Commit
•
7430914
1
Parent(s):
bac01b2
Training in progress, epoch 1
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- model.safetensors +1 -1
- run-5/checkpoint-120/config.json +80 -0
- run-5/checkpoint-120/model.safetensors +3 -0
- run-5/checkpoint-120/optimizer.pt +3 -0
- run-5/checkpoint-120/preprocessor_config.json +9 -0
- run-5/checkpoint-120/rng_state.pth +3 -0
- run-5/checkpoint-120/scheduler.pt +3 -0
- run-5/checkpoint-120/trainer_state.json +237 -0
- run-5/checkpoint-120/training_args.bin +3 -0
- run-5/checkpoint-144/config.json +80 -0
- run-5/checkpoint-144/model.safetensors +3 -0
- run-5/checkpoint-144/optimizer.pt +3 -0
- run-5/checkpoint-144/preprocessor_config.json +9 -0
- run-5/checkpoint-144/rng_state.pth +3 -0
- run-5/checkpoint-144/scheduler.pt +3 -0
- run-5/checkpoint-144/trainer_state.json +274 -0
- run-5/checkpoint-144/training_args.bin +3 -0
- run-5/checkpoint-168/config.json +80 -0
- run-5/checkpoint-168/model.safetensors +3 -0
- run-5/checkpoint-168/optimizer.pt +3 -0
- run-5/checkpoint-168/preprocessor_config.json +9 -0
- run-5/checkpoint-168/rng_state.pth +3 -0
- run-5/checkpoint-168/scheduler.pt +3 -0
- run-5/checkpoint-168/trainer_state.json +318 -0
- run-5/checkpoint-168/training_args.bin +3 -0
- run-5/checkpoint-48/config.json +80 -0
- run-5/checkpoint-48/model.safetensors +3 -0
- run-5/checkpoint-48/optimizer.pt +3 -0
- run-5/checkpoint-48/preprocessor_config.json +9 -0
- run-5/checkpoint-48/rng_state.pth +3 -0
- run-5/checkpoint-48/scheduler.pt +3 -0
- run-5/checkpoint-48/trainer_state.json +105 -0
- run-5/checkpoint-48/training_args.bin +3 -0
- run-5/checkpoint-72/config.json +80 -0
- run-5/checkpoint-72/model.safetensors +3 -0
- run-5/checkpoint-72/optimizer.pt +3 -0
- run-5/checkpoint-72/preprocessor_config.json +9 -0
- run-5/checkpoint-72/rng_state.pth +3 -0
- run-5/checkpoint-72/scheduler.pt +3 -0
- run-5/checkpoint-72/trainer_state.json +149 -0
- run-5/checkpoint-72/training_args.bin +3 -0
- run-5/checkpoint-96/config.json +80 -0
- run-5/checkpoint-96/model.safetensors +3 -0
- run-5/checkpoint-96/optimizer.pt +3 -0
- run-5/checkpoint-96/preprocessor_config.json +9 -0
- run-5/checkpoint-96/rng_state.pth +3 -0
- run-5/checkpoint-96/scheduler.pt +3 -0
- run-5/checkpoint-96/trainer_state.json +193 -0
- run-5/checkpoint-96/training_args.bin +3 -0
- run-7/checkpoint-16/config.json +80 -0
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 94763496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bcd38943271d452fb41e0cc1c9ac715cd036eaf1fd6859055f6d2b6c6d5d0ccb
|
| 3 |
size 94763496
|
run-5/checkpoint-120/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|
run-5/checkpoint-120/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:454d504da2d40a1ac5002465cf473307c1be7b971d301e4f92ffa50b202f16fc
|
| 3 |
+
size 94763496
|
run-5/checkpoint-120/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f06f2ae8c05c541ef9c93df0a93bccced30f1b4dc10eadbccef60a373d68af39
|
| 3 |
+
size 189552570
|
run-5/checkpoint-120/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
run-5/checkpoint-120/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea9b93979db9d187112d0877ca456edd569c23f080722f2fbbe337c28c1a6935
|
| 3 |
+
size 14244
|
run-5/checkpoint-120/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54d977ec7e793480ff2dd7eef923a400fad294e8389923d1392ea0eb694ad4d2
|
| 3 |
+
size 1064
|
run-5/checkpoint-120/trainer_state.json
ADDED
|
@@ -0,0 +1,237 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7326732673267327,
|
| 3 |
+
"best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-5/checkpoint-24",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 120,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.6560618877410889,
|
| 14 |
+
"learning_rate": 9.928194518007172e-07,
|
| 15 |
+
"loss": 0.6997,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42,
|
| 20 |
+
"grad_norm": 0.6409028768539429,
|
| 21 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 22 |
+
"loss": 0.6971,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"grad_norm": 0.6359347105026245,
|
| 28 |
+
"learning_rate": 2.978458355402151e-06,
|
| 29 |
+
"loss": 0.6953,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.83,
|
| 34 |
+
"grad_norm": 1.56245756149292,
|
| 35 |
+
"learning_rate": 3.971277807202869e-06,
|
| 36 |
+
"loss": 0.6889,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_f1": 0.7326732673267327,
|
| 42 |
+
"eval_loss": 0.6865768432617188,
|
| 43 |
+
"eval_runtime": 1.3652,
|
| 44 |
+
"eval_samples_per_second": 46.879,
|
| 45 |
+
"eval_steps_per_second": 5.86,
|
| 46 |
+
"step": 24
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.04,
|
| 50 |
+
"grad_norm": 0.6976048946380615,
|
| 51 |
+
"learning_rate": 4.743470714158982e-06,
|
| 52 |
+
"loss": 0.6865,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.25,
|
| 57 |
+
"grad_norm": 2.008615732192993,
|
| 58 |
+
"learning_rate": 4.63315744173668e-06,
|
| 59 |
+
"loss": 0.6681,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.46,
|
| 64 |
+
"grad_norm": 1.9885936975479126,
|
| 65 |
+
"learning_rate": 4.522844169314378e-06,
|
| 66 |
+
"loss": 0.6531,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.67,
|
| 71 |
+
"grad_norm": 0.5047381520271301,
|
| 72 |
+
"learning_rate": 4.412530896892076e-06,
|
| 73 |
+
"loss": 0.6664,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"grad_norm": 0.9776627421379089,
|
| 79 |
+
"learning_rate": 4.302217624469774e-06,
|
| 80 |
+
"loss": 0.6545,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_f1": 0.7326732673267327,
|
| 86 |
+
"eval_loss": 0.680206298828125,
|
| 87 |
+
"eval_runtime": 1.3708,
|
| 88 |
+
