carlosaccardi
commited on
Commit
·
9295cba
1
Parent(s):
826f672
Upload 9 files
Browse files- config.json +42 -0
- optimizer.pt +3 -0
- preprocessor_config.json +0 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1012 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-base",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 512,
|
15 |
+
"decoder_attention_heads": 8,
|
16 |
+
"decoder_ffn_dim": 2048,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 6,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 8,
|
22 |
+
"encoder_ffn_dim": 2048,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 6,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 6,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.26.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:08029994fd4a9533aed27ea848f6de72f14a8675e8c2cafcd2579ace1aaa7a69
|
3 |
+
size 580893677
|
preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:431b9abf1caf70c0889225b4059c7593e6be0d3e5a5dd5adbe53a3a389037ed1
|
3 |
+
size 290456599
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:73ebd5cc03d2eb58905344c3401d23db896d5d52d470d6eea70f014854ec6f63
|
3 |
+
size 14503
|
scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e482ce49e1cc88705344ab368cf77604c9679f5d4ca9ec3a33b6a9edfdfe92fc
|
3 |
+
size 559
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:358d3cc5378fafc026eb4d987f7a358d0f6d6bfc9501f710e91d8952b2fb645b
|
3 |
+
size 623
|
trainer_state.json
ADDED
@@ -0,0 +1,1012 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 34.12384716732543,
|
3 |
+
"best_model_checkpoint": "/content/gdrive/MyDrive/whisper/whisper-swedish/checkpoint-4000",
|
4 |
+
"epoch": 5.174644243208279,
|
5 |
+
"global_step": 4000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 4.4e-07,
|
13 |
+
"loss": 2.6585,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.06,
|
18 |
+
"learning_rate": 9.400000000000001e-07,
|
19 |
+
"loss": 2.4329,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.1,
|
24 |
+
"learning_rate": 1.42e-06,
|
25 |
+
"loss": 2.1236,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.13,
|
30 |
+
"learning_rate": 1.9200000000000003e-06,
|
31 |
+
"loss": 1.5842,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.16,
|
36 |
+
"learning_rate": 2.42e-06,
|
37 |
+
"loss": 1.0271,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.19,
|
42 |
+
"learning_rate": 2.92e-06,
|
43 |
+
"loss": 0.86,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.23,
|
48 |
+
"learning_rate": 3.4200000000000007e-06,
|
49 |
+
"loss": 0.7182,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.26,
|
54 |
+
"learning_rate": 3.920000000000001e-06,
|
55 |
+
"loss": 0.6769,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.29,
|
60 |
+
"learning_rate": 4.42e-06,
|
61 |
+
"loss": 0.6119,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.32,
|
66 |
+
"learning_rate": 4.92e-06,
|
67 |
+
"loss": 0.6063,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.36,
|
72 |
+
"learning_rate": 5.420000000000001e-06,
|
73 |
+
"loss": 0.6213,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.39,
|
78 |
+
"learning_rate": 5.92e-06,
|
79 |
+
"loss": 0.5603,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.42,
|
84 |
+
"learning_rate": 6.42e-06,
|
85 |
+
"loss": 0.6051,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.45,
|
90 |
+
"learning_rate": 6.92e-06,
|
91 |
+
"loss": 0.534,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.49,
|
96 |
+
"learning_rate": 7.420000000000001e-06,
|
97 |
+
"loss": 0.5559,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.52,
|
102 |
+
"learning_rate": 7.92e-06,
|
103 |
+
"loss": 0.5403,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.55,
|
108 |
+
"learning_rate": 8.42e-06,
|
109 |
+
"loss": 0.5321,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.58,
|
114 |
+
"learning_rate": 8.920000000000001e-06,
|
115 |
+
"loss": 0.5398,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.61,
|
120 |
+
"learning_rate": 9.42e-06,
|
121 |
+
"loss": 0.4873,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.65,
|
126 |
+
"learning_rate": 9.920000000000002e-06,
|
127 |
+
"loss": 0.5191,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.68,
|
132 |
+
"learning_rate": 9.940000000000001e-06,
|
133 |
+
"loss": 0.5117,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.71,
|
138 |
+
"learning_rate": 9.86857142857143e-06,
|
139 |
+
"loss": 0.5096,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.74,
|
144 |
+
"learning_rate": 9.797142857142858e-06,
|
145 |
+
"loss": 0.5201,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.78,
|
150 |
+
"learning_rate": 9.725714285714287e-06,
|
151 |
+
"loss": 0.5157,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.81,
|
156 |
+
"learning_rate": 9.654285714285716e-06,
|
157 |
+
"loss": 0.524,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.84,
|
162 |
+
"learning_rate": 9.582857142857143e-06,
|
163 |
+
"loss": 0.4901,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.87,
|
168 |
+
"learning_rate": 9.511428571428572e-06,
|
169 |
+
"loss": 0.4796,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.91,
|
174 |
+
"learning_rate": 9.440000000000001e-06,
|
175 |
+
"loss": 0.4879,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.94,
|
180 |
+
"learning_rate": 9.368571428571428e-06,
|
181 |
+
"loss": 0.4943,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.97,
|
186 |
+
"learning_rate": 9.297142857142857e-06,
|
187 |
+
"loss": 0.4609,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.0,
|
192 |
+
"learning_rate": 9.225714285714286e-06,
|
193 |
+
"loss": 0.4669,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.03,
|
198 |
+
"learning_rate": 9.154285714285715e-06,
|
199 |
+
