss": 2.2084,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.86,
|
| 30 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 31 |
+
"loss": 1.7218,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.57,
|
| 36 |
+
"learning_rate": 4.4e-06,
|
| 37 |
+
"loss": 1.3608,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.29,
|
| 42 |
+
"learning_rate": 5.400000000000001e-06,
|
| 43 |
+
"loss": 1.046,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 5.0,
|
| 48 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 49 |
+
"loss": 0.8672,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.71,
|
| 54 |
+
"learning_rate": 7.4e-06,
|
| 55 |
+
"loss": 0.6926,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 6.43,
|
| 60 |
+
"learning_rate": 8.400000000000001e-06,
|
| 61 |
+
"loss": 0.5528,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 7.14,
|
| 66 |
+
"learning_rate": 9.4e-06,
|
| 67 |
+
"loss": 0.4595,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 7.86,
|
| 72 |
+
"learning_rate": 9.955555555555556e-06,
|
| 73 |
+
"loss": 0.3677,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 8.57,
|
| 78 |
+
"learning_rate": 9.844444444444446e-06,
|
| 79 |
+
"loss": 0.2566,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 9.29,
|
| 84 |
+
"learning_rate": 9.733333333333334e-06,
|
| 85 |
+
"loss": 0.2176,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 10.0,
|
| 90 |
+
"learning_rate": 9.622222222222222e-06,
|
| 91 |
+
"loss": 0.1514,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.71,
|
| 96 |
+
"learning_rate": 9.511111111111112e-06,
|
| 97 |
+
"loss": 0.0989,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.43,
|
| 102 |
+
"learning_rate": 9.4e-06,
|
| 103 |
+
"loss": 0.0644,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 12.14,
|
| 108 |
+
"learning_rate": 9.28888888888889e-06,
|
| 109 |
+
"loss": 0.0445,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.86,
|
| 114 |
+
"learning_rate": 9.17777777777778e-06,
|
| 115 |
+
"loss": 0.0279,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 13.57,
|
| 120 |
+
"learning_rate": 9.066666666666667e-06,
|
| 121 |
+
"loss": 0.0191,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 14.29,
|
| 126 |
+
"learning_rate": 8.955555555555555e-06,
|
| 127 |
+
"loss": 0.0139,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 14.29,
|
| 132 |
+
"eval_loss": 1.0302417278289795,
|
| 133 |
+
"eval_runtime": 1389.5619,
|
| 134 |
+
"eval_samples_per_second": 0.368,
|
| 135 |
+
"eval_steps_per_second": 0.023,
|
| 136 |
+
"eval_wer": 50.12106537530266,
|
| 137 |
+
"step": 100
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 15.0,
|
| 141 |
+
"learning_rate": 8.844444444444445e-06,
|
| 142 |
+
"loss": 0.0104,
|
| 143 |
+
"step": 105
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 15.71,
|
| 147 |
+
"learning_rate": 8.733333333333333e-06,
|
| 148 |
+
"loss": 0.0091,
|
| 149 |
+
"step": 110
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.43,
|
| 153 |
+
"learning_rate": 8.622222222222223e-06,
|
| 154 |
+
"loss": 0.0062,
|
| 155 |
+
"step": 115
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 17.14,
|
| 159 |
+
"learning_rate": 8.511111111111113e-06,
|
| 160 |
+
"loss": 0.0056,
|
| 161 |
+
"step": 120
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 17.86,
|
| 165 |
+
"learning_rate": 8.400000000000001e-06,
|
| 166 |
+
"loss": 0.0052,
|
| 167 |
+
"step": 125
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 18.57,
|
| 171 |
+
"learning_rate": 8.288888888888889e-06,
|
| 172 |
+
"loss": 0.0047,
|
| 173 |
+
"step": 130
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 19.29,
|
| 177 |
+
"learning_rate": 8.177777777777779e-06,
|
| 178 |
+
"loss": 0.0049,
|
| 179 |
+
"step": 135
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 20.0,
|
| 183 |
+
"learning_rate": 8.066666666666667e-06,
|
| 184 |
+
"loss": 0.0042,
|
| 185 |
+
"step": 140
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 20.71,
|
| 189 |
+
"learning_rate": 7.955555555555557e-06,
|
| 190 |
+
"loss": 0.0033,
|
| 191 |
+
"step": 145
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 21.43,
|
| 195 |
+
"learning_rate": 7.844444444444446e-06,
|
| 196 |
+
"loss": 0.0028,
|
| 197 |
+
"step": 150
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 22.14,
|
| 201 |
+
"learning_rate": 7.733333333333334e-06,
|
| 202 |
+
"loss": 0.002,
|
| 203 |
+
"step": 155
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 22.86,
|
| 207 |
+
"learning_rate": 7.622222222222223e-06,
|
| 208 |
+
"loss": 0.0018,
|
| 209 |
+
"step": 160
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 23.57,
|
| 213 |
+
"learning_rate": 7.511111111111111e-06,
|
| 214 |
+
"loss": 0.0016,
|
| 215 |
+
"step": 165
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 24.29,
|
| 219 |
+
"learning_rate": 7.4e-06,
|
| 220 |
+
"loss": 0.0014,
|
| 221 |
+
"step": 170
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 25.0,
|
| 225 |
+
"learning_rate": 7.28888888888889e-06,
|
| 226 |
+
"loss": 0.0013,
|
| 227 |
+
"step": 175
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 25.71,
|
| 231 |
+
"learning_rate": 7.177777777777778e-06,
|
| 232 |
+
"loss": 0.0013,
|
| 233 |
+
"step": 180
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 26.43,
|
| 237 |
+
"learning_rate": 7.066666666666667e-06,
|
| 238 |
+
"loss": 0.0012,
|
| 239 |
+
"step": 185
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 27.14,
|
| 243 |
+
"learning_rate": 6.955555555555557e-06,
|
| 244 |
+
"loss": 0.0011,
|
| 245 |
+
"step": 190
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 27.86,