"eval_samples_per_second": 46.689,
|
| 89 |
+
"eval_steps_per_second": 5.836,
|
| 90 |
+
"step": 48
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.08,
|
| 94 |
+
"grad_norm": 0.46269264817237854,
|
| 95 |
+
"learning_rate": 4.191904352047472e-06,
|
| 96 |
+
"loss": 0.6518,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.29,
|
| 101 |
+
"grad_norm": 0.8560582995414734,
|
| 102 |
+
"learning_rate": 4.081591079625171e-06,
|
| 103 |
+
"loss": 0.6333,
|
| 104 |
+
"step": 55
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.5,
|
| 108 |
+
"grad_norm": 0.9135796427726746,
|
| 109 |
+
"learning_rate": 3.971277807202869e-06,
|
| 110 |
+
"loss": 0.6494,
|
| 111 |
+
"step": 60
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.71,
|
| 115 |
+
"grad_norm": 0.71152263879776,
|
| 116 |
+
"learning_rate": 3.8609645347805665e-06,
|
| 117 |
+
"loss": 0.6444,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.92,
|
| 122 |
+
"grad_norm": 1.3949508666992188,
|
| 123 |
+
"learning_rate": 3.750651262358265e-06,
|
| 124 |
+
"loss": 0.5726,
|
| 125 |
+
"step": 70
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.0,
|
| 129 |
+
"eval_f1": 0.7326732673267327,
|
| 130 |
+
"eval_loss": 0.6824684143066406,
|
| 131 |
+
"eval_runtime": 1.373,
|
| 132 |
+
"eval_samples_per_second": 46.614,
|
| 133 |
+
"eval_steps_per_second": 5.827,
|
| 134 |
+
"step": 72
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.12,
|
| 138 |
+
"grad_norm": 0.7415557503700256,
|
| 139 |
+
"learning_rate": 3.6403379899359627e-06,
|
| 140 |
+
"loss": 0.6271,
|
| 141 |
+
"step": 75
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.33,
|
| 145 |
+
"grad_norm": 0.6959993243217468,
|
| 146 |
+
"learning_rate": 3.5300247175136605e-06,
|
| 147 |
+
"loss": 0.6316,
|
| 148 |
+
"step": 80
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.54,
|
| 152 |
+
"grad_norm": 1.2839587926864624,
|
| 153 |
+
"learning_rate": 3.419711445091359e-06,
|
| 154 |
+
"loss": 0.5607,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.75,
|
| 159 |
+
"grad_norm": 1.2485283613204956,
|
| 160 |
+
"learning_rate": 3.309398172669057e-06,
|
| 161 |
+
"loss": 0.6403,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.96,
|
| 166 |
+
"grad_norm": 1.1759371757507324,
|
| 167 |
+
"learning_rate": 3.199084900246755e-06,
|
| 168 |
+
"loss": 0.6242,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.0,
|
| 173 |
+
"eval_f1": 0.7326732673267327,
|
| 174 |
+
"eval_loss": 0.6892623901367188,
|
| 175 |
+
"eval_runtime": 1.3822,
|
| 176 |
+
"eval_samples_per_second": 46.302,
|
| 177 |
+
"eval_steps_per_second": 5.788,
|
| 178 |
+
"step": 96
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.17,
|
| 182 |
+
"grad_norm": 0.6085842251777649,
|
| 183 |
+
"learning_rate": 3.0887716278244532e-06,
|
| 184 |
+
"loss": 0.564,
|
| 185 |
+
"step": 100
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.38,
|
| 189 |
+
"grad_norm": 0.9830076098442078,
|
| 190 |
+
"learning_rate": 2.978458355402151e-06,
|
| 191 |
+
"loss": 0.641,
|
| 192 |
+
"step": 105
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.58,
|
| 196 |
+
"grad_norm": 0.6557525992393494,
|
| 197 |
+
"learning_rate": 2.8681450829798494e-06,
|
| 198 |
+
"loss": 0.5943,
|
| 199 |
+
"step": 110
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.79,
|
| 203 |
+
"grad_norm": 0.7436282634735107,
|
| 204 |
+
"learning_rate": 2.7578318105575477e-06,
|
| 205 |
+
"loss": 0.614,
|
| 206 |
+
"step": 115
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 5.0,
|
| 210 |
+
"grad_norm": 0.7114256620407104,
|
| 211 |
+
"learning_rate": 2.6475185381352455e-06,
|
| 212 |
+
"loss": 0.6081,
|
| 213 |
+
"step": 120
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 5.0,
|
| 217 |
+
"eval_f1": 0.7326732673267327,
|
| 218 |
+
"eval_loss": 0.6939811706542969,
|
| 219 |
+
"eval_runtime": 1.3699,
|
| 220 |
+
"eval_samples_per_second": 46.719,
|
| 221 |
+
"eval_steps_per_second": 5.84,
|
| 222 |
+
"step": 120
|
| 223 |
+
}
|
| 224 |
+
],
|
| 225 |
+
"logging_steps": 5,
|
| 226 |
+
"max_steps": 240,
|
| 227 |
+
"num_input_tokens_seen": 0,
|
| 228 |
+
"num_train_epochs": 10,
|
| 229 |
+
"save_steps": 500,
|
| 230 |
+
"total_flos": 4193401989215328.0,
|
| 231 |
+
"train_batch_size": 8,
|
| 232 |
+
"trial_name": null,
|
| 233 |
+
"trial_params": {
|
| 234 |
+
"learning_rate": 4.765533368643442e-06,
|
| 235 |
+
"per_device_train_batch_size": 8
|
| 236 |
+
}
|
| 237 |
+
}
|
run-5/checkpoint-120/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064d1923d7ff14a10cb0625f0bc3dd479a67056d28f933cdfbfdecd320e675db
|
| 3 |
+
size 4920
|
run-5/checkpoint-144/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|
run-5/checkpoint-144/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:401a1802884ad357b5301a2252ce0f1ce6c69288db1ddd155757b374073dc7aa
|
| 3 |
+
size 94763496
|
run-5/checkpoint-144/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a13a9395bc8e69da9e62af79e7b6270dc16576123bae7e84c80f3643f4cefe5c
|
| 3 |
+
size 189552570
|
run-5/checkpoint-144/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
run-5/checkpoint-144/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eddddd854f935bdc6bf7e33f19c011df64d161505c6be9aa447288219fa3010a
|
| 3 |
+
size 14244
|
run-5/checkpoint-144/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7632d13f0f8122a5f2aa2d52d94d9212cd5ebcf002004e470d98c2451054c23
|
| 3 |
+
size 1064
|
run-5/checkpoint-144/trainer_state.json
ADDED
|
@@ -0,0 +1,274 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7326732673267327,
|
| 3 |
+
"best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-5/checkpoint-24",
