Upload 10 files
Browse files- config.json +116 -0
- preprocessor_config.json +9 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +739 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "C:\\workspace\\checkpoints\\",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"adapter_kernel_size": 3,
|
5 |
+
"adapter_stride": 2,
|
6 |
+
"add_adapter": false,
|
7 |
+
"apply_spec_augment": true,
|
8 |
+
"architectures": [
|
9 |
+
"Wav2Vec2ForCTC"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.0,
|
12 |
+
"bos_token_id": 1,
|
13 |
+
"classifier_proj_size": 256,
|
14 |
+
"codevector_dim": 768,
|
15 |
+
"contrastive_logits_temperature": 0.1,
|
16 |
+
"conv_bias": true,
|
17 |
+
"conv_dim": [
|
18 |
+
512,
|
19 |
+
512,
|
20 |
+
512,
|
21 |
+
512,
|
22 |
+
512,
|
23 |
+
512,
|
24 |
+
512
|
25 |
+
],
|
26 |
+
"conv_kernel": [
|
27 |
+
10,
|
28 |
+
3,
|
29 |
+
3,
|
30 |
+
3,
|
31 |
+
3,
|
32 |
+
2,
|
33 |
+
2
|
34 |
+
],
|
35 |
+
"conv_stride": [
|
36 |
+
5,
|
37 |
+
2,
|
38 |
+
2,
|
39 |
+
2,
|
40 |
+
2,
|
41 |
+
2,
|
42 |
+
2
|
43 |
+
],
|
44 |
+
"ctc_loss_reduction": "mean",
|
45 |
+
"ctc_zero_infinity": false,
|
46 |
+
"diversity_loss_weight": 0.1,
|
47 |
+
"do_stable_layer_norm": true,
|
48 |
+
"eos_token_id": 2,
|
49 |
+
"feat_extract_activation": "gelu",
|
50 |
+
"feat_extract_dropout": 0.0,
|
51 |
+
"feat_extract_norm": "layer",
|
52 |
+
"feat_proj_dropout": 0,
|
53 |
+
"feat_quantizer_dropout": 0.0,
|
54 |
+
"final_dropout": 0.0,
|
55 |
+
"gradient_checkpointing": false,
|
56 |
+
"hidden_act": "gelu",
|
57 |
+
"hidden_dropout": 0.0,
|
58 |
+
"hidden_size": 1024,
|
59 |
+
"initializer_range": 0.02,
|
60 |
+
"intermediate_size": 4096,
|
61 |
+
"layer_norm_eps": 1e-05,
|
62 |
+
"layerdrop": 0.0,
|
63 |
+
"mask_channel_length": 10,
|
64 |
+
"mask_channel_min_space": 1,
|
65 |
+
"mask_channel_other": 0.0,
|
66 |
+
"mask_channel_prob": 0.0,
|
67 |
+
"mask_channel_selection": "static",
|
68 |
+
"mask_feature_length": 10,
|
69 |
+
"mask_feature_min_masks": 0,
|
70 |
+
"mask_feature_prob": 0.0,
|
71 |
+
"mask_time_length": 10,
|
72 |
+
"mask_time_min_masks": 2,
|
73 |
+
"mask_time_min_space": 1,
|
74 |
+
"mask_time_other": 0.0,
|
75 |
+
"mask_time_prob": 0.0,
|
76 |
+
"mask_time_selection": "static",
|
77 |
+
"model_type": "wav2vec2",
|
78 |
+
"num_adapter_layers": 3,
|
79 |
+
"num_attention_heads": 16,
|
80 |
+
"num_codevector_groups": 2,
|
81 |
+
"num_codevectors_per_group": 320,
|
82 |
+
"num_conv_pos_embedding_groups": 16,
|
83 |
+
"num_conv_pos_embeddings": 128,
|
84 |
+
"num_feat_extract_layers": 7,
|
85 |
+
"num_hidden_layers": 24,
|
86 |
+
"num_negatives": 100,
|
87 |
+
"output_hidden_size": 1024,
|
88 |
+
"pad_token_id": 64,
|
89 |
+
"proj_codevector_dim": 768,
|
90 |
+
"tdnn_dilation": [
|
91 |
+
1,
|
92 |
+
2,
|
93 |
+
3,
|
94 |
+
1,
|
95 |
+
1
|
96 |
+
],
|
97 |
+
"tdnn_dim": [
|
98 |
+
512,
|
99 |
+
512,
|
100 |
+
512,
|
101 |
+
512,
|
102 |
+
1500
|
103 |
+
],
|
104 |
+
"tdnn_kernel": [
|
105 |
+
5,
|
106 |
+
3,
|
107 |
+
3,
|
108 |
+
1,
|
109 |
+
1
|
110 |
+
],
|
111 |
+
"torch_dtype": "float32",
|
112 |
+
"transformers_version": "4.21.1",
|
113 |
+
"use_weighted_layer_sum": false,
|
114 |
+
"vocab_size": 65,
|
115 |
+
"xvector_output_dim": 512
|
116 |
+
}
|
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": 0.0,
|
7 |
+
"return_attention_mask": true,
|
8 |
+
"sampling_rate": 16000
|
9 |
+