|
| 249 |
+
"learning_rate": 6.844444444444445e-06,
|
| 250 |
+
"loss": 0.0011,
|
| 251 |
+
"step": 195
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 28.57,
|
| 255 |
+
"learning_rate": 6.733333333333334e-06,
|
| 256 |
+
"loss": 0.0011,
|
| 257 |
+
"step": 200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 28.57,
|
| 261 |
+
"eval_loss": 1.2129117250442505,
|
| 262 |
+
"eval_runtime": 1370.5005,
|
| 263 |
+
"eval_samples_per_second": 0.374,
|
| 264 |
+
"eval_steps_per_second": 0.023,
|
| 265 |
+
"eval_wer": 49.78056900726393,
|
| 266 |
+
"step": 200
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 29.29,
|
| 270 |
+
"learning_rate": 6.6222222222222236e-06,
|
| 271 |
+
"loss": 0.001,
|
| 272 |
+
"step": 205
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 30.0,
|
| 276 |
+
"learning_rate": 6.511111111111112e-06,
|
| 277 |
+
"loss": 0.001,
|
| 278 |
+
"step": 210
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 30.71,
|
| 282 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 283 |
+
"loss": 0.001,
|
| 284 |
+
"step": 215
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 31.43,
|
| 288 |
+
"learning_rate": 6.28888888888889e-06,
|
| 289 |
+
"loss": 0.001,
|
| 290 |
+
"step": 220
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 32.14,
|
| 294 |
+
"learning_rate": 6.177777777777778e-06,
|
| 295 |
+
"loss": 0.001,
|
| 296 |
+
"step": 225
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 32.86,
|
| 300 |
+
"learning_rate": 6.066666666666667e-06,
|
| 301 |
+
"loss": 0.0009,
|
| 302 |
+
"step": 230
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 33.57,
|
| 306 |
+
"learning_rate": 5.955555555555555e-06,
|
| 307 |
+
"loss": 0.0009,
|
| 308 |
+
"step": 235
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 34.29,
|
| 312 |
+
"learning_rate": 5.844444444444445e-06,
|
| 313 |
+
"loss": 0.0009,
|
| 314 |
+
"step": 240
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 35.0,
|
| 318 |
+
"learning_rate": 5.733333333333334e-06,
|
| 319 |
+
"loss": 0.0009,
|
| 320 |
+
"step": 245
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 35.71,
|
| 324 |
+
"learning_rate": 5.622222222222222e-06,
|
| 325 |
+
"loss": 0.0009,
|
| 326 |
+
"step": 250
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 36.43,
|
| 330 |
+
"learning_rate": 5.511111111111112e-06,
|
| 331 |
+
"loss": 0.0009,
|
| 332 |
+
"step": 255
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 37.14,
|
| 336 |
+
"learning_rate": 5.400000000000001e-06,
|
| 337 |
+
"loss": 0.0008,
|
| 338 |
+
"step": 260
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 37.86,
|
| 342 |
+
"learning_rate": 5.288888888888889e-06,
|
| 343 |
+
"loss": 0.0008,
|
| 344 |
+
"step": 265
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 38.57,
|
| 348 |
+
"learning_rate": 5.177777777777779e-06,
|
| 349 |
+
"loss": 0.0008,
|
| 350 |
+
"step": 270
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 39.29,
|
| 354 |
+
"learning_rate": 5.0666666666666676e-06,
|
| 355 |
+
"loss": 0.0008,
|
| 356 |
+
"step": 275
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 40.0,
|
| 360 |
+
"learning_rate": 4.9555555555555565e-06,
|
| 361 |
+
"loss": 0.0008,
|
| 362 |
+
"step": 280
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 40.71,
|
| 366 |
+
"learning_rate": 4.8444444444444446e-06,
|
| 367 |
+
"loss": 0.0008,
|
| 368 |
+
"step": 285
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 41.43,
|
| 372 |
+
"learning_rate": 4.7333333333333335e-06,
|
| 373 |
+
"loss": 0.0008,
|
| 374 |
+
"step": 290
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 42.14,
|
| 378 |
+
"learning_rate": 4.622222222222222e-06,
|
| 379 |
+
"loss": 0.0008,
|
| 380 |
+
"step": 295
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 42.86,
|
| 384 |
+
"learning_rate": 4.511111111111111e-06,
|
| 385 |
+
"loss": 0.0008,
|
| 386 |
+
"step": 300
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 42.86,
|
| 390 |
+
"eval_loss": 1.2581015825271606,
|
| 391 |
+
"eval_runtime": 1407.7801,
|
| 392 |
+
"eval_samples_per_second": 0.364,
|
| 393 |
+
"eval_steps_per_second": 0.023,
|
| 394 |
+
"eval_wer": 50.317796610169495,
|
| 395 |
+
"step": 300
|
| 396 |
+
}
|
| 397 |
+
],
|
| 398 |
+
"max_steps": 500,
|
| 399 |
+
"num_train_epochs": 72,
|
| 400 |
+
"total_flos": 1.886690232041472e+19,
|
| 401 |
+
"trial_name": null,
|
| 402 |
+
"trial_params": null
|
| 403 |
+
}
|
checkpoint-300/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb79c84ed9cdcdf9027db44a0b593b1eebe5ee0791d07229cec597018c5ec2ca
|
| 3 |
+
size 3515
|
checkpoint-400/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
checkpoint-400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0335d7ccc7dd1a249958e509497fd02cda7a8e3e0094b5ef6ec98bb6bec65021
|
| 3 |
+
size 6111428695
|
checkpoint-400/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-400/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:22ff2c60776a9d28c239afed909dbffb11393107cb19c2b8e79dc0508b7cdb69
|
| 3 |
+
size 3055754841
|
checkpoint-400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a002adf0ecf251fa60d893b50d0a42a0f9750970dec3c860fd2a1ff59ba59089
|
| 3 |
+
size 14575
|
checkpoint-400/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d2260cc8eedb64ad9eeb9e9720d1880ff939090257cdf6a6b9eec0a85adbf17
|
| 3 |
+
size 557
|
checkpoint-400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a532bfa83abb6d3c76be2f96ae5b8d0ea5dd1f35b8422b5072484dbee62ebabf
|
| 3 |
+
size 627
|
checkpoint-400/trainer_state.json
ADDED
|
@@ -0,0 +1,532 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 57.142857142857146,