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 144,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.6560618877410889,
|
| 14 |
+
"learning_rate": 9.928194518007172e-07,
|
| 15 |
+
"loss": 0.6997,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42,
|
| 20 |
+
"grad_norm": 0.6409028768539429,
|
| 21 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 22 |
+
"loss": 0.6971,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"grad_norm": 0.6359347105026245,
|
| 28 |
+
"learning_rate": 2.978458355402151e-06,
|
| 29 |
+
"loss": 0.6953,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.83,
|
| 34 |
+
"grad_norm": 1.56245756149292,
|
| 35 |
+
"learning_rate": 3.971277807202869e-06,
|
| 36 |
+
"loss": 0.6889,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_f1": 0.7326732673267327,
|
| 42 |
+
"eval_loss": 0.6865768432617188,
|
| 43 |
+
"eval_runtime": 1.3652,
|
| 44 |
+
"eval_samples_per_second": 46.879,
|
| 45 |
+
"eval_steps_per_second": 5.86,
|
| 46 |
+
"step": 24
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.04,
|
| 50 |
+
"grad_norm": 0.6976048946380615,
|
| 51 |
+
"learning_rate": 4.743470714158982e-06,
|
| 52 |
+
"loss": 0.6865,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.25,
|
| 57 |
+
"grad_norm": 2.008615732192993,
|
| 58 |
+
"learning_rate": 4.63315744173668e-06,
|
| 59 |
+
"loss": 0.6681,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.46,
|
| 64 |
+
"grad_norm": 1.9885936975479126,
|
| 65 |
+
"learning_rate": 4.522844169314378e-06,
|
| 66 |
+
"loss": 0.6531,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.67,
|
| 71 |
+
"grad_norm": 0.5047381520271301,
|
| 72 |
+
"learning_rate": 4.412530896892076e-06,
|
| 73 |
+
"loss": 0.6664,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"grad_norm": 0.9776627421379089,
|
| 79 |
+
"learning_rate": 4.302217624469774e-06,
|
| 80 |
+
"loss": 0.6545,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_f1": 0.7326732673267327,
|
| 86 |
+
"eval_loss": 0.680206298828125,
|
| 87 |
+
"eval_runtime": 1.3708,
|
| 88 |
+
"eval_samples_per_second": 46.689,
|
| 89 |
+
"eval_steps_per_second": 5.836,
|
| 90 |
+
"step": 48
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.08,
|
| 94 |
+
"grad_norm": 0.46269264817237854,
|
| 95 |
+
"learning_rate": 4.191904352047472e-06,
|
| 96 |
+
"loss": 0.6518,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.29,
|
| 101 |
+
"grad_norm": 0.8560582995414734,
|
| 102 |
+
"learning_rate": 4.081591079625171e-06,
|
| 103 |
+
"loss": 0.6333,
|
| 104 |
+
"step": 55
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.5,
|
| 108 |
+
"grad_norm": 0.9135796427726746,
|
| 109 |
+
"learning_rate": 3.971277807202869e-06,
|
| 110 |
+
"loss": 0.6494,
|
| 111 |
+
"step": 60
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.71,
|
| 115 |
+
"grad_norm": 0.71152263879776,
|
| 116 |
+
"learning_rate": 3.8609645347805665e-06,
|
| 117 |
+
"loss": 0.6444,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.92,
|
| 122 |
+
"grad_norm": 1.3949508666992188,
|
| 123 |
+
"learning_rate": 3.750651262358265e-06,
|
| 124 |
+
"loss": 0.5726,
|
| 125 |
+
"step": 70
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.0,
|
| 129 |
+
"eval_f1": 0.7326732673267327,
|
| 130 |
+
"eval_loss": 0.6824684143066406,
|
| 131 |
+
"eval_runtime": 1.373,
|
| 132 |
+
"eval_samples_per_second": 46.614,
|
| 133 |
+
"eval_steps_per_second": 5.827,
|
| 134 |
+
"step": 72
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.12,
|
| 138 |
+
"grad_norm": 0.7415557503700256,
|
| 139 |
+
"learning_rate": 3.6403379899359627e-06,
|
| 140 |
+
"loss": 0.6271,
|
| 141 |
+
"step": 75
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.33,
|
| 145 |
+
"grad_norm": 0.6959993243217468,
|
| 146 |
+
"learning_rate": 3.5300247175136605e-06,
|
| 147 |
+
"loss": 0.6316,
|
| 148 |
+
"step": 80
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.54,
|
| 152 |
+
"grad_norm": 1.2839587926864624,
|
| 153 |
+
"learning_rate": 3.419711445091359e-06,
|
| 154 |
+
"loss": 0.5607,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.75,
|
| 159 |
+
"grad_norm": 1.2485283613204956,
|
| 160 |
+
"learning_rate": 3.309398172669057e-06,
|
| 161 |
+
"loss": 0.6403,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.96,
|
| 166 |
+
"grad_norm": 1.1759371757507324,
|
| 167 |
+
"learning_rate": 3.199084900246755e-06,
|
| 168 |
+
"loss": 0.6242,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.0,
|
| 173 |
+
"eval_f1": 0.7326732673267327,
|
| 174 |
+
"eval_loss": 0.6892623901367188,
|
| 175 |
+
"eval_runtime": 1.3822,
|
| 176 |
+
"eval_samples_per_second": 46.302,
|
| 177 |
+
"eval_steps_per_second": 5.788,
|
| 178 |
+
"step": 96
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.17,
|
| 182 |
+
"grad_norm": 0.6085842251777649,
|
| 183 |
+
"learning_rate": 3.0887716278244532e-06,
|
| 184 |
+
"loss": 0.564,
|
| 185 |
+
"step": 100
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.38,
|
| 189 |
+
"grad_norm": 0.9830076098442078,
|
| 190 |
+
"learning_rate": 2.978458355402151e-06,
|
| 191 |
+
"loss": 0.641,
|
| 192 |
+
"step": 105
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.58,
|
| 196 |
+
"grad_norm": 0.6557525992393494,
|
| 197 |
+
"learning_rate": 2.8681450829798494e-06,
|
| 198 |
+
"loss": 0.5943,
|
| 199 |
+
"step": 110
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.79,
|
| 203 |
+
"grad_norm": 0.7436282634735107,
|
| 204 |
+
"learning_rate": 2.7578318105575477e-06,
|
| 205 |
+
"loss": 0.614,
|
| 206 |
+
"step": 115
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 5.0,
|
| 210 |
+
"grad_norm": 0.7114256620407104,
|
| 211 |
+
"learning_rate": 2.6475185381352455e-06,
|
| 212 |
+
