Training in progress, step 700
Browse files- checkpoint-700/config.json +42 -0
- checkpoint-700/optimizer.pt +3 -0
- checkpoint-700/preprocessor_config.json +0 -0
- checkpoint-700/pytorch_model.bin +3 -0
- checkpoint-700/rng_state.pth +3 -0
- checkpoint-700/scaler.pt +3 -0
- checkpoint-700/scheduler.pt +3 -0
- checkpoint-700/trainer_state.json +506 -0
- checkpoint-700/training_args.bin +3 -0
- pytorch_model.bin +1 -1
- runs/Dec20_23-04-41_129-146-32-172/1671579286.7936676/events.out.tfevents.1671579286.129-146-32-172.156136.3 +3 -0
- runs/Dec20_23-04-41_129-146-32-172/events.out.tfevents.1671577490.129-146-32-172.156136.0 +2 -2
- runs/Dec20_23-04-41_129-146-32-172/events.out.tfevents.1671579286.129-146-32-172.156136.2 +3 -0
- runs/Dec20_23-41-43_129-146-32-172/1671579713.6389546/events.out.tfevents.1671579713.129-146-32-172.157116.1 +3 -0
- runs/Dec20_23-41-43_129-146-32-172/1671579714.680834/events.out.tfevents.1671579714.129-146-32-172.157116.3 +3 -0
- runs/Dec20_23-41-43_129-146-32-172/events.out.tfevents.1671579713.129-146-32-172.157116.0 +3 -0
- runs/Dec20_23-41-43_129-146-32-172/events.out.tfevents.1671579714.129-146-32-172.157116.2 +3 -0
- runs/Dec20_23-45-37_129-146-32-172/1671579948.0116377/events.out.tfevents.1671579948.129-146-32-172.157273.1 +3 -0
- runs/Dec20_23-45-37_129-146-32-172/events.out.tfevents.1671579948.129-146-32-172.157273.0 +3 -0
- training_args.bin +1 -1
- whisper_small_ps_augmented.py +1 -1
checkpoint-700/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-small",
|
| 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": 768,
|
| 15 |
+
"decoder_attention_heads": 12,
|
| 16 |
+
"decoder_ffn_dim": 3072,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 12,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 12,
|
| 22 |
+
"encoder_ffn_dim": 3072,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 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": 12,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"suppress_tokens": [],
|
| 38 |
+
"torch_dtype": "float32",
|
| 39 |
+
"transformers_version": "4.26.0.dev0",
|
| 40 |
+
"use_cache": false,
|
| 41 |
+
"vocab_size": 51865
|
| 42 |
+
}
|
checkpoint-700/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5985bf1c6cd1b9111c71913d4a28958a07970c52ce6cc7d091abb2727a4a67c5
|
| 3 |
+
size 1934161093
|
checkpoint-700/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-700/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30b85a1146393e276dd0a4b5ab17e7366d30b253434accf996fa0c68be77e7f5
|
| 3 |
+
size 967102601
|
checkpoint-700/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6a5100f14bc7224ca972cfe6c6c076a5f2e5025521f8076964781a2f074b628
|
| 3 |
+
size 14575
|
checkpoint-700/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10d91726a9390d4d710621633e528427453ab73445a6d233735f9a37b9c09789
|
| 3 |
+
size 557
|
checkpoint-700/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db9f3ccf26771ce344864b125486879d4a70391a50b5fbb51d75527361400177
|
| 3 |
+
size 627
|
checkpoint-700/trainer_state.json
ADDED
|
@@ -0,0 +1,506 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 49.03147699757869,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-400",
|
| 4 |
+
"epoch": 16.658823529411766,
|
| 5 |
+
"global_step": 700,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.24,
|
| 12 |
+