|
| 5 |
+
"global_step": 400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.71,
|
| 12 |
+
"learning_rate": 6.000000000000001e-07,
|
| 13 |
+
"loss": 2.8734,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.43,
|
| 18 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 19 |
+
"loss": 2.7517,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.14,
|
| 24 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 25 |
+
"loss": 2.2084,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.86,
|
| 30 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 31 |
+
"loss": 1.7218,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.57,
|
| 36 |
+
"learning_rate": 4.4e-06,
|
| 37 |
+
"loss": 1.3608,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.29,
|
| 42 |
+
"learning_rate": 5.400000000000001e-06,
|
| 43 |
+
"loss": 1.046,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 5.0,
|
| 48 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 49 |
+
"loss": 0.8672,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.71,
|
| 54 |
+
"learning_rate": 7.4e-06,
|
| 55 |
+
"loss": 0.6926,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 6.43,
|
| 60 |
+
"learning_rate": 8.400000000000001e-06,
|
| 61 |
+
"loss": 0.5528,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 7.14,
|
| 66 |
+
"learning_rate": 9.4e-06,
|
| 67 |
+
"loss": 0.4595,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 7.86,
|
| 72 |
+
"learning_rate": 9.955555555555556e-06,
|
| 73 |
+
"loss": 0.3677,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 8.57,
|
| 78 |
+
"learning_rate": 9.844444444444446e-06,
|
| 79 |
+
"loss": 0.2566,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 9.29,
|
| 84 |
+
"learning_rate": 9.733333333333334e-06,
|
| 85 |
+
"loss": 0.2176,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 10.0,
|
| 90 |
+
"learning_rate": 9.622222222222222e-06,
|
| 91 |
+
"loss": 0.1514,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.71,
|
| 96 |
+
"learning_rate": 9.511111111111112e-06,
|
| 97 |
+
"loss": 0.0989,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.43,
|
| 102 |
+
"learning_rate": 9.4e-06,
|
| 103 |
+
"loss": 0.0644,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 12.14,
|
| 108 |
+
"learning_rate": 9.28888888888889e-06,
|
| 109 |
+
"loss": 0.0445,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.86,
|
| 114 |
+
"learning_rate": 9.17777777777778e-06,
|
| 115 |
+
"loss": 0.0279,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 13.57,
|
| 120 |
+
"learning_rate": 9.066666666666667e-06,
|
| 121 |
+
"loss": 0.0191,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 14.29,
|
| 126 |
+
"learning_rate": 8.955555555555555e-06,
|
| 127 |
+
"loss": 0.0139,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 14.29,
|
| 132 |
+
"eval_loss": 1.0302417278289795,
|
| 133 |
+
"eval_runtime": 1389.5619,
|
| 134 |
+
"eval_samples_per_second": 0.368,
|
| 135 |
+
"eval_steps_per_second": 0.023,
|
| 136 |
+
"eval_wer": 50.12106537530266,
|
| 137 |
+
"step": 100
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 15.0,
|
| 141 |
+
"learning_rate": 8.844444444444445e-06,
|
| 142 |
+
"loss": 0.0104,
|
| 143 |
+
"step": 105
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 15.71,
|
| 147 |
+
"learning_rate": 8.733333333333333e-06,
|
| 148 |
+
"loss": 0.0091,
|
| 149 |
+
"step": 110
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.43,
|
| 153 |
+
"learning_rate": 8.622222222222223e-06,
|
| 154 |
+
"loss": 0.0062,
|
| 155 |
+
"step": 115
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 17.14,
|
| 159 |
+
"learning_rate": 8.511111111111113e-06,
|
| 160 |
+
"loss": 0.0056,
|
| 161 |
+
"step": 120
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 17.86,
|
| 165 |
+
"learning_rate": 8.400000000000001e-06,
|
| 166 |
+
"loss": 0.0052,
|
| 167 |
+
"step": 125
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 18.57,
|
| 171 |
+
"learning_rate": 8.288888888888889e-06,
|
| 172 |
+
"loss": 0.0047,
|
| 173 |
+
"step": 130
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 19.29,
|
| 177 |
+
"learning_rate": 8.177777777777779e-06,
|
| 178 |
+
"loss": 0.0049,
|
| 179 |
+
"step": 135
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 20.0,
|
| 183 |
+
"learning_rate": 8.066666666666667e-06,
|
| 184 |
+
"loss": 0.0042,
|
| 185 |
+
"step": 140
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 20.71,
|
| 189 |
+
"learning_rate": 7.955555555555557e-06,
|
| 190 |
+
"loss": 0.0033,
|
| 191 |
+
"step": 145
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 21.43,
|
| 195 |
+
"learning_rate": 7.844444444444446e-06,
|
| 196 |
+
"loss": 0.0028,
|
| 197 |
+
"step": 150
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 22.14,
|
| 201 |
+
"learning_rate": 7.733333333333334e-06,
|
| 202 |
+
"loss": 0.002,
|
| 203 |
+
"step": 155
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 22.86,
|
| 207 |
+
"learning_rate": 7.622222222222223e-06,
|
| 208 |
+
"loss": 0.0018,
|
| 209 |
+
"step": 160
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 23.57,
|
| 213 |
+
"learning_rate": 7.511111111111111e-06,
|
| 214 |
+
"loss": 0.0016,
|
| 215 |
+
"step": 165
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 24.29,
|
| 219 |
+
"learning_rate": 7.4e-06,
|
| 220 |
+
"loss": 0.0014,
|
| 221 |
+
"step": 170
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 25.0,
|
| 225 |
+
"learning_rate": 7.28888888888889e-06,
|
| 226 |
+
"loss": 0.0013,
|
| 227 |
+
"step": 175
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 25.71,
|
| 231 |
+