"loss": 0.6081,
|
| 213 |
+
"step": 120
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 5.0,
|
| 217 |
+
"eval_f1": 0.7326732673267327,
|
| 218 |
+
"eval_loss": 0.6939811706542969,
|
| 219 |
+
"eval_runtime": 1.3699,
|
| 220 |
+
"eval_samples_per_second": 46.719,
|
| 221 |
+
"eval_steps_per_second": 5.84,
|
| 222 |
+
"step": 120
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.21,
|
| 226 |
+
"grad_norm": 0.6939758062362671,
|
| 227 |
+
"learning_rate": 2.537205265712944e-06,
|
| 228 |
+
"loss": 0.5683,
|
| 229 |
+
"step": 125
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 5.42,
|
| 233 |
+
"grad_norm": 0.7150554656982422,
|
| 234 |
+
"learning_rate": 2.426891993290642e-06,
|
| 235 |
+
"loss": 0.6389,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.62,
|
| 240 |
+
"grad_norm": 0.5391159057617188,
|
| 241 |
+
"learning_rate": 2.31657872086834e-06,
|
| 242 |
+
"loss": 0.6185,
|
| 243 |
+
"step": 135
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 5.83,
|
| 247 |
+
"grad_norm": 0.657724142074585,
|
| 248 |
+
"learning_rate": 2.206265448446038e-06,
|
| 249 |
+
"loss": 0.6216,
|
| 250 |
+
"step": 140
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 6.0,
|
| 254 |
+
"eval_f1": 0.7326732673267327,
|
| 255 |
+
"eval_loss": 0.69482421875,
|
| 256 |
+
"eval_runtime": 1.3614,
|
| 257 |
+
"eval_samples_per_second": 47.009,
|
| 258 |
+
"eval_steps_per_second": 5.876,
|
| 259 |
+
"step": 144
|
| 260 |
+
}
|
| 261 |
+
],
|
| 262 |
+
"logging_steps": 5,
|
| 263 |
+
"max_steps": 240,
|
| 264 |
+
"num_input_tokens_seen": 0,
|
| 265 |
+
"num_train_epochs": 10,
|
| 266 |
+
"save_steps": 500,
|
| 267 |
+
"total_flos": 4911381340990080.0,
|
| 268 |
+
"train_batch_size": 8,
|
| 269 |
+
"trial_name": null,
|
| 270 |
+
"trial_params": {
|
| 271 |
+
"learning_rate": 4.765533368643442e-06,
|
| 272 |
+
"per_device_train_batch_size": 8
|
| 273 |
+
}
|
| 274 |
+
}
|
run-5/checkpoint-144/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064d1923d7ff14a10cb0625f0bc3dd479a67056d28f933cdfbfdecd320e675db
|
| 3 |
+
size 4920
|
run-5/checkpoint-168/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|
run-5/checkpoint-168/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:efc7bd641365c6786592c3515017b5ed2ae6a8819f32b77ec681108ed8e76dcb
|
| 3 |
+
size 94763496
|
run-5/checkpoint-168/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:815babcbbaa2ea7cdb1bc66bb94cf7e3cdb8ac6fd8db97a412629826b8eed8fc
|
| 3 |
+
size 189552570
|
run-5/checkpoint-168/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
run-5/checkpoint-168/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3d1e8300999dd1e32b96ea9da5218bf1661f524b97023fae45e0a2d78d5309f
|
| 3 |
+
size 14244
|
run-5/checkpoint-168/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f0c5b02573b46c9f41bc7a751eec9891494db2257743c0802d91b804b452d9a
|
| 3 |
+
size 1064
|
run-5/checkpoint-168/trainer_state.json
ADDED
|
@@ -0,0 +1,318 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7326732673267327,
|
| 3 |
+
"best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-5/checkpoint-24",
|
| 4 |
+
"epoch": 7.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 168,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.6560618877410889,
|
| 14 |
+
"learning_rate": 9.928194518007172e-07,
|
| 15 |
+
"loss": 0.6997,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42,
|
| 20 |
+
"grad_norm": 0.6409028768539429,
|
| 21 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 22 |
+
"loss": 0.6971,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"grad_norm": 0.6359347105026245,
|
| 28 |
+
"learning_rate": 2.978458355402151e-06,
|
| 29 |
+
"loss": 0.6953,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.83,
|
| 34 |
+
"grad_norm": 1.56245756149292,
|
| 35 |
+
"learning_rate": 3.971277807202869e-06,
|
| 36 |
+
"loss": 0.6889,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_f1": 0.7326732673267327,
|
| 42 |
+
"eval_loss": 0.6865768432617188,
|
| 43 |
+
"eval_runtime": 1.3652,
|
| 44 |
+
"eval_samples_per_second": 46.879,
|
| 45 |
+
"eval_steps_per_second": 5.86,
|
| 46 |
+
"step": 24
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.04,
|
| 50 |
+
"grad_norm": 0.6976048946380615,
|
| 51 |
+
"learning_rate": 4.743470714158982e-06,
|
| 52 |
+
"loss": 0.6865,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.25,
|
| 57 |
+
"grad_norm": 2.008615732192993,
|
| 58 |
+
"learning_rate": 4.63315744173668e-06,
|
| 59 |
+
"loss": 0.6681,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.46,
|
| 64 |
+
"grad_norm": 1.9885936975479126,
|
| 65 |
+
"learning_rate": 4.522844169314378e-06,
|
| 66 |
+
"loss": 0.6531,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.67,
|
| 71 |
+
"grad_norm": 0.5047381520271301,
|
| 72 |
+
"learning_rate": 4.412530896892076e-06,
|
| 73 |
+
"loss": 0.6664,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"grad_norm": 0.9776627421379089,
|
| 79 |
+
"learning_rate": 4.302217624469774e-06,
|
| 80 |
+
"loss": 0.6545,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_f1": 0.7326732673267327,
|
| 86 |
+
"eval_loss": 0.680206298828125,
|
| 87 |
+
"eval_runtime": 1.3708,
|
| 88 |
+
"eval_samples_per_second": 46.689,
|
| 89 |
+
"eval_steps_per_second": 5.836,
|
| 90 |
+
"step": 48
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.08,
|
| 94 |
+
"grad_norm": 0.46269264817237854,
|
| 95 |
+
"learning_rate": 4.191904352047472e-06,
|
| 96 |
+
"loss": 0.6518,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.29,
|
| 101 |
+
"grad_norm": 0.8560582995414734,
|
| 102 |
+
"learning_rate": 4.081591079625171e-06,
|
| 103 |
+
"loss": 0.6333,
|
| 104 |
+
"step": 55