"learning_rate": 2.666666666666667e-06,
|
| 13 |
+
"loss": 4.3134,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.47,
|
| 18 |
+
"learning_rate": 6e-06,
|
| 19 |
+
"loss": 2.9329,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.71,
|
| 24 |
+
"learning_rate": 9.333333333333334e-06,
|
| 25 |
+
"loss": 2.0584,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.94,
|
| 30 |
+
"learning_rate": 9.829787234042554e-06,
|
| 31 |
+
"loss": 1.566,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.19,
|
| 36 |
+
"learning_rate": 9.617021276595745e-06,
|
| 37 |
+
"loss": 1.3777,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.42,
|
| 42 |
+
"learning_rate": 9.404255319148937e-06,
|
| 43 |
+
"loss": 1.1469,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.66,
|
| 48 |
+
"learning_rate": 9.191489361702128e-06,
|
| 49 |
+
"loss": 1.0638,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.89,
|
| 54 |
+
"learning_rate": 8.97872340425532e-06,
|
| 55 |
+
"loss": 0.9974,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.14,
|
| 60 |
+
"learning_rate": 8.765957446808512e-06,
|
| 61 |
+
"loss": 0.9615,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.38,
|
| 66 |
+
"learning_rate": 8.553191489361703e-06,
|
| 67 |
+
"loss": 0.8262,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.38,
|
| 72 |
+
"eval_cer": 149.33157314260887,
|
| 73 |
+
"eval_loss": 0.8188337683677673,
|
| 74 |
+
"eval_runtime": 459.9395,
|
| 75 |
+
"eval_samples_per_second": 1.113,
|
| 76 |
+
"eval_steps_per_second": 0.139,
|
| 77 |
+
"eval_wer": 146.85230024213075,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.61,
|
| 82 |
+
"learning_rate": 8.340425531914894e-06,
|
| 83 |
+
"loss": 0.7986,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 2.85,
|
| 88 |
+
"learning_rate": 8.127659574468085e-06,
|
| 89 |
+
"loss": 0.7707,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 3.09,
|
| 94 |
+
"learning_rate": 7.914893617021278e-06,
|
| 95 |
+
"loss": 0.7623,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.33,
|
| 100 |
+
"learning_rate": 7.702127659574469e-06,
|
| 101 |
+
"loss": 0.6451,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 3.56,
|
| 106 |
+
"learning_rate": 7.48936170212766e-06,
|
| 107 |
+
"loss": 0.6205,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 3.8,
|
| 112 |
+
"learning_rate": 7.2765957446808524e-06,
|
| 113 |
+
"loss": 0.6097,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 4.05,
|
| 118 |
+
"learning_rate": 7.0638297872340434e-06,
|
| 119 |
+
"loss": 0.6217,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 4.28,
|
| 124 |
+
"learning_rate": 6.8510638297872344e-06,
|
| 125 |
+
"loss": 0.4584,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 4.52,
|
| 130 |
+
"learning_rate": 6.6382978723404254e-06,
|
| 131 |
+
"loss": 0.4929,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.75,
|
| 136 |
+
"learning_rate": 6.425531914893618e-06,
|
| 137 |
+
"loss": 0.4843,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 4.75,
|
| 142 |
+
"eval_cer": 22.099862972494236,
|
| 143 |
+
"eval_loss": 0.6699215173721313,
|
| 144 |
+
"eval_runtime": 1062.3991,
|
| 145 |
+
"eval_samples_per_second": 0.482,
|
| 146 |
+
"eval_steps_per_second": 0.06,
|
| 147 |
+
"eval_wer": 52.66343825665859,
|
| 148 |
+
"step": 200
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 4.99,