"learning_rate": 7.177777777777778e-06,
|
| 232 |
+
"loss": 0.0013,
|
| 233 |
+
"step": 180
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 26.43,
|
| 237 |
+
"learning_rate": 7.066666666666667e-06,
|
| 238 |
+
"loss": 0.0012,
|
| 239 |
+
"step": 185
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 27.14,
|
| 243 |
+
"learning_rate": 6.955555555555557e-06,
|
| 244 |
+
"loss": 0.0011,
|
| 245 |
+
"step": 190
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 27.86,
|
| 249 |
+
"learning_rate": 6.844444444444445e-06,
|
| 250 |
+
"loss": 0.0011,
|
| 251 |
+
"step": 195
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 28.57,
|
| 255 |
+
"learning_rate": 6.733333333333334e-06,
|
| 256 |
+
"loss": 0.0011,
|
| 257 |
+
"step": 200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 28.57,
|
| 261 |
+
"eval_loss": 1.2129117250442505,
|
| 262 |
+
"eval_runtime": 1370.5005,
|
| 263 |
+
"eval_samples_per_second": 0.374,
|
| 264 |
+
"eval_steps_per_second": 0.023,
|
| 265 |
+
"eval_wer": 49.78056900726393,
|
| 266 |
+
"step": 200
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 29.29,
|
| 270 |
+
"learning_rate": 6.6222222222222236e-06,
|
| 271 |
+
"loss": 0.001,
|
| 272 |
+
"step": 205
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 30.0,
|
| 276 |
+
"learning_rate": 6.511111111111112e-06,
|
| 277 |
+
"loss": 0.001,
|
| 278 |
+
"step": 210
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 30.71,
|
| 282 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 283 |
+
"loss": 0.001,
|
| 284 |
+
"step": 215
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 31.43,
|
| 288 |
+
"learning_rate": 6.28888888888889e-06,
|
| 289 |
+
"loss": 0.001,
|
| 290 |
+
"step": 220
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 32.14,
|
| 294 |
+
"learning_rate": 6.177777777777778e-06,
|
| 295 |
+
"loss": 0.001,
|
| 296 |
+
"step": 225
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 32.86,
|
| 300 |
+
"learning_rate": 6.066666666666667e-06,
|
| 301 |
+
"loss": 0.0009,
|
| 302 |
+
"step": 230
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 33.57,
|
| 306 |
+
"learning_rate": 5.955555555555555e-06,
|
| 307 |
+
"loss": 0.0009,
|
| 308 |
+
"step": 235
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 34.29,
|
| 312 |
+
"learning_rate": 5.844444444444445e-06,
|
| 313 |
+
"loss": 0.0009,
|
| 314 |
+
"step": 240
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 35.0,
|
| 318 |
+
"learning_rate": 5.733333333333334e-06,
|
| 319 |
+
"loss": 0.0009,
|
| 320 |
+
"step": 245
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 35.71,
|
| 324 |
+
"learning_rate": 5.622222222222222e-06,
|
| 325 |
+
"loss": 0.0009,
|
| 326 |
+
"step": 250
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 36.43,
|
| 330 |
+
"learning_rate": 5.511111111111112e-06,
|
| 331 |
+
"loss": 0.0009,
|
| 332 |
+
"step": 255
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 37.14,
|
| 336 |
+
"learning_rate": 5.400000000000001e-06,
|
| 337 |
+
"loss": 0.0008,
|
| 338 |
+
"step": 260
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 37.86,
|
| 342 |
+
"learning_rate": 5.288888888888889e-06,
|
| 343 |
+
"loss": 0.0008,
|
| 344 |
+
"step": 265
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 38.57,
|
| 348 |
+
"learning_rate": 5.177777777777779e-06,
|
| 349 |
+
"loss": 0.0008,
|
| 350 |
+
"step": 270
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 39.29,
|
| 354 |
+
"learning_rate": 5.0666666666666676e-06,
|
| 355 |
+
"loss": 0.0008,
|
| 356 |
+
"step": 275
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 40.0,
|
| 360 |
+
"learning_rate": 4.9555555555555565e-06,
|
| 361 |
+
"loss": 0.0008,
|
| 362 |
+
"step": 280
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 40.71,
|
| 366 |
+
"learning_rate": 4.8444444444444446e-06,
|
| 367 |
+
"loss": 0.0008,
|
| 368 |
+
"step": 285
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 41.43,
|
| 372 |
+
"learning_rate": 4.7333333333333335e-06,
|
| 373 |
+
"loss": 0.0008,
|
| 374 |
+
"step": 290
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 42.14,
|
| 378 |
+
"learning_rate": 4.622222222222222e-06,
|
| 379 |
+
"loss": 0.0008,
|
| 380 |
+
"step": 295
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 42.86,
|
| 384 |
+
"learning_rate": 4.511111111111111e-06,
|
| 385 |
+
"loss": 0.0008,
|
| 386 |
+
"step": 300
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 42.86,
|
| 390 |
+
"eval_loss": 1.2581015825271606,
|
| 391 |
+
"eval_runtime": 1407.7801,
|
| 392 |
+
"eval_samples_per_second": 0.364,
|
| 393 |
+
"eval_steps_per_second": 0.023,
|
| 394 |
+
"eval_wer": 50.317796610169495,
|
| 395 |
+
"step": 300
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 43.57,
|
| 399 |
+
"learning_rate": 4.4e-06,
|
| 400 |
+
"loss": 0.0008,
|
| 401 |
+
"step": 305
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 44.29,
|
| 405 |
+
"learning_rate": 4.288888888888889e-06,
|
| 406 |
+
"loss": 0.0008,
|
| 407 |
+
"step": 310
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 45.0,
|
| 411 |
+
"learning_rate": 4.177777777777778e-06,
|
| 412 |
+
"loss": 0.0007,
|
| 413 |
+
"step": 315
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 45.71,
|
| 417 |
+
"learning_rate": 4.066666666666667e-06,
|
| 418 |
+
"loss": 0.0008,
|
| 419 |
+
"step": 320
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 46.43,
|
| 423 |
+
"learning_rate": 3.955555555555556e-06,
|
| 424 |
+
"loss": 0.0007,
|
| 425 |
+
"step": 325
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 47.14,
|
| 429 |
+
"learning_rate": 3.844444444444445e-06,
|
| 430 |
+