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.5,
|
| 108 |
+
"grad_norm": 0.9135796427726746,
|
| 109 |
+
"learning_rate": 3.971277807202869e-06,
|
| 110 |
+
"loss": 0.6494,
|
| 111 |
+
"step": 60
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.71,
|
| 115 |
+
"grad_norm": 0.71152263879776,
|
| 116 |
+
"learning_rate": 3.8609645347805665e-06,
|
| 117 |
+
"loss": 0.6444,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.92,
|
| 122 |
+
"grad_norm": 1.3949508666992188,
|
| 123 |
+
"learning_rate": 3.750651262358265e-06,
|
| 124 |
+
"loss": 0.5726,
|
| 125 |
+
"step": 70
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.0,
|
| 129 |
+
"eval_f1": 0.7326732673267327,
|
| 130 |
+
"eval_loss": 0.6824684143066406,
|
| 131 |
+
"eval_runtime": 1.373,
|
| 132 |
+
"eval_samples_per_second": 46.614,
|
| 133 |
+
"eval_steps_per_second": 5.827,
|
| 134 |
+
"step": 72
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.12,
|
| 138 |
+
"grad_norm": 0.7415557503700256,
|
| 139 |
+
"learning_rate": 3.6403379899359627e-06,
|
| 140 |
+
"loss": 0.6271,
|
| 141 |
+
"step": 75
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.33,
|
| 145 |
+
"grad_norm": 0.6959993243217468,
|
| 146 |
+
"learning_rate": 3.5300247175136605e-06,
|
| 147 |
+
"loss": 0.6316,
|
| 148 |
+
"step": 80
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.54,
|
| 152 |
+
"grad_norm": 1.2839587926864624,
|
| 153 |
+
"learning_rate": 3.419711445091359e-06,
|
| 154 |
+
"loss": 0.5607,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.75,
|
| 159 |
+
"grad_norm": 1.2485283613204956,
|
| 160 |
+
"learning_rate": 3.309398172669057e-06,
|
| 161 |
+
"loss": 0.6403,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.96,
|
| 166 |
+
"grad_norm": 1.1759371757507324,
|
| 167 |
+
"learning_rate": 3.199084900246755e-06,
|
| 168 |
+
"loss": 0.6242,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.0,
|
| 173 |
+
"eval_f1": 0.7326732673267327,
|
| 174 |
+
"eval_loss": 0.6892623901367188,
|
| 175 |
+
"eval_runtime": 1.3822,
|
| 176 |
+
"eval_samples_per_second": 46.302,
|
| 177 |
+
"eval_steps_per_second": 5.788,
|
| 178 |
+
"step": 96
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.17,
|
| 182 |
+
"grad_norm": 0.6085842251777649,
|
| 183 |
+
"learning_rate": 3.0887716278244532e-06,
|
| 184 |
+
"loss": 0.564,
|
| 185 |
+
"step": 100
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.38,
|
| 189 |
+
"grad_norm": 0.9830076098442078,
|
| 190 |
+
"learning_rate": 2.978458355402151e-06,
|
| 191 |
+
"loss": 0.641,
|
| 192 |
+
"step": 105
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.58,
|
| 196 |
+
"grad_norm": 0.6557525992393494,
|
| 197 |
+
"learning_rate": 2.8681450829798494e-06,
|
| 198 |
+
"loss": 0.5943,
|
| 199 |
+
"step": 110
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.79,
|
| 203 |
+
"grad_norm": 0.7436282634735107,
|
| 204 |
+
"learning_rate": 2.7578318105575477e-06,
|
| 205 |
+
"loss": 0.614,
|
| 206 |
+
"step": 115
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 5.0,
|
| 210 |
+
"grad_norm": 0.7114256620407104,
|
| 211 |
+
"learning_rate": 2.6475185381352455e-06,
|
| 212 |
+
"loss": 0.6081,
|
| 213 |
+
"step": 120
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 5.0,
|
| 217 |
+
"eval_f1": 0.7326732673267327,
|
| 218 |
+
"eval_loss": 0.6939811706542969,
|
| 219 |
+
"eval_runtime": 1.3699,
|
| 220 |
+
"eval_samples_per_second": 46.719,
|
| 221 |
+
"eval_steps_per_second": 5.84,
|
| 222 |
+
"step": 120
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.21,
|
| 226 |
+
"grad_norm": 0.6939758062362671,
|
| 227 |
+
"learning_rate": 2.537205265712944e-06,
|
| 228 |
+
"loss": 0.5683,
|
| 229 |
+
"step": 125
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 5.42,
|
| 233 |
+
"grad_norm": 0.7150554656982422,
|
| 234 |
+
"learning_rate": 2.426891993290642e-06,
|
| 235 |
+
"loss": 0.6389,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.62,
|
| 240 |
+
"grad_norm": 0.5391159057617188,
|
| 241 |
+
"learning_rate": 2.31657872086834e-06,
|
| 242 |
+
"loss": 0.6185,
|
| 243 |
+
"step": 135
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 5.83,
|
| 247 |
+
"grad_norm": 0.657724142074585,
|
| 248 |
+
"learning_rate": 2.206265448446038e-06,
|
| 249 |
+
"loss": 0.6216,
|
| 250 |
+
"step": 140
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 6.0,
|
| 254 |
+
"eval_f1": 0.7326732673267327,
|
| 255 |
+
"eval_loss": 0.69482421875,
|
| 256 |
+
"eval_runtime": 1.3614,
|
| 257 |
+
"eval_samples_per_second": 47.009,
|
| 258 |
+
"eval_steps_per_second": 5.876,
|
| 259 |
+
"step": 144
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.04,
|
| 263 |
+
"grad_norm": 1.1818499565124512,
|
| 264 |
+
"learning_rate": 2.095952176023736e-06,
|
| 265 |
+
"loss": 0.5331,
|
| 266 |
+
"step": 145
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 6.25,
|
| 270 |
+
"grad_norm": 0.6219866871833801,
|
| 271 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 272 |
+
"loss": 0.5781,
|
| 273 |
+
"step": 150
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 6.46,
|
| 277 |
+
"grad_norm": 0.954688310623169,
|
| 278 |
+
"learning_rate": 1.8753256311791324e-06,
|
| 279 |
+
"loss": 0.6719,
|
| 280 |
+
"step": 155
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 6.67,
|
| 284 |
+
"grad_norm": 0.5371306538581848,
|
| 285 |
+
"learning_rate": 1.7650123587568303e-06,
|
| 286 |
+
"loss": 0.6174,
|
| 287 |
+
"step": 160
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 6.88,
|
| 291 |
+
"grad_norm": 1.098739743232727,
|
| 292 |
+
"learning_rate": 1.6546990863345285e-06,
|
| 293 |
+
"loss": 0.6005,
|
| 294 |
+
"step": 165