|
| 152 |
+
"learning_rate": 6.212765957446809e-06,
|
| 153 |
+
"loss": 0.4678,
|
| 154 |
+
"step": 210
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.24,
|
| 158 |
+
"learning_rate": 6e-06,
|
| 159 |
+
"loss": 0.4014,
|
| 160 |
+
"step": 220
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 5.47,
|
| 164 |
+
"learning_rate": 5.787234042553191e-06,
|
| 165 |
+
"loss": 0.3926,
|
| 166 |
+
"step": 230
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 5.71,
|
| 170 |
+
"learning_rate": 5.574468085106384e-06,
|
| 171 |
+
"loss": 0.3806,
|
| 172 |
+
"step": 240
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 5.94,
|
| 176 |
+
"learning_rate": 5.361702127659575e-06,
|
| 177 |
+
"loss": 0.3855,
|
| 178 |
+
"step": 250
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 6.19,
|
| 182 |
+
"learning_rate": 5.148936170212766e-06,
|
| 183 |
+
"loss": 0.3478,
|
| 184 |
+
"step": 260
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 6.42,
|
| 188 |
+
"learning_rate": 4.936170212765958e-06,
|
| 189 |
+
"loss": 0.3078,
|
| 190 |
+
"step": 270
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.66,
|
| 194 |
+
"learning_rate": 4.7234042553191496e-06,
|
| 195 |
+
"loss": 0.3147,
|
| 196 |
+
"step": 280
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 6.89,
|
| 200 |
+
"learning_rate": 4.5106382978723406e-06,
|
| 201 |
+
"loss": 0.3213,
|
| 202 |
+
"step": 290
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 7.14,
|
| 206 |
+
"learning_rate": 4.297872340425532e-06,
|
| 207 |
+
"loss": 0.287,
|
| 208 |
+
"step": 300
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 7.14,
|
| 212 |
+
"eval_cer": 20.677784833394604,
|
| 213 |
+
"eval_loss": 0.6913720369338989,
|
| 214 |
+
"eval_runtime": 1003.603,
|
| 215 |
+
"eval_samples_per_second": 0.51,
|
| 216 |
+
"eval_steps_per_second": 0.064,
|
| 217 |
+
"eval_wer": 50.13619854721549,
|
| 218 |
+
"step": 300
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 7.38,
|
| 222 |
+
"learning_rate": 4.085106382978723e-06,
|
| 223 |
+
"loss": 0.2602,
|
| 224 |
+
"step": 310
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 7.61,
|
| 228 |
+
"learning_rate": 3.872340425531915e-06,
|
| 229 |
+
"loss": 0.2612,
|
| 230 |
+
"step": 320
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 7.85,
|
| 234 |
+
"learning_rate": 3.6595744680851063e-06,
|
| 235 |
+
"loss": 0.2618,
|
| 236 |
+
"step": 330
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 8.09,
|
| 240 |
+
"learning_rate": 3.446808510638298e-06,
|
| 241 |
+
"loss": 0.2494,
|
| 242 |
+
"step": 340
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 8.33,
|
| 246 |
+
"learning_rate": 3.23404255319149e-06,
|
| 247 |
+
"loss": 0.2123,
|
| 248 |
+
"step": 350
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.56,
|
| 252 |
+
"learning_rate": 3.021276595744681e-06,
|
| 253 |
+
"loss": 0.2173,
|
| 254 |
+
"step": 360
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 8.8,
|
| 258 |
+
"learning_rate": 2.808510638297873e-06,
|
| 259 |
+
"loss": 0.2093,
|
| 260 |
+
"step": 370
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 9.05,
|
| 264 |
+
"learning_rate": 2.595744680851064e-06,
|
| 265 |
+
"loss": 0.2209,
|
| 266 |
+
"step": 380
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 9.28,
|
| 270 |
+
"learning_rate": 2.3829787234042557e-06,