"loss": 0.0007,
|
| 431 |
+
"step": 330
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 47.86,
|
| 435 |
+
"learning_rate": 3.7333333333333337e-06,
|
| 436 |
+
"loss": 0.0007,
|
| 437 |
+
"step": 335
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 48.57,
|
| 441 |
+
"learning_rate": 3.6222222222222226e-06,
|
| 442 |
+
"loss": 0.0007,
|
| 443 |
+
"step": 340
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 49.29,
|
| 447 |
+
"learning_rate": 3.511111111111111e-06,
|
| 448 |
+
"loss": 0.0007,
|
| 449 |
+
"step": 345
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 50.0,
|
| 453 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 454 |
+
"loss": 0.0007,
|
| 455 |
+
"step": 350
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 50.71,
|
| 459 |
+
"learning_rate": 3.2888888888888894e-06,
|
| 460 |
+
"loss": 0.0007,
|
| 461 |
+
"step": 355
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 51.43,
|
| 465 |
+
"learning_rate": 3.177777777777778e-06,
|
| 466 |
+
"loss": 0.0007,
|
| 467 |
+
"step": 360
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 52.14,
|
| 471 |
+
"learning_rate": 3.066666666666667e-06,
|
| 472 |
+
"loss": 0.0007,
|
| 473 |
+
"step": 365
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 52.86,
|
| 477 |
+
"learning_rate": 2.955555555555556e-06,
|
| 478 |
+
"loss": 0.0007,
|
| 479 |
+
"step": 370
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 53.57,
|
| 483 |
+
"learning_rate": 2.8444444444444446e-06,
|
| 484 |
+
"loss": 0.0007,
|
| 485 |
+
"step": 375
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 54.29,
|
| 489 |
+
"learning_rate": 2.7333333333333336e-06,
|
| 490 |
+
"loss": 0.0007,
|
| 491 |
+
"step": 380
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 55.0,
|
| 495 |
+
"learning_rate": 2.6222222222222225e-06,
|
| 496 |
+
"loss": 0.0007,
|
| 497 |
+
"step": 385
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 55.71,
|
| 501 |
+
"learning_rate": 2.5111111111111114e-06,
|
| 502 |
+
"loss": 0.0007,
|
| 503 |
+
"step": 390
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 56.43,
|
| 507 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 508 |
+
"loss": 0.0007,
|
| 509 |
+
"step": 395
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 57.14,
|
| 513 |
+
"learning_rate": 2.2888888888888892e-06,
|
| 514 |
+
"loss": 0.0007,
|
| 515 |
+
"step": 400
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 57.14,
|
| 519 |
+
"eval_loss": 1.2849922180175781,
|
| 520 |
+
"eval_runtime": 1412.1951,
|
| 521 |
+
"eval_samples_per_second": 0.363,
|
| 522 |
+
"eval_steps_per_second": 0.023,
|
| 523 |
+
"eval_wer": 50.5523607748184,
|
| 524 |
+
"step": 400
|
| 525 |
+
}
|
| 526 |
+
],
|
| 527 |
+
"max_steps": 500,
|
| 528 |
+
"num_train_epochs": 72,
|
| 529 |
+
"total_flos": 2.513851947712512e+19,
|
| 530 |
+
"trial_name": null,
|
| 531 |
+
"trial_params": null
|
| 532 |
+
}
|
checkpoint-400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb79c84ed9cdcdf9027db44a0b593b1eebe5ee0791d07229cec597018c5ec2ca
|
| 3 |
+
size 3515
|
checkpoint-500/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
checkpoint-500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9f40a61d1f7d1a48ffe82745f696f7ee8b481ea7f57256d056da18cd8c9390c
|
| 3 |
+
size 6111428695
|
checkpoint-500/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-500/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1594cd9be21d3f1736cc43603de142c7df4ad15e4f3da42917b4058f4d3d58b9
|
| 3 |
+
size 3055754841
|
checkpoint-500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e0175230e09141322f0373e0b8c4cff27757ba01e18572280adf1ec24803b30
|
| 3 |
+
size 14575
|
checkpoint-500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbeed1f30e404a2e8d31dc0a7522a26a110838fe2ff3e24899f22a484324e7e5
|
| 3 |
+
size 557
|
checkpoint-500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:412f5825576caa68c361a1d662cc5a44fa7320a4b99333583b9052f378871539
|
| 3 |
+
size 627
|
checkpoint-500/trainer_state.json
ADDED
|
@@ -0,0 +1,661 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 71.42857142857143,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.71,
|
| 12 |
+
"learning_rate": 6.000000000000001e-07,
|
| 13 |
+
"loss": 2.8734,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.43,
|
| 18 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 19 |
+
"loss": 2.7517,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.14,
|
| 24 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 25 |
+
"loss": 2.2084,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.86,
|
| 30 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 31 |
+
"loss": 1.7218,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.57,
|
| 36 |
+
"learning_rate": 4.4e-06,
|
| 37 |
+
"loss": 1.3608,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.29,
|
| 42 |
+
"learning_rate": 5.400000000000001e-06,
|
| 43 |
+
"loss": 1.046,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 5.0,
|
| 48 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 49 |
+
"loss": 0.8672,
|
| 50 |
+
"step": 35
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.71,
|
| 54 |
+
"learning_rate": 7.4e-06,
|
| 55 |
+
"loss": 0.6926,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 6.43,
|
| 60 |
+
"learning_rate": 8.400000000000001e-06,
|
| 61 |
+
"loss": 0.5528,
|
| 62 |
+
"step": 45
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 7.14,
|
| 66 |
+
"learning_rate": 9.4e-06,
|
| 67 |
+
"loss": 0.4595,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 7.86,
|
| 72 |
+
"learning_rate": 9.955555555555556e-06,
|
| 73 |
+