}
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9a54bbcf5ef5655a8844305d9b29c66fcb2137a007e342d2845355587ac5bf0
|
3 |
+
size 14575
|
scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53300b627bba7b77b6ab9a05ee948365e033b79f771001ca10632d15bc81dff4
|
3 |
+
size 557
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:81a01193337416f3456a183afeeafe6b67674def2d37e1c7594967f72ed79839
|
3 |
+
size 627
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|"}
|
trainer_state.json
ADDED
@@ -0,0 +1,739 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9996852376455776,
|
5 |
+
"global_step": 1191,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 9.999991603694376e-06,
|
13 |
+
"loss": 0.0036,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.02,
|
18 |
+
"learning_rate": 9.999983207388749e-06,
|
19 |
+
"loss": 0.0026,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.03,
|
24 |
+
"learning_rate": 9.999974811083124e-06,
|
25 |
+
"loss": 0.0013,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.03,
|
30 |
+
"learning_rate": 9.999966414777499e-06,
|
31 |
+
"loss": 0.0008,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.04,
|
36 |
+
"learning_rate": 9.999958018471874e-06,
|
37 |
+
"loss": 0.0003,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.05,
|
42 |
+
"learning_rate": 9.999949622166247e-06,
|
43 |
+
"loss": 0.003,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06,
|
48 |
+
"learning_rate": 9.999941225860622e-06,
|
49 |
+
"loss": 0.0031,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.07,
|
54 |
+
"learning_rate": 9.999932829554996e-06,
|
55 |
+
"loss": 0.0012,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.08,
|
60 |
+
"learning_rate": 9.999924433249371e-06,
|
61 |
+
"loss": 0.0006,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.08,
|
66 |
+
"learning_rate": 9.999916036943745e-06,
|
67 |
+
"loss": 0.0003,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.09,
|
72 |
+
"learning_rate": 9.99990764063812e-06,
|
73 |
+
"loss": 0.0041,
|
74 |
+
"step": 110
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.1,
|
78 |
+
"learning_rate": 9.999899244332494e-06,
|
79 |
+
"loss": 0.0018,
|
80 |
+
"step": 120
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.11,
|
84 |
+
"learning_rate": 9.99989084802687e-06,
|
85 |
+
"loss": 0.0011,
|
86 |
+
"step": 130
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.12,
|
90 |
+
"learning_rate": 9.999882451721244e-06,
|
91 |
+
"loss": 0.0006,
|
92 |
+
"step": 140
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.13,
|
96 |
+
"learning_rate": 9.999874895046181e-06,
|
97 |
+
"loss": 0.0013,
|
98 |
+
"step": 150
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.13,
|
102 |
+
"learning_rate": 9.999866498740554e-06,
|
103 |
+
"loss": 0.0039,
|
104 |
+
"step": 160
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.14,
|
108 |
+
"learning_rate": 9.999858102434929e-06,
|
109 |
+
"loss": 0.0019,
|
110 |
+
"step": 170
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.15,
|
114 |
+
"learning_rate": 9.999849706129304e-06,
|
115 |
+
"loss": 0.0011,
|
116 |
+
"step": 180
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.16,
|
120 |
+
"learning_rate": 9.999841309823679e-06,
|
121 |
+
"loss": 0.0004,
|
122 |
+
"step": 190
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.17,
|
126 |
+
"learning_rate": 9.999832913518052e-06,