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 7.0,
|
| 298 |
+
"eval_f1": 0.7326732673267327,
|
| 299 |
+
"eval_loss": 0.6969451904296875,
|
| 300 |
+
"eval_runtime": 1.371,
|
| 301 |
+
"eval_samples_per_second": 46.681,
|
| 302 |
+
"eval_steps_per_second": 5.835,
|
| 303 |
+
"step": 168
|
| 304 |
+
}
|
| 305 |
+
],
|
| 306 |
+
"logging_steps": 5,
|
| 307 |
+
"max_steps": 240,
|
| 308 |
+
"num_input_tokens_seen": 0,
|
| 309 |
+
"num_train_epochs": 10,
|
| 310 |
+
"save_steps": 500,
|
| 311 |
+
"total_flos": 5784921236870880.0,
|
| 312 |
+
"train_batch_size": 8,
|
| 313 |
+
"trial_name": null,
|
| 314 |
+
"trial_params": {
|
| 315 |
+
"learning_rate": 4.765533368643442e-06,
|
| 316 |
+
"per_device_train_batch_size": 8
|
| 317 |
+
}
|
| 318 |
+
}
|
run-5/checkpoint-168/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064d1923d7ff14a10cb0625f0bc3dd479a67056d28f933cdfbfdecd320e675db
|
| 3 |
+
size 4920
|
run-5/checkpoint-48/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|
run-5/checkpoint-48/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44374b58af3c51b5eac7f464c67b1a6c5e2197656be426f7a0caf6bd7ebd43eb
|
| 3 |
+
size 94763496
|
run-5/checkpoint-48/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:560210c0f70f7f9718c0ae200103724f2f497040cd6f57db30cdeada4fbe508e
|
| 3 |
+
size 189552570
|
run-5/checkpoint-48/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
run-5/checkpoint-48/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ea0f2253490889e9b5dbea96976236c3c1f025b5a24179f4bcaa44eec621be7
|
| 3 |
+
size 14244
|
run-5/checkpoint-48/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49e1e8bf8425998d14aca500c6e7194f5af9e802d88054b4e269eeac25555423
|
| 3 |
+
size 1064
|
run-5/checkpoint-48/trainer_state.json
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7326732673267327,
|
| 3 |
+
"best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-5/checkpoint-24",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 48,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.6560618877410889,
|
| 14 |
+
"learning_rate": 9.928194518007172e-07,
|
| 15 |
+
"loss": 0.6997,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42,
|
| 20 |
+
"grad_norm": 0.6409028768539429,
|
| 21 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 22 |
+
"loss": 0.6971,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"grad_norm": 0.6359347105026245,
|
| 28 |
+
"learning_rate": 2.978458355402151e-06,
|
| 29 |
+
"loss": 0.6953,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.83,
|
| 34 |
+
"grad_norm": 1.56245756149292,
|
| 35 |
+
"learning_rate": 3.971277807202869e-06,
|
| 36 |
+
"loss": 0.6889,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_f1": 0.7326732673267327,
|
| 42 |
+
"eval_loss": 0.6865768432617188,
|
| 43 |
+
"eval_runtime": 1.3652,
|
| 44 |
+
"eval_samples_per_second": 46.879,
|
| 45 |
+
"eval_steps_per_second": 5.86,
|
| 46 |
+
"step": 24
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.04,
|
| 50 |
+
"grad_norm": 0.6976048946380615,
|
| 51 |
+
"learning_rate": 4.743470714158982e-06,
|
| 52 |
+
"loss": 0.6865,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.25,
|
| 57 |
+
"grad_norm": 2.008615732192993,
|
| 58 |
+
"learning_rate": 4.63315744173668e-06,
|
| 59 |
+
"loss": 0.6681,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.46,
|
| 64 |
+
"grad_norm": 1.9885936975479126,
|
| 65 |
+
"learning_rate": 4.522844169314378e-06,
|
| 66 |
+
"loss": 0.6531,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.67,
|
| 71 |
+
"grad_norm": 0.5047381520271301,
|
| 72 |
+
"learning_rate": 4.412530896892076e-06,
|
| 73 |
+
"loss": 0.6664,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"grad_norm": 0.9776627421379089,
|
| 79 |
+
"learning_rate": 4.302217624469774e-06,
|
| 80 |
+
"loss": 0.6545,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_f1": 0.7326732673267327,
|
| 86 |
+
"eval_loss": 0.680206298828125,
|
| 87 |
+
"eval_runtime": 1.3708,
|
| 88 |
+
"eval_samples_per_second": 46.689,
|
| 89 |
+
"eval_steps_per_second": 5.836,
|
| 90 |
+
"step": 48
|
| 91 |
+
}
|
| 92 |
+
],
|
| 93 |
+
"logging_steps": 5,
|
| 94 |
+
"max_steps": 240,
|
| 95 |
+
"num_input_tokens_seen": 0,
|
| 96 |
+
"num_train_epochs": 10,
|
| 97 |
+
"save_steps": 500,
|
| 98 |
+
"total_flos": 1570339835242944.0,
|
| 99 |
+
"train_batch_size": 8,
|
| 100 |
+
"trial_name": null,
|
| 101 |
+
"trial_params": {
|
| 102 |
+
"learning_rate": 4.765533368643442e-06,
|
| 103 |
+
"per_device_train_batch_size": 8
|
| 104 |
+
}
|
| 105 |
+
}
|
run-5/checkpoint-48/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064d1923d7ff14a10cb0625f0bc3dd479a67056d28f933cdfbfdecd320e675db
|
| 3 |
+
size 4920
|
run-5/checkpoint-72/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|
run-5/checkpoint-72/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a00e0256d1ea1849afaec15d21adc0869da29e07ca765d0d9625d8cef6c46ae9
|
| 3 |
+
size 94763496
|
run-5/checkpoint-72/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:054e0c4c58449144ae1ba77189cbb26a582773d33a97bd494ab34c74ae905776
|
| 3 |
+
size 189552570
|
run-5/checkpoint-72/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
run-5/checkpoint-72/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ee3b4c1fd10f5b7dc9cd3892663bc52a1bc7290b86a11225b56750a1f3c0adf
|
| 3 |
+
size 14244
|
run-5/checkpoint-72/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8bfb2ec2dd0d59fe1c6c48a5eb057fb5f63d5cde1db459bdc9745595437e122
|
| 3 |
+
size 1064
|
run-5/checkpoint-72/trainer_state.json
ADDED
|
@@ -0,0 +1,149 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7326732673267327,
|
| 3 |
+