"loss": 0.3677,
|
| 74 |
+
"step": 55
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 8.57,
|
| 78 |
+
"learning_rate": 9.844444444444446e-06,
|
| 79 |
+
"loss": 0.2566,
|
| 80 |
+
"step": 60
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 9.29,
|
| 84 |
+
"learning_rate": 9.733333333333334e-06,
|
| 85 |
+
"loss": 0.2176,
|
| 86 |
+
"step": 65
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 10.0,
|
| 90 |
+
"learning_rate": 9.622222222222222e-06,
|
| 91 |
+
"loss": 0.1514,
|
| 92 |
+
"step": 70
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.71,
|
| 96 |
+
"learning_rate": 9.511111111111112e-06,
|
| 97 |
+
"loss": 0.0989,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.43,
|
| 102 |
+
"learning_rate": 9.4e-06,
|
| 103 |
+
"loss": 0.0644,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 12.14,
|
| 108 |
+
"learning_rate": 9.28888888888889e-06,
|
| 109 |
+
"loss": 0.0445,
|
| 110 |
+
"step": 85
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.86,
|
| 114 |
+
"learning_rate": 9.17777777777778e-06,
|
| 115 |
+
"loss": 0.0279,
|
| 116 |
+
"step": 90
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 13.57,
|
| 120 |
+
"learning_rate": 9.066666666666667e-06,
|
| 121 |
+
"loss": 0.0191,
|
| 122 |
+
"step": 95
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 14.29,
|
| 126 |
+
"learning_rate": 8.955555555555555e-06,
|
| 127 |
+
"loss": 0.0139,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 14.29,
|
| 132 |
+
"eval_loss": 1.0302417278289795,
|
| 133 |
+
"eval_runtime": 1389.5619,
|
| 134 |
+
"eval_samples_per_second": 0.368,
|
| 135 |
+
"eval_steps_per_second": 0.023,
|
| 136 |
+
"eval_wer": 50.12106537530266,
|
| 137 |
+
"step": 100
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 15.0,
|
| 141 |
+
"learning_rate": 8.844444444444445e-06,
|
| 142 |
+
"loss": 0.0104,
|
| 143 |
+
"step": 105
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 15.71,
|
| 147 |
+
"learning_rate": 8.733333333333333e-06,
|
| 148 |
+
"loss": 0.0091,
|
| 149 |
+
"step": 110
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.43,
|
| 153 |
+
"learning_rate": 8.622222222222223e-06,
|
| 154 |
+
"loss": 0.0062,
|
| 155 |
+
"step": 115
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 17.14,
|
| 159 |
+
"learning_rate": 8.511111111111113e-06,
|
| 160 |
+
"loss": 0.0056,
|
| 161 |
+
"step": 120
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 17.86,
|
| 165 |
+
"learning_rate": 8.400000000000001e-06,
|
| 166 |
+
"loss": 0.0052,
|
| 167 |
+
"step": 125
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 18.57,
|
| 171 |
+
"learning_rate": 8.288888888888889e-06,
|
| 172 |
+
"loss": 0.0047,
|
| 173 |
+
"step": 130
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 19.29,
|
| 177 |
+
"learning_rate": 8.177777777777779e-06,
|
| 178 |
+
"loss": 0.0049,
|
| 179 |
+
"step": 135
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 20.0,
|
| 183 |
+
"learning_rate": 8.066666666666667e-06,
|
| 184 |
+
"loss": 0.0042,
|
| 185 |
+
"step": 140
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 20.71,
|
| 189 |
+
"learning_rate": 7.955555555555557e-06,
|
| 190 |
+
"loss": 0.0033,
|
| 191 |
+
"step": 145
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 21.43,
|
| 195 |
+
"learning_rate": 7.844444444444446e-06,
|
| 196 |
+
"loss": 0.0028,
|
| 197 |
+
"step": 150
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 22.14,
|
| 201 |
+
"learning_rate": 7.733333333333334e-06,
|
| 202 |
+
"loss": 0.002,
|
| 203 |
+
"step": 155
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 22.86,
|
| 207 |
+
"learning_rate": 7.622222222222223e-06,
|
| 208 |
+
"loss": 0.0018,
|
| 209 |
+
"step": 160
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 23.57,
|
| 213 |
+
"learning_rate": 7.511111111111111e-06,
|
| 214 |
+
"loss": 0.0016,
|
| 215 |
+
"step": 165
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 24.29,
|
| 219 |
+
"learning_rate": 7.4e-06,
|
| 220 |
+
"loss": 0.0014,
|
| 221 |
+
"step": 170
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 25.0,
|
| 225 |
+
"learning_rate": 7.28888888888889e-06,
|
| 226 |
+
"loss": 0.0013,
|
| 227 |
+
"step": 175
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 25.71,
|
| 231 |
+
"learning_rate": 7.177777777777778e-06,
|
| 232 |
+
"loss": 0.0013,
|
| 233 |
+
"step": 180
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 26.43,
|
| 237 |
+
"learning_rate": 7.066666666666667e-06,
|
| 238 |
+
"loss": 0.0012,
|
| 239 |
+
"step": 185
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 27.14,
|
| 243 |
+
"learning_rate": 6.955555555555557e-06,
|
| 244 |
+
"loss": 0.0011,
|
| 245 |
+
"step": 190
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 27.86,
|
| 249 |
+
"learning_rate": 6.844444444444445e-06,
|
| 250 |
+
"loss": 0.0011,
|
| 251 |
+
"step": 195
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 28.57,
|
| 255 |
+
"learning_rate": 6.733333333333334e-06,
|
| 256 |
+
"loss": 0.0011,
|
| 257 |
+
"step": 200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 28.57,
|
| 261 |
+
"eval_loss": 1.2129117250442505,
|
| 262 |
+
"eval_runtime": 1370.5005,
|
| 263 |
+
"eval_samples_per_second": 0.374,
|
| 264 |
+
"eval_steps_per_second": 0.023,
|
| 265 |
+
"eval_wer": 49.78056900726393,
|
| 266 |
+
"step": 200
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 29.29,
|
| 270 |
+
"learning_rate": 6.6222222222222236e-06,
|
| 271 |
+
"loss": 0.001,
|
| 272 |
+
"step": 205
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 30.0,
|
| 276 |
+
"learning_rate": 6.511111111111112e-06,
|
| 277 |
+
"loss": 0.001,
|