|
127 |
+
"loss": 0.0004,
|
128 |
+
"step": 200
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.18,
|
132 |
+
"learning_rate": 9.999824517212427e-06,
|
133 |
+
"loss": 0.0028,
|
134 |
+
"step": 210
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.18,
|
138 |
+
"learning_rate": 9.999816120906802e-06,
|
139 |
+
"loss": 0.002,
|
140 |
+
"step": 220
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.19,
|
144 |
+
"learning_rate": 9.999807724601177e-06,
|
145 |
+
"loss": 0.0013,
|
146 |
+
"step": 230
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.2,
|
150 |
+
"learning_rate": 9.999799328295551e-06,
|
151 |
+
"loss": 0.0003,
|
152 |
+
"step": 240
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.21,
|
156 |
+
"learning_rate": 9.999790931989926e-06,
|
157 |
+
"loss": 0.0035,
|
158 |
+
"step": 250
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.22,
|
162 |
+
"learning_rate": 9.9997825356843e-06,
|
163 |
+
"loss": 0.0037,
|
164 |
+
"step": 260
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.23,
|
168 |
+
"learning_rate": 9.999774139378674e-06,
|
169 |
+
"loss": 0.0027,
|
170 |
+
"step": 270
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.24,
|
174 |
+
"learning_rate": 9.99976574307305e-06,
|
175 |
+
"loss": 0.001,
|
176 |
+
"step": 280
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.24,
|
180 |
+
"learning_rate": 9.999757346767424e-06,
|
181 |
+
"loss": 0.0005,
|
182 |
+
"step": 290
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.25,
|
186 |
+
"learning_rate": 9.999748950461797e-06,
|
187 |
+
"loss": 0.0011,
|
188 |
+
"step": 300
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.26,
|
192 |
+
"learning_rate": 9.999740554156172e-06,
|
193 |
+
"loss": 0.0032,
|
194 |
+
"step": 310
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.27,
|
198 |
+
"learning_rate": 9.999732157850547e-06,
|
199 |
+
"loss": 0.002,
|
200 |
+
"step": 320
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.28,
|
204 |
+
"learning_rate": 9.99972376154492e-06,
|
205 |
+
"loss": 0.0009,
|
206 |
+
"step": 330
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.29,
|
210 |
+
"learning_rate": 9.999715365239295e-06,
|
211 |
+
"loss": 0.0006,
|
212 |
+
"step": 340
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.29,
|
216 |
+
"learning_rate": 9.99970696893367e-06,
|
217 |
+
"loss": 0.0012,
|
218 |
+
"step": 350
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.3,
|
222 |
+
"learning_rate": 9.999698572628045e-06,
|
223 |
+
"loss": 0.0041,
|
224 |
+
"step": 360
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.31,
|
228 |
+
"learning_rate": 9.999690176322418e-06,
|
229 |
+
"loss": 0.0021,
|
230 |
+
"step": 370
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.32,
|
234 |
+
"learning_rate": 9.999681780016793e-06,
|
235 |
+
"loss": 0.0015,
|
236 |
+
"step": 380
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.33,
|
240 |
+
"learning_rate": 9.999673383711168e-06,
|
241 |
+
"loss": 0.0008,
|
242 |
+
"step": 390
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.34,
|
246 |
+
"learning_rate": 9.999664987405543e-06,
|
247 |
+
"loss": 0.0015,
|
248 |
+
"step": 400
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.34,
|
252 |
+
"learning_rate": 9.999656591099916e-06,
|
253 |
+
"loss": 0.003,
|
254 |
+
"step": 410
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.35,