|
| 271 |
+
"loss": 0.1742,
|
| 272 |
+
"step": 390
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 9.52,
|
| 276 |
+
"learning_rate": 2.170212765957447e-06,
|
| 277 |
+
"loss": 0.1828,
|
| 278 |
+
"step": 400
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 9.52,
|
| 282 |
+
"eval_cer": 20.460546104742487,
|
| 283 |
+
"eval_loss": 0.7369570136070251,
|
| 284 |
+
"eval_runtime": 1004.0546,
|
| 285 |
+
"eval_samples_per_second": 0.51,
|
| 286 |
+
"eval_steps_per_second": 0.064,
|
| 287 |
+
"eval_wer": 49.03147699757869,
|
| 288 |
+
"step": 400
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 9.75,
|
| 292 |
+
"learning_rate": 1.9574468085106385e-06,
|
| 293 |
+
"loss": 0.1902,
|
| 294 |
+
"step": 410
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 9.99,
|
| 298 |
+
"learning_rate": 1.74468085106383e-06,
|
| 299 |
+
"loss": 0.1809,
|
| 300 |
+
"step": 420
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 10.24,
|
| 304 |
+
"learning_rate": 1.5319148936170214e-06,
|
| 305 |
+
"loss": 0.1705,
|
| 306 |
+
"step": 430
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.47,
|
| 310 |
+
"learning_rate": 1.3191489361702128e-06,
|
| 311 |
+
"loss": 0.1593,
|
| 312 |
+
"step": 440
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 10.71,
|
| 316 |
+
"learning_rate": 1.1063829787234042e-06,
|
| 317 |
+
"loss": 0.1596,
|
| 318 |
+
"step": 450
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 10.94,
|
| 322 |
+
"learning_rate": 8.936170212765959e-07,
|
| 323 |
+
"loss": 0.161,
|
| 324 |
+
"step": 460
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 11.19,
|
| 328 |
+
"learning_rate": 6.808510638297873e-07,
|
| 329 |
+
"loss": 0.1594,
|
| 330 |
+
"step": 470
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 11.42,
|
| 334 |
+
"learning_rate": 4.6808510638297873e-07,
|
| 335 |
+
"loss": 0.1436,
|
| 336 |
+
"step": 480
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 11.66,
|
| 340 |
+
"learning_rate": 2.553191489361702e-07,
|
| 341 |
+
"loss": 0.1517,
|
| 342 |
+
"step": 490
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 11.89,
|
| 346 |
+
"learning_rate": 4.2553191489361707e-08,
|
| 347 |
+
"loss": 0.1493,
|
| 348 |
+
"step": 500
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 11.89,
|
| 352 |
+
"eval_cer": 20.8064570034424,
|
| 353 |
+
"eval_loss": 0.7532366514205933,
|
| 354 |
+
"eval_runtime": 925.6199,
|
| 355 |
+
"eval_samples_per_second": 0.553,
|
| 356 |
+
"eval_steps_per_second": 0.069,
|
| 357 |
+
"eval_wer": 49.583837772397096,
|
| 358 |
+
"step": 500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 12.14,
|
| 362 |
+
"learning_rate": 1.6140350877192983e-06,
|
| 363 |
+
"loss": 0.1527,
|
| 364 |
+
"step": 510
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 12.38,
|
| 368 |
+
"learning_rate": 1.4385964912280704e-06,
|
| 369 |
+
"loss": 0.1453,
|
| 370 |
+
"step": 520
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 12.61,
|
| 374 |
+
"learning_rate": 1.2631578947368422e-06,
|
| 375 |
+
"loss": 0.1408,
|
| 376 |
+
"step": 530
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 12.85,
|
| 380 |
+
"learning_rate": 1.0877192982456142e-06,
|
| 381 |
+
"loss": 0.1413,
|
| 382 |
+
"step": 540
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 13.09,
|
| 386 |
+
"learning_rate": 9.122807017543861e-07,