| 278 |
+
"step": 210
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 30.71,
|
| 282 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 283 |
+
"loss": 0.001,
|
| 284 |
+
"step": 215
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 31.43,
|
| 288 |
+
"learning_rate": 6.28888888888889e-06,
|
| 289 |
+
"loss": 0.001,
|
| 290 |
+
"step": 220
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 32.14,
|
| 294 |
+
"learning_rate": 6.177777777777778e-06,
|
| 295 |
+
"loss": 0.001,
|
| 296 |
+
"step": 225
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 32.86,
|
| 300 |
+
"learning_rate": 6.066666666666667e-06,
|
| 301 |
+
"loss": 0.0009,
|
| 302 |
+
"step": 230
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 33.57,
|
| 306 |
+
"learning_rate": 5.955555555555555e-06,
|
| 307 |
+
"loss": 0.0009,
|
| 308 |
+
"step": 235
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 34.29,
|
| 312 |
+
"learning_rate": 5.844444444444445e-06,
|
| 313 |
+
"loss": 0.0009,
|
| 314 |
+
"step": 240
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 35.0,
|
| 318 |
+
"learning_rate": 5.733333333333334e-06,
|
| 319 |
+
"loss": 0.0009,
|
| 320 |
+
"step": 245
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 35.71,
|
| 324 |
+
"learning_rate": 5.622222222222222e-06,
|
| 325 |
+
"loss": 0.0009,
|
| 326 |
+
"step": 250
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 36.43,
|
| 330 |
+
"learning_rate": 5.511111111111112e-06,
|
| 331 |
+
"loss": 0.0009,
|
| 332 |
+
"step": 255
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 37.14,
|
| 336 |
+
"learning_rate": 5.400000000000001e-06,
|
| 337 |
+
"loss": 0.0008,
|
| 338 |
+
"step": 260
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 37.86,
|
| 342 |
+
"learning_rate": 5.288888888888889e-06,
|
| 343 |
+
"loss": 0.0008,
|
| 344 |
+
"step": 265
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 38.57,
|
| 348 |
+
"learning_rate": 5.177777777777779e-06,
|
| 349 |
+
"loss": 0.0008,
|
| 350 |
+
"step": 270
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 39.29,
|
| 354 |
+
"learning_rate": 5.0666666666666676e-06,
|
| 355 |
+
"loss": 0.0008,
|
| 356 |
+
"step": 275
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 40.0,
|
| 360 |
+
"learning_rate": 4.9555555555555565e-06,
|
| 361 |
+
"loss": 0.0008,
|
| 362 |
+
"step": 280
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 40.71,
|
| 366 |
+
"learning_rate": 4.8444444444444446e-06,
|
| 367 |
+
"loss": 0.0008,
|
| 368 |
+
"step": 285
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 41.43,
|
| 372 |
+
"learning_rate": 4.7333333333333335e-06,
|
| 373 |
+
"loss": 0.0008,
|
| 374 |
+
"step": 290
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 42.14,
|
| 378 |
+
"learning_rate": 4.622222222222222e-06,
|
| 379 |
+
"loss": 0.0008,
|
| 380 |
+
"step": 295
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 42.86,
|
| 384 |
+
"learning_rate": 4.511111111111111e-06,
|
| 385 |
+
"loss": 0.0008,
|
| 386 |
+
"step": 300
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 42.86,
|
| 390 |
+
"eval_loss": 1.2581015825271606,
|
| 391 |
+
"eval_runtime": 1407.7801,
|
| 392 |
+
"eval_samples_per_second": 0.364,
|
| 393 |
+
"eval_steps_per_second": 0.023,
|
| 394 |
+
"eval_wer": 50.317796610169495,
|
| 395 |
+
"step": 300
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 43.57,
|
| 399 |
+
"learning_rate": 4.4e-06,
|
| 400 |
+
"loss": 0.0008,
|
| 401 |
+
"step": 305
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 44.29,
|
| 405 |
+
"learning_rate": 4.288888888888889e-06,
|
| 406 |
+
"loss": 0.0008,
|
| 407 |
+
"step": 310
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 45.0,
|
| 411 |
+
"learning_rate": 4.177777777777778e-06,
|
| 412 |
+
"loss": 0.0007,
|
| 413 |
+
"step": 315
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 45.71,
|
| 417 |
+
"learning_rate": 4.066666666666667e-06,
|
| 418 |
+
"loss": 0.0008,
|
| 419 |
+
"step": 320
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 46.43,
|
| 423 |
+
"learning_rate": 3.955555555555556e-06,
|
| 424 |
+
"loss": 0.0007,
|
| 425 |
+
"step": 325
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 47.14,
|
| 429 |
+
"learning_rate": 3.844444444444445e-06,
|
| 430 |
+
"loss": 0.0007,
|
| 431 |
+
"step": 330
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 47.86,
|
| 435 |
+
"learning_rate": 3.7333333333333337e-06,
|
| 436 |
+
"loss": 0.0007,
|
| 437 |
+
"step": 335
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 48.57,
|
| 441 |
+
"learning_rate": 3.6222222222222226e-06,
|
| 442 |
+
"loss": 0.0007,
|
| 443 |
+
"step": 340
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 49.29,
|
| 447 |
+
"learning_rate": 3.511111111111111e-06,
|
| 448 |
+
"loss": 0.0007,
|
| 449 |
+
"step": 345
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 50.0,
|
| 453 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 454 |
+
"loss": 0.0007,
|
| 455 |
+
"step": 350
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 50.71,
|
| 459 |
+
"learning_rate": 3.2888888888888894e-06,
|
| 460 |
+
"loss": 0.0007,
|
| 461 |
+
"step": 355
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 51.43,
|
| 465 |
+
"learning_rate": 3.177777777777778e-06,
|
| 466 |
+
"loss": 0.0007,
|
| 467 |
+
"step": 360
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 52.14,
|
| 471 |
+
"learning_rate": 3.066666666666667e-06,
|
| 472 |
+
"loss": 0.0007,
|
| 473 |
+
"step": 365
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 52.86,
|
| 477 |
+
"learning_rate": 2.955555555555556e-06,
|
| 478 |
+
"loss": 0.0007,
|
| 479 |