|
258 |
+
"learning_rate": 9.99964819479429e-06,
|
259 |
+
"loss": 0.0015,
|
260 |
+
"step": 420
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.36,
|
264 |
+
"learning_rate": 9.999639798488666e-06,
|
265 |
+
"loss": 0.0022,
|
266 |
+
"step": 430
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.37,
|
270 |
+
"learning_rate": 9.99963140218304e-06,
|
271 |
+
"loss": 0.0004,
|
272 |
+
"step": 440
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.38,
|
276 |
+
"learning_rate": 9.999623005877415e-06,
|
277 |
+
"loss": 0.0001,
|
278 |
+
"step": 450
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.39,
|
282 |
+
"learning_rate": 9.999614609571789e-06,
|
283 |
+
"loss": 0.0033,
|
284 |
+
"step": 460
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.39,
|
288 |
+
"learning_rate": 9.999606213266164e-06,
|
289 |
+
"loss": 0.0013,
|
290 |
+
"step": 470
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.4,
|
294 |
+
"learning_rate": 9.999597816960538e-06,
|
295 |
+
"loss": 0.001,
|
296 |
+
"step": 480
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.41,
|
300 |
+
"learning_rate": 9.999589420654913e-06,
|
301 |
+
"loss": 0.0004,
|
302 |
+
"step": 490
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.42,
|
306 |
+
"learning_rate": 9.999581024349288e-06,
|
307 |
+
"loss": 0.0003,
|
308 |
+
"step": 500
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.43,
|
312 |
+
"learning_rate": 9.999572628043661e-06,
|
313 |
+
"loss": 0.0038,
|
314 |
+
"step": 510
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.44,
|
318 |
+
"learning_rate": 9.999564231738036e-06,
|
319 |
+
"loss": 0.0018,
|
320 |
+
"step": 520
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.44,
|
324 |
+
"learning_rate": 9.999555835432411e-06,
|
325 |
+
"loss": 0.0014,
|
326 |
+
"step": 530
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.45,
|
330 |
+
"learning_rate": 9.999547439126786e-06,
|
331 |
+
"loss": 0.0003,
|
332 |
+
"step": 540
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.46,
|
336 |
+
"learning_rate": 9.99953904282116e-06,
|
337 |
+
"loss": 0.0003,
|
338 |
+
"step": 550
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.47,
|
342 |
+
"learning_rate": 9.999530646515534e-06,
|
343 |
+
"loss": 0.0044,
|
344 |
+
"step": 560
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.48,
|
348 |
+
"learning_rate": 9.999522250209909e-06,
|
349 |
+
"loss": 0.0022,
|
350 |
+
"step": 570
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 0.49,
|
354 |
+
"learning_rate": 9.999513853904282e-06,
|
355 |
+
"loss": 0.0009,
|
356 |
+
"step": 580
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.5,
|
360 |
+
"learning_rate": 9.999505457598657e-06,
|
361 |
+
"loss": 0.0006,
|
362 |
+
"step": 590
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 0.5,
|
366 |
+
"learning_rate": 9.999497061293032e-06,
|
367 |
+
"loss": 0.0004,
|
368 |
+
"step": 600
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.51,
|
372 |
+
"learning_rate": 9.999488664987407e-06,
|
373 |
+
"loss": 0.0044,
|
374 |
+
"step": 610
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.52,
|
378 |
+
"learning_rate": 9.99948026868178e-06,
|
379 |
+
"loss": 0.002,
|
380 |
+
"step": 620
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.53,
|
384 |
+
"learning_rate": 9.999471872376155e-06,
|
385 |
+
"loss": 0.0013,
|