|
| 387 |
+
"loss": 0.142,
|
| 388 |
+
"step": 550
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 13.33,
|
| 392 |
+
"learning_rate": 7.368421052631579e-07,
|
| 393 |
+
"loss": 0.1351,
|
| 394 |
+
"step": 560
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 13.56,
|
| 398 |
+
"learning_rate": 5.614035087719299e-07,
|
| 399 |
+
"loss": 0.1288,
|
| 400 |
+
"step": 570
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 13.8,
|
| 404 |
+
"learning_rate": 3.859649122807018e-07,
|
| 405 |
+
"loss": 0.1308,
|
| 406 |
+
"step": 580
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 14.05,
|
| 410 |
+
"learning_rate": 2.105263157894737e-07,
|
| 411 |
+
"loss": 0.1357,
|
| 412 |
+
"step": 590
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 14.28,
|
| 416 |
+
"learning_rate": 3.508771929824562e-08,
|
| 417 |
+
"loss": 0.1249,
|
| 418 |
+
"step": 600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 14.28,
|
| 422 |
+
"eval_cer": 21.225894856455334,
|
| 423 |
+
"eval_loss": 0.7713016271591187,
|
| 424 |
+
"eval_runtime": 1016.0283,
|
| 425 |
+
"eval_samples_per_second": 0.504,
|
| 426 |
+
"eval_steps_per_second": 0.063,
|
| 427 |
+
"eval_wer": 49.69733656174334,
|
| 428 |
+
"step": 600
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 14.52,
|
| 432 |
+
"learning_rate": 1.373134328358209e-06,
|
| 433 |
+
"loss": 0.1206,
|
| 434 |
+
"step": 610
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 14.75,
|
| 438 |
+
"learning_rate": 1.2238805970149254e-06,
|
| 439 |
+
"loss": 0.1227,
|
| 440 |
+
"step": 620
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 14.99,
|
| 444 |
+
"learning_rate": 1.0746268656716418e-06,
|
| 445 |
+
"loss": 0.1312,
|
| 446 |
+
"step": 630
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 15.24,
|
| 450 |
+
"learning_rate": 9.253731343283583e-07,
|
| 451 |
+
"loss": 0.1208,
|
| 452 |
+
"step": 640
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 15.47,
|
| 456 |
+
"learning_rate": 7.761194029850747e-07,
|
| 457 |
+
"loss": 0.1132,
|
| 458 |
+
"step": 650
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 15.71,
|
| 462 |
+
"learning_rate": 6.26865671641791e-07,
|
| 463 |
+
"loss": 0.1186,
|
| 464 |
+
"step": 660
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 15.94,
|
| 468 |
+
"learning_rate": 4.776119402985075e-07,
|
| 469 |
+
"loss": 0.1177,
|
| 470 |
+
"step": 670
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 16.19,
|
| 474 |
+
"learning_rate": 3.283582089552239e-07,
|
| 475 |
+
"loss": 0.1205,
|
| 476 |
+
"step": 680
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 16.42,
|
| 480 |
+
"learning_rate": 1.7910447761194033e-07,
|
| 481 |
+
"loss": 0.111,
|
| 482 |
+
"step": 690
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 16.66,
|
| 486 |
+
"learning_rate": 2.985074626865672e-08,
|
| 487 |
+
"loss": 0.1146,
|
| 488 |
+
"step": 700
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 16.66,
|
| 492 |
+
"eval_cer": 20.823167674877176,
|
| 493 |
+
"eval_loss": 0.7877429127693176,
|
| 494 |
+
"eval_runtime": 1023.8344,
|
| 495 |
+
"eval_samples_per_second": 0.5,
|
| 496 |
+
"eval_steps_per_second": 0.063,
|
| 497 |
+
"eval_wer": 49.56870460048426,
|
| 498 |
+
"step": 700
|
| 499 |
+
}
|
| 500 |
+
],
|
| 501 |
+
"max_steps": 700,
|
| 502 |
+
"num_train_epochs": 17,
|
| 503 |
+