"best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-5/checkpoint-24",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 72,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.6560618877410889,
|
| 14 |
+
"learning_rate": 9.928194518007172e-07,
|
| 15 |
+
"loss": 0.6997,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42,
|
| 20 |
+
"grad_norm": 0.6409028768539429,
|
| 21 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 22 |
+
"loss": 0.6971,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"grad_norm": 0.6359347105026245,
|
| 28 |
+
"learning_rate": 2.978458355402151e-06,
|
| 29 |
+
"loss": 0.6953,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.83,
|
| 34 |
+
"grad_norm": 1.56245756149292,
|
| 35 |
+
"learning_rate": 3.971277807202869e-06,
|
| 36 |
+
"loss": 0.6889,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_f1": 0.7326732673267327,
|
| 42 |
+
"eval_loss": 0.6865768432617188,
|
| 43 |
+
"eval_runtime": 1.3652,
|
| 44 |
+
"eval_samples_per_second": 46.879,
|
| 45 |
+
"eval_steps_per_second": 5.86,
|
| 46 |
+
"step": 24
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.04,
|
| 50 |
+
"grad_norm": 0.6976048946380615,
|
| 51 |
+
"learning_rate": 4.743470714158982e-06,
|
| 52 |
+
"loss": 0.6865,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.25,
|
| 57 |
+
"grad_norm": 2.008615732192993,
|
| 58 |
+
"learning_rate": 4.63315744173668e-06,
|
| 59 |
+
"loss": 0.6681,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.46,
|
| 64 |
+
"grad_norm": 1.9885936975479126,
|
| 65 |
+
"learning_rate": 4.522844169314378e-06,
|
| 66 |
+
"loss": 0.6531,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.67,
|
| 71 |
+
"grad_norm": 0.5047381520271301,
|
| 72 |
+
"learning_rate": 4.412530896892076e-06,
|
| 73 |
+
"loss": 0.6664,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"grad_norm": 0.9776627421379089,
|
| 79 |
+
"learning_rate": 4.302217624469774e-06,
|
| 80 |
+
"loss": 0.6545,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_f1": 0.7326732673267327,
|
| 86 |
+
"eval_loss": 0.680206298828125,
|
| 87 |
+
"eval_runtime": 1.3708,
|
| 88 |
+
"eval_samples_per_second": 46.689,
|
| 89 |
+
"eval_steps_per_second": 5.836,
|
| 90 |
+
"step": 48
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.08,
|
| 94 |
+
"grad_norm": 0.46269264817237854,
|
| 95 |
+
"learning_rate": 4.191904352047472e-06,
|
| 96 |
+
"loss": 0.6518,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.29,
|
| 101 |
+
"grad_norm": 0.8560582995414734,
|
| 102 |
+
"learning_rate": 4.081591079625171e-06,
|
| 103 |
+
"loss": 0.6333,
|
| 104 |
+
"step": 55
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.5,
|
| 108 |
+
"grad_norm": 0.9135796427726746,
|
| 109 |
+
"learning_rate": 3.971277807202869e-06,
|
| 110 |
+
"loss": 0.6494,
|
| 111 |
+
"step": 60
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.71,
|
| 115 |
+
"grad_norm": 0.71152263879776,
|
| 116 |
+
"learning_rate": 3.8609645347805665e-06,
|
| 117 |
+
"loss": 0.6444,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.92,
|
| 122 |
+
"grad_norm": 1.3949508666992188,
|
| 123 |
+
"learning_rate": 3.750651262358265e-06,
|
| 124 |
+
"loss": 0.5726,
|
| 125 |
+
"step": 70
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.0,
|
| 129 |
+
"eval_f1": 0.7326732673267327,
|
| 130 |
+
"eval_loss": 0.6824684143066406,
|
| 131 |
+
"eval_runtime": 1.373,
|
| 132 |
+
"eval_samples_per_second": 46.614,
|
| 133 |
+
"eval_steps_per_second": 5.827,
|
| 134 |
+
"step": 72
|
| 135 |
+
}
|
| 136 |
+
],
|
| 137 |
+
"logging_steps": 5,
|
| 138 |
+
"max_steps": 240,
|
| 139 |
+
"num_input_tokens_seen": 0,
|
| 140 |
+
"num_train_epochs": 10,
|
| 141 |
+
"save_steps": 500,
|
| 142 |
+
"total_flos": 2449725503657472.0,
|
| 143 |
+
"train_batch_size": 8,
|
| 144 |
+
"trial_name": null,
|
| 145 |
+
"trial_params": {
|
| 146 |
+
"learning_rate": 4.765533368643442e-06,
|
| 147 |
+
"per_device_train_batch_size": 8
|
| 148 |
+
}
|
| 149 |
+
}
|
run-5/checkpoint-72/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064d1923d7ff14a10cb0625f0bc3dd479a67056d28f933cdfbfdecd320e675db
|
| 3 |
+
size 4920
|
run-5/checkpoint-96/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|
run-5/checkpoint-96/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d80eb03f980b18c53130ffe20612f24f0ad6b932416946a4a72ebe164dedf2d2
|
| 3 |
+
size 94763496
|
run-5/checkpoint-96/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0343c275626ac316997263b3116cf0e218509562bd2f7af00c7a09411a332e2e
|
| 3 |
+
size 189552570
|
run-5/checkpoint-96/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
run-5/checkpoint-96/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b456b0645d03f4fc8a479113b5caa8bb02734d9c9788d3ded452d4ca10da7ef
|
| 3 |
+
size 14244
|
run-5/checkpoint-96/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e35dfd154520d23e4be85bc337f841c5a10afc0834c04f9c8d53994824829a57
|
| 3 |
+
size 1064
|
run-5/checkpoint-96/trainer_state.json
ADDED
|
@@ -0,0 +1,193 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7326732673267327,
|
| 3 |
+
"best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-5/checkpoint-24",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 96,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21,
|
| 13 |
+
"grad_norm": 1.6560618877410889,
|
| 14 |
+
"learning_rate": 9.928194518007172e-07,
|
| 15 |
+
"loss": 0.6997,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.42,
|
| 20 |
+
"grad_norm": 0.6409028768539429,
|
| 21 |
+
"learning_rate": 1.9856389036014343e-06,
|
| 22 |
+
"loss": 0.6971,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.62,
|
| 27 |
+
"grad_norm": 0.6359347105026245,
|
| 28 |
+
"learning_rate": 2.978458355402151e-06,