386 |
+
"step": 630
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 0.54,
|
390 |
+
"learning_rate": 9.99946347607053e-06,
|
391 |
+
"loss": 0.0004,
|
392 |
+
"step": 640
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 0.55,
|
396 |
+
"learning_rate": 9.999455079764905e-06,
|
397 |
+
"loss": 0.0004,
|
398 |
+
"step": 650
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 0.55,
|
402 |
+
"learning_rate": 9.999446683459278e-06,
|
403 |
+
"loss": 0.0034,
|
404 |
+
"step": 660
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 0.56,
|
408 |
+
"learning_rate": 9.999438287153653e-06,
|
409 |
+
"loss": 0.0023,
|
410 |
+
"step": 670
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 0.57,
|
414 |
+
"learning_rate": 9.999429890848028e-06,
|
415 |
+
"loss": 0.0017,
|
416 |
+
"step": 680
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.58,
|
420 |
+
"learning_rate": 9.999421494542402e-06,
|
421 |
+
"loss": 0.0004,
|
422 |
+
"step": 690
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.59,
|
426 |
+
"learning_rate": 9.999413098236777e-06,
|
427 |
+
"loss": 0.0012,
|
428 |
+
"step": 700
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.6,
|
432 |
+
"learning_rate": 9.999404701931152e-06,
|
433 |
+
"loss": 0.0046,
|
434 |
+
"step": 710
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.6,
|
438 |
+
"learning_rate": 9.999396305625525e-06,
|
439 |
+
"loss": 0.002,
|
440 |
+
"step": 720
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.61,
|
444 |
+
"learning_rate": 9.9993879093199e-06,
|
445 |
+
"loss": 0.0012,
|
446 |
+
"step": 730
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.62,
|
450 |
+
"learning_rate": 9.999379513014275e-06,
|
451 |
+
"loss": 0.0007,
|
452 |
+
"step": 740
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.63,
|
456 |
+
"learning_rate": 9.99937111670865e-06,
|
457 |
+
"loss": 0.0002,
|
458 |
+
"step": 750
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.64,
|
462 |
+
"learning_rate": 9.999362720403023e-06,
|
463 |
+
"loss": 0.0025,
|
464 |
+
"step": 760
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.65,
|
468 |
+
"learning_rate": 9.999354324097398e-06,
|
469 |
+
"loss": 0.0022,
|
470 |
+
"step": 770
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.65,
|
474 |
+
"learning_rate": 9.999345927791773e-06,
|
475 |
+
"loss": 0.0011,
|
476 |
+
"step": 780
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.66,
|
480 |
+
"learning_rate": 9.999337531486148e-06,
|
481 |
+
"loss": 0.0004,
|
482 |
+
"step": 790
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.67,
|
486 |
+
"learning_rate": 9.999329135180521e-06,
|
487 |
+
"loss": 0.0002,
|
488 |
+
"step": 800
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 0.68,
|
492 |
+
"learning_rate": 9.999320738874896e-06,
|
493 |
+
"loss": 0.0024,
|
494 |
+
"step": 810
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 0.69,
|
498 |
+
"learning_rate": 9.99931234256927e-06,
|
499 |
+
"loss": 0.0028,
|
500 |
+
"step": 820
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 0.7,
|
504 |
+
"learning_rate": 9.999303946263644e-06,
|
505 |
+
"loss": 0.0007,
|
506 |
+
"step": 830
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.71,
|
510 |
+
"learning_rate": 9.999295549958019e-06,
|
511 |
+
"loss": 0.0004,
|
512 |
+
"step": 840
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 0.71,
|
516 |