"total_flos": 1.303482536165376e+19,
|
| 504 |
+
"trial_name": null,
|
| 505 |
+
"trial_params": null
|
| 506 |
+
}
|
checkpoint-700/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ac880f769c6c8c222243cd146ac46a2cb7368d0196625844f7aa9d536c159d8
|
| 3 |
+
size 3643
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 967102601
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30b85a1146393e276dd0a4b5ab17e7366d30b253434accf996fa0c68be77e7f5
|
| 3 |
size 967102601
|
runs/Dec20_23-04-41_129-146-32-172/1671579286.7936676/events.out.tfevents.1671579286.129-146-32-172.156136.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7205383401960bd3a122a146bb01000e78f61a02ad028a732fc07d2c6ea6c6e2
|
| 3 |
+
size 5862
|
runs/Dec20_23-04-41_129-146-32-172/events.out.tfevents.1671577490.129-146-32-172.156136.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3dff3f7c6b8938cc51c18705bf4766c3d8829b13d58ce7c8bd51c2d00e16f4ee
|
| 3 |
+
size 6576
|
runs/Dec20_23-04-41_129-146-32-172/events.out.tfevents.1671579286.129-146-32-172.156136.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d494bb809213112f01c253438ee792b9141f4fe3a09de1df5f00bceb3af45abb
|
| 3 |
+
size 4287
|
runs/Dec20_23-41-43_129-146-32-172/1671579713.6389546/events.out.tfevents.1671579713.129-146-32-172.157116.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a502219f160f7d2b73601b9e2c25690c3f93c9a4472a623106fd3aa7a032d8f1
|
| 3 |
+
size 5862
|
runs/Dec20_23-41-43_129-146-32-172/1671579714.680834/events.out.tfevents.1671579714.129-146-32-172.157116.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ec4207062238f88633e0ddc2a0d84dd3be7c2962d16978a632ee1386142e164
|
| 3 |
+
size 5862
|
runs/Dec20_23-41-43_129-146-32-172/events.out.tfevents.1671579713.129-146-32-172.157116.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5edf2c3834c08656c430f6018057f266f3baa0e870c3692cdc401bb28f65ae3d
|
| 3 |
+
size 4641
|
runs/Dec20_23-41-43_129-146-32-172/events.out.tfevents.1671579714.129-146-32-172.157116.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a52645dbf6c20a433185717c37abfd7f2fbb9002ae87e1e6a2b8b3c4470d63d
|
| 3 |
+
size 4441
|
runs/Dec20_23-45-37_129-146-32-172/1671579948.0116377/events.out.tfevents.1671579948.129-146-32-172.157273.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a68872241f0d653734d3778c153868f5c18bbef511a358eea6a5d44701dcc077
|
| 3 |
+
size 5862
|
runs/Dec20_23-45-37_129-146-32-172/events.out.tfevents.1671579948.129-146-32-172.157273.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e510467099222dd9fad92012ce8bbbbaba33b4852b1baf5b06afdb0bdc8ac0b9
|
| 3 |
+
size 6222
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3643
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ac880f769c6c8c222243cd146ac46a2cb7368d0196625844f7aa9d536c159d8
|
| 3 |
size 3643
|
whisper_small_ps_augmented.py
CHANGED
|
@@ -256,7 +256,7 @@ training_args = Seq2SeqTrainingArguments(
|
|
| 256 |
gradient_accumulation_steps=4,
|
| 257 |
learning_rate=1e-5,
|
| 258 |
warmup_steps=30,
|
| 259 |
-
max_steps=
|
| 260 |
gradient_checkpointing=True,
|
| 261 |
fp16=True,
|
| 262 |
evaluation_strategy="steps",
|
|
|
|
| 256 |
gradient_accumulation_steps=4,
|
| 257 |
learning_rate=1e-5,
|
| 258 |
warmup_steps=30,
|
| 259 |
+
max_steps=700,
|
| 260 |
gradient_checkpointing=True,
|
| 261 |
fp16=True,
|
| 262 |
evaluation_strategy="steps",
|