+
"step": 370
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 53.57,
|
| 483 |
+
"learning_rate": 2.8444444444444446e-06,
|
| 484 |
+
"loss": 0.0007,
|
| 485 |
+
"step": 375
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 54.29,
|
| 489 |
+
"learning_rate": 2.7333333333333336e-06,
|
| 490 |
+
"loss": 0.0007,
|
| 491 |
+
"step": 380
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 55.0,
|
| 495 |
+
"learning_rate": 2.6222222222222225e-06,
|
| 496 |
+
"loss": 0.0007,
|
| 497 |
+
"step": 385
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 55.71,
|
| 501 |
+
"learning_rate": 2.5111111111111114e-06,
|
| 502 |
+
"loss": 0.0007,
|
| 503 |
+
"step": 390
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 56.43,
|
| 507 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 508 |
+
"loss": 0.0007,
|
| 509 |
+
"step": 395
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 57.14,
|
| 513 |
+
"learning_rate": 2.2888888888888892e-06,
|
| 514 |
+
"loss": 0.0007,
|
| 515 |
+
"step": 400
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 57.14,
|
| 519 |
+
"eval_loss": 1.2849922180175781,
|
| 520 |
+
"eval_runtime": 1412.1951,
|
| 521 |
+
"eval_samples_per_second": 0.363,
|
| 522 |
+
"eval_steps_per_second": 0.023,
|
| 523 |
+
"eval_wer": 50.5523607748184,
|
| 524 |
+
"step": 400
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 57.86,
|
| 528 |
+
"learning_rate": 2.1777777777777777e-06,
|
| 529 |
+
"loss": 0.0007,
|
| 530 |
+
"step": 405
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 58.57,
|
| 534 |
+
"learning_rate": 2.0666666666666666e-06,
|
| 535 |
+
"loss": 0.0007,
|
| 536 |
+
"step": 410
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 59.29,
|
| 540 |
+
"learning_rate": 1.955555555555556e-06,
|
| 541 |
+
"loss": 0.0006,
|
| 542 |
+
"step": 415
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 60.0,
|
| 546 |
+
"learning_rate": 1.8444444444444445e-06,
|
| 547 |
+
"loss": 0.0007,
|
| 548 |
+
"step": 420
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 60.71,
|
| 552 |
+
"learning_rate": 1.7333333333333336e-06,
|
| 553 |
+
"loss": 0.0007,
|
| 554 |
+
"step": 425
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 61.43,
|
| 558 |
+
"learning_rate": 1.6222222222222223e-06,
|
| 559 |
+
"loss": 0.0006,
|
| 560 |
+
"step": 430
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 62.14,
|
| 564 |
+
"learning_rate": 1.5111111111111112e-06,
|
| 565 |
+
"loss": 0.0007,
|
| 566 |
+
"step": 435
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 62.86,
|
| 570 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 571 |
+
"loss": 0.0007,
|
| 572 |
+
"step": 440
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 63.57,
|
| 576 |
+
"learning_rate": 1.288888888888889e-06,
|
| 577 |
+
"loss": 0.0007,
|
| 578 |
+
"step": 445
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 64.29,
|
| 582 |
+
"learning_rate": 1.1777777777777778e-06,
|
| 583 |
+
"loss": 0.0006,
|
| 584 |
+
"step": 450
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 65.0,
|
| 588 |
+
"learning_rate": 1.066666666666667e-06,
|
| 589 |
+
"loss": 0.0006,
|
| 590 |
+
"step": 455
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 65.71,
|
| 594 |
+
"learning_rate": 9.555555555555556e-07,
|
| 595 |
+
"loss": 0.0006,
|
| 596 |
+
"step": 460
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 66.43,
|
| 600 |
+
"learning_rate": 8.444444444444445e-07,
|
| 601 |
+
"loss": 0.0006,
|
| 602 |
+
"step": 465
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 67.14,
|
| 606 |
+
"learning_rate": 7.333333333333334e-07,
|
| 607 |
+
"loss": 0.0006,
|
| 608 |
+
"step": 470
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 67.86,
|
| 612 |
+
"learning_rate": 6.222222222222223e-07,
|
| 613 |
+
"loss": 0.0007,
|
| 614 |
+
"step": 475
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 68.57,
|
| 618 |
+
"learning_rate": 5.111111111111112e-07,
|
| 619 |
+
"loss": 0.0006,
|
| 620 |
+
"step": 480
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 69.29,
|
| 624 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 625 |
+
"loss": 0.0006,
|
| 626 |
+
"step": 485
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 70.0,
|
| 630 |
+
"learning_rate": 2.888888888888889e-07,
|
| 631 |
+
"loss": 0.0006,
|
| 632 |
+
"step": 490
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 70.71,
|
| 636 |
+
"learning_rate": 1.777777777777778e-07,
|
| 637 |
+
"loss": 0.0006,
|
| 638 |
+
"step": 495
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 71.43,
|
| 642 |
+
"learning_rate": 6.666666666666668e-08,
|
| 643 |
+
"loss": 0.0007,
|
| 644 |
+
"step": 500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 71.43,
|
| 648 |
+
"eval_loss": 1.2950359582901,
|
| 649 |
+
"eval_runtime": 1407.5689,
|
| 650 |
+
"eval_samples_per_second": 0.364,
|
| 651 |
+
"eval_steps_per_second": 0.023,
|
| 652 |
+
"eval_wer": 50.6431598062954,
|
| 653 |
+
"step": 500
|
| 654 |
+
}
|
| 655 |
+
],
|
| 656 |
+
"max_steps": 500,
|
| 657 |
+
"num_train_epochs": 72,
|
| 658 |
+
"total_flos": 3.142748691726336e+19,
|
| 659 |
+
"trial_name": null,
|
| 660 |
+
"trial_params": null
|
| 661 |
+
}
|
checkpoint-500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb79c84ed9cdcdf9027db44a0b593b1eebe5ee0791d07229cec597018c5ec2ca
|
| 3 |
+
size 3515
|
config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 71.43,
|
| 3 |
+
"eval_loss": 1.2950359582901,
|
| 4 |
+
"eval_runtime": 1406.7383,
|
| 5 |
+
"eval_samples_per_second": 0.364,
|
| 6 |
+
"eval_steps_per_second": 0.023,
|
| 7 |
+
"eval_wer": 50.6431598062954
|
| 8 |
+
}
|