+
"learning_rate": 9.999287153652394e-06,
|
517 |
+
"loss": 0.0003,
|
518 |
+
"step": 850
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 0.72,
|
522 |
+
"learning_rate": 9.999278757346769e-06,
|
523 |
+
"loss": 0.0031,
|
524 |
+
"step": 860
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 0.73,
|
528 |
+
"learning_rate": 9.999270361041142e-06,
|
529 |
+
"loss": 0.0016,
|
530 |
+
"step": 870
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 0.74,
|
534 |
+
"learning_rate": 9.999261964735517e-06,
|
535 |
+
"loss": 0.0009,
|
536 |
+
"step": 880
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 0.75,
|
540 |
+
"learning_rate": 9.999253568429892e-06,
|
541 |
+
"loss": 0.0006,
|
542 |
+
"step": 890
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 0.76,
|
546 |
+
"learning_rate": 9.999245172124266e-06,
|
547 |
+
"loss": 0.0009,
|
548 |
+
"step": 900
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.76,
|
552 |
+
"learning_rate": 9.99923677581864e-06,
|
553 |
+
"loss": 0.0038,
|
554 |
+
"step": 910
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 0.77,
|
558 |
+
"learning_rate": 9.999228379513014e-06,
|
559 |
+
"loss": 0.0016,
|
560 |
+
"step": 920
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 0.78,
|
564 |
+
"learning_rate": 9.99921998320739e-06,
|
565 |
+
"loss": 0.0005,
|
566 |
+
"step": 930
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 0.79,
|
570 |
+
"learning_rate": 9.999211586901764e-06,
|
571 |
+
"loss": 0.0009,
|
572 |
+
"step": 940
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 0.8,
|
576 |
+
"learning_rate": 9.999203190596139e-06,
|
577 |
+
"loss": 0.0002,
|
578 |
+
"step": 950
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 0.81,
|
582 |
+
"learning_rate": 9.999194794290514e-06,
|
583 |
+
"loss": 0.0038,
|
584 |
+
"step": 960
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 0.81,
|
588 |
+
"learning_rate": 9.999186397984887e-06,
|
589 |
+
"loss": 0.0024,
|
590 |
+
"step": 970
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.82,
|
594 |
+
"learning_rate": 9.999178001679262e-06,
|
595 |
+
"loss": 0.0006,
|
596 |
+
"step": 980
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.83,
|
600 |
+
"learning_rate": 9.999169605373637e-06,
|
601 |
+
"loss": 0.0004,
|
602 |
+
"step": 990
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 0.84,
|
606 |
+
"learning_rate": 9.999161209068012e-06,
|
607 |
+
"loss": 0.0004,
|
608 |
+
"step": 1000
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 0.85,
|
612 |
+
"learning_rate": 9.999152812762385e-06,
|
613 |
+
"loss": 0.0026,
|
614 |
+
"step": 1010
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 0.86,
|
618 |
+
"learning_rate": 9.99914441645676e-06,
|
619 |
+
"loss": 0.002,
|
620 |
+
"step": 1020
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.86,
|
624 |
+
"learning_rate": 9.999136020151135e-06,
|
625 |
+
"loss": 0.0009,
|
626 |
+
"step": 1030
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.87,
|
630 |
+
"learning_rate": 9.99912762384551e-06,
|
631 |
+
"loss": 0.0007,
|
632 |
+
"step": 1040
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.88,
|
636 |
+
"learning_rate": 9.999119227539883e-06,
|
637 |
+
"loss": 0.0002,
|
638 |
+
"step": 1050
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 0.89,
|
642 |
+
"learning_rate": 9.999110831234258e-06,
|
643 |
+