|
| 29 |
+
"loss": 0.6953,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.83,
|
| 34 |
+
"grad_norm": 1.56245756149292,
|
| 35 |
+
"learning_rate": 3.971277807202869e-06,
|
| 36 |
+
"loss": 0.6889,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_f1": 0.7326732673267327,
|
| 42 |
+
"eval_loss": 0.6865768432617188,
|
| 43 |
+
"eval_runtime": 1.3652,
|
| 44 |
+
"eval_samples_per_second": 46.879,
|
| 45 |
+
"eval_steps_per_second": 5.86,
|
| 46 |
+
"step": 24
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.04,
|
| 50 |
+
"grad_norm": 0.6976048946380615,
|
| 51 |
+
"learning_rate": 4.743470714158982e-06,
|
| 52 |
+
"loss": 0.6865,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.25,
|
| 57 |
+
"grad_norm": 2.008615732192993,
|
| 58 |
+
"learning_rate": 4.63315744173668e-06,
|
| 59 |
+
"loss": 0.6681,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.46,
|
| 64 |
+
"grad_norm": 1.9885936975479126,
|
| 65 |
+
"learning_rate": 4.522844169314378e-06,
|
| 66 |
+
"loss": 0.6531,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.67,
|
| 71 |
+
"grad_norm": 0.5047381520271301,
|
| 72 |
+
"learning_rate": 4.412530896892076e-06,
|
| 73 |
+
"loss": 0.6664,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"grad_norm": 0.9776627421379089,
|
| 79 |
+
"learning_rate": 4.302217624469774e-06,
|
| 80 |
+
"loss": 0.6545,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_f1": 0.7326732673267327,
|
| 86 |
+
"eval_loss": 0.680206298828125,
|
| 87 |
+
"eval_runtime": 1.3708,
|
| 88 |
+
"eval_samples_per_second": 46.689,
|
| 89 |
+
"eval_steps_per_second": 5.836,
|
| 90 |
+
"step": 48
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.08,
|
| 94 |
+
"grad_norm": 0.46269264817237854,
|
| 95 |
+
"learning_rate": 4.191904352047472e-06,
|
| 96 |
+
"loss": 0.6518,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.29,
|
| 101 |
+
"grad_norm": 0.8560582995414734,
|
| 102 |
+
"learning_rate": 4.081591079625171e-06,
|
| 103 |
+
"loss": 0.6333,
|
| 104 |
+
"step": 55
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.5,
|
| 108 |
+
"grad_norm": 0.9135796427726746,
|
| 109 |
+
"learning_rate": 3.971277807202869e-06,
|
| 110 |
+
"loss": 0.6494,
|
| 111 |
+
"step": 60
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.71,
|
| 115 |
+
"grad_norm": 0.71152263879776,
|
| 116 |
+
"learning_rate": 3.8609645347805665e-06,
|
| 117 |
+
"loss": 0.6444,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.92,
|
| 122 |
+
"grad_norm": 1.3949508666992188,
|
| 123 |
+
"learning_rate": 3.750651262358265e-06,
|
| 124 |
+
"loss": 0.5726,
|
| 125 |
+
"step": 70
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.0,
|
| 129 |
+
"eval_f1": 0.7326732673267327,
|
| 130 |
+
"eval_loss": 0.6824684143066406,
|
| 131 |
+
"eval_runtime": 1.373,
|
| 132 |
+
"eval_samples_per_second": 46.614,
|
| 133 |
+
"eval_steps_per_second": 5.827,
|
| 134 |
+
"step": 72
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.12,
|
| 138 |
+
"grad_norm": 0.7415557503700256,
|
| 139 |
+
"learning_rate": 3.6403379899359627e-06,
|
| 140 |
+
"loss": 0.6271,
|
| 141 |
+
"step": 75
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.33,
|
| 145 |
+
"grad_norm": 0.6959993243217468,
|
| 146 |
+
"learning_rate": 3.5300247175136605e-06,
|
| 147 |
+
"loss": 0.6316,
|
| 148 |
+
"step": 80
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.54,
|
| 152 |
+
"grad_norm": 1.2839587926864624,
|
| 153 |
+
"learning_rate": 3.419711445091359e-06,
|
| 154 |
+
"loss": 0.5607,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.75,
|
| 159 |
+
"grad_norm": 1.2485283613204956,
|
| 160 |
+
"learning_rate": 3.309398172669057e-06,
|
| 161 |
+
"loss": 0.6403,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.96,
|
| 166 |
+
"grad_norm": 1.1759371757507324,
|
| 167 |
+
"learning_rate": 3.199084900246755e-06,
|
| 168 |
+
"loss": 0.6242,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.0,
|
| 173 |
+
"eval_f1": 0.7326732673267327,
|
| 174 |
+
"eval_loss": 0.6892623901367188,
|
| 175 |
+
"eval_runtime": 1.3822,
|
| 176 |
+
"eval_samples_per_second": 46.302,
|
| 177 |
+
"eval_steps_per_second": 5.788,
|
| 178 |
+
"step": 96
|
| 179 |
+
}
|
| 180 |
+
],
|
| 181 |
+
"logging_steps": 5,
|
| 182 |
+
"max_steps": 240,
|
| 183 |
+
"num_input_tokens_seen": 0,
|
| 184 |
+
"num_train_epochs": 10,
|
| 185 |
+
"save_steps": 500,
|
| 186 |
+
"total_flos": 3314382463332576.0,
|
| 187 |
+
"train_batch_size": 8,
|
| 188 |
+
"trial_name": null,
|
| 189 |
+
"trial_params": {
|
| 190 |
+
"learning_rate": 4.765533368643442e-06,
|
| 191 |
+
"per_device_train_batch_size": 8
|
| 192 |
+
}
|
| 193 |
+
}
|
run-5/checkpoint-96/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:064d1923d7ff14a10cb0625f0bc3dd479a67056d28f933cdfbfdecd320e675db
|
| 3 |
+
size 4920
|
run-7/checkpoint-16/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ntu-spml/distilhubert",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": false,
|
| 47 |
+
"final_dropout": 0.0,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"id2label": {
|
| 52 |
+
"0": "NOT_WORD",
|
| 53 |
+
"1": "WORD"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"NOT_WORD": "0",
|
| 59 |
+
"WORD": "1"
|
| 60 |
+
},
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "hubert",
|
| 70 |
+
"num_attention_heads": 12,
|
| 71 |
+
"num_conv_pos_embedding_groups": 16,
|
| 72 |
+
"num_conv_pos_embeddings": 128,
|
| 73 |
+
"num_feat_extract_layers": 7,
|
| 74 |
+
"num_hidden_layers": 2,
|
| 75 |
+
"pad_token_id": 0,
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.38.1",
|
| 78 |
+
"use_weighted_layer_sum": false,
|
| 79 |
+
"vocab_size": 32
|
| 80 |
+
}
|