"loss": 0.0043,
|
644 |
+
"step": 1060
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 0.9,
|
648 |
+
"learning_rate": 9.999102434928633e-06,
|
649 |
+
"loss": 0.003,
|
650 |
+
"step": 1070
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 0.91,
|
654 |
+
"learning_rate": 9.999094038623006e-06,
|
655 |
+
"loss": 0.0012,
|
656 |
+
"step": 1080
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.91,
|
660 |
+
"learning_rate": 9.99908564231738e-06,
|
661 |
+
"loss": 0.0006,
|
662 |
+
"step": 1090
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 0.92,
|
666 |
+
"learning_rate": 9.999077246011755e-06,
|
667 |
+
"loss": 0.0003,
|
668 |
+
"step": 1100
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.93,
|
672 |
+
"learning_rate": 9.99906884970613e-06,
|
673 |
+
"loss": 0.0032,
|
674 |
+
"step": 1110
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.94,
|
678 |
+
"learning_rate": 9.999060453400504e-06,
|
679 |
+
"loss": 0.0018,
|
680 |
+
"step": 1120
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 0.95,
|
684 |
+
"learning_rate": 9.999052057094878e-06,
|
685 |
+
"loss": 0.0012,
|
686 |
+
"step": 1130
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 0.96,
|
690 |
+
"learning_rate": 9.999043660789253e-06,
|
691 |
+
"loss": 0.0004,
|
692 |
+
"step": 1140
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 0.97,
|
696 |
+
"learning_rate": 9.999035264483628e-06,
|
697 |
+
"loss": 0.0002,
|
698 |
+
"step": 1150
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 0.97,
|
702 |
+
"learning_rate": 9.999026868178003e-06,
|
703 |
+
"loss": 0.0039,
|
704 |
+
"step": 1160
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 0.98,
|
708 |
+
"learning_rate": 9.999018471872376e-06,
|
709 |
+
"loss": 0.0019,
|
710 |
+
"step": 1170
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 0.99,
|
714 |
+
"learning_rate": 9.999010075566751e-06,
|
715 |
+
"loss": 0.0021,
|
716 |
+
"step": 1180
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 1.0,
|
720 |
+
"learning_rate": 9.999001679261126e-06,
|
721 |
+
"loss": 0.0003,
|
722 |
+
"step": 1190
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 1.0,
|
726 |
+
"eval_loss": 0.22505426406860352,
|
727 |
+
"eval_runtime": 703.0693,
|
728 |
+
"eval_samples_per_second": 19.22,
|
729 |
+
"eval_steps_per_second": 2.404,
|
730 |
+
"eval_wer": 0.1582263181749509,
|
731 |
+
"step": 1191
|
732 |
+
}
|
733 |
+
],
|
734 |
+
"max_steps": 11910000,
|
735 |
+
"num_train_epochs": 10000,
|
736 |
+
"total_flos": 6.75741609918661e+18,
|
737 |
+
"trial_name": null,
|
738 |
+
"trial_params": null
|
739 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:af14b43dafad80dd620481a9fcb9b7bbd93d09b6f45088c1764f05bb92ce113a
|
3 |
+
size 3323
|
vocab.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"ا": 0, "_": 1, "ك": 3, "ن": 4, "@": 5, "ذ": 6, "ٖ": 7, "ش": 8, "غ": 9, "آ": 10, "ص": 11, "ي": 12, "ث": 13, "n": 14, "ج": 15, "d": 16, "ئ": 17, "ة": 18, "ه": 19, "ز": 20, "ع": 21, "ف": 22, "i": 23, "r": 24, "v": 25, "ڸ": 26, "k": 27, "ِ": 28, "ء": 29, "ر": 30, "s": 31, "t": 32, "ى": 33, "ـ": 34, " ": 35, "ؤ": 36, "ق": 37, "ض": 38, "م": 39, "ل": 40, "…": 41, "ط": 42, "ت": 43, "ّ": 44, "c": 45, "ظ": 46, "e": 47, "؟": 48, "h": 49, "ب": 50, "o": 51, "س": 52, "a": 53, "د": 54, "p": 55, "أ": 56, "ْ": 57, "\n": 58, "خ": 59, "ح": 60, "و": 61, "إ": 62, "|": 2, "[UNK]": 63, "[PAD]": 64}
|