"loss": 0.3327,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 1.07,
|
204 |
+
"learning_rate": 9.082857142857143e-06,
|
205 |
+
"loss": 0.358,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 1.1,
|
210 |
+
"learning_rate": 9.011428571428572e-06,
|
211 |
+
"loss": 0.3451,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 1.13,
|
216 |
+
"learning_rate": 8.94e-06,
|
217 |
+
"loss": 0.329,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 1.16,
|
222 |
+
"learning_rate": 8.86857142857143e-06,
|
223 |
+
"loss": 0.3382,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.2,
|
228 |
+
"learning_rate": 8.797142857142857e-06,
|
229 |
+
"loss": 0.3448,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 1.23,
|
234 |
+
"learning_rate": 8.725714285714286e-06,
|
235 |
+
"loss": 0.3523,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 1.26,
|
240 |
+
"learning_rate": 8.654285714285715e-06,
|
241 |
+
"loss": 0.3514,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 1.29,
|
246 |
+
"learning_rate": 8.582857142857144e-06,
|
247 |
+
"loss": 0.3268,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.29,
|
252 |
+
"eval_loss": 0.4857306182384491,
|
253 |
+
"eval_runtime": 1095.0614,
|
254 |
+
"eval_samples_per_second": 4.629,
|
255 |
+
"eval_steps_per_second": 0.579,
|
256 |
+
"eval_wer": 35.86620418918556,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.33,
|
261 |
+
"learning_rate": 8.511428571428571e-06,
|
262 |
+
"loss": 0.3551,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.36,
|
267 |
+
"learning_rate": 8.44e-06,
|
268 |
+
"loss": 0.3179,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.39,
|
273 |
+
"learning_rate": 8.36857142857143e-06,
|
274 |
+
"loss": 0.3324,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.42,
|
279 |
+
"learning_rate": 8.297142857142859e-06,
|
280 |
+
"loss": 0.3432,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.46,
|
285 |
+
"learning_rate": 8.225714285714288e-06,
|
286 |
+
"loss": 0.3353,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 1.49,
|
291 |
+
"learning_rate": 8.154285714285715e-06,
|
292 |
+
"loss": 0.3206,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.52,
|
297 |
+
"learning_rate": 8.082857142857144e-06,
|
298 |
+
"loss": 0.3407,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.55,
|
303 |
+
"learning_rate": 8.011428571428573e-06,
|
304 |
+
"loss": 0.3458,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 1.58,
|
309 |
+
"learning_rate": 7.94e-06,
|
310 |
+
"loss": 0.3517,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.62,
|
315 |
+
"learning_rate": 7.86857142857143e-06,
|
316 |
+
"loss": 0.3184,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.65,
|
321 |
+
"learning_rate": 7.797142857142858e-06,
|
322 |
+
"loss": 0.3402,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 1.68,
|
327 |
+
"learning_rate": 7.725714285714286e-06,
|
328 |
+
"loss": 0.317,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 1.71,
|
333 |
+
"learning_rate": 7.654285714285715e-06,
|
334 |
+
"loss": 0.3213,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.75,
|
339 |
+
"learning_rate": 7.5828571428571444e-06,
|
340 |
+
"loss": 0.3158,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 1.78,
|
345 |
+
"learning_rate": 7.511428571428572e-06,
|
346 |
+
"loss": 0.3359,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 1.81,
|
351 |
+
"learning_rate": 7.440000000000001e-06,
|
352 |
+
"loss": 0.3314,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.84,
|
357 |
+
"learning_rate": 7.36857142857143e-06,
|
358 |
+
"loss": 0.3023,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 1.88,
|
363 |
+
"learning_rate": 7.297142857142858e-06,
|
364 |
+
"loss": 0.3067,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 1.91,
|
369 |
+
"learning_rate": 7.225714285714286e-06,
|
370 |
+
"loss": 0.3386,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 1.94,
|
375 |
+
"learning_rate": 7.154285714285715e-06,
|
376 |
+
"loss": 0.3093,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 1.97,
|
381 |
+
"learning_rate": 7.082857142857143e-06,
|
382 |
+
"loss": 0.3233,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 2.01,
|
387 |
+
"learning_rate": 7.011428571428572e-06,
|
388 |
+
"loss": 0.2847,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 2.04,
|
393 |
+
"learning_rate": 6.9400000000000005e-06,
|
394 |
+
"loss": 0.2002,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 2.07,
|
399 |
+
"learning_rate": 6.868571428571429e-06,
|
400 |
+
"loss": 0.1967,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 2.1,
|
405 |
+
"learning_rate": 6.797142857142858e-06,
|
406 |
+
"loss": 0.2005,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 2.13,
|
411 |
+
"learning_rate": 6.725714285714287e-06,
|
412 |
+
"loss": 0.1875,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 2.17,
|
417 |
+
"learning_rate": 6.654285714285716e-06,
|
418 |
+
"loss": 0.2139,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 2.2,
|
423 |
+
"learning_rate": 6.582857142857143e-06,
|
424 |
+
"loss": 0.2113,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 2.23,
|
429 |
+
"learning_rate": 6.511428571428572e-06,
|
430 |
+
"loss": 0.1913,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 2.26,
|
435 |
+
"learning_rate": 6.440000000000001e-06,
|
436 |
+
"loss": 0.1955,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 2.3,
|
441 |
+
"learning_rate": 6.368571428571429e-06,
|
442 |
+
"loss": 0.1984,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 2.33,
|
447 |
+
"learning_rate": 6.297142857142857e-06,
|
448 |
+
"loss": 0.1959,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 2.36,
|
453 |
+
"learning_rate": 6.225714285714286e-06,
|
454 |
+
"loss": 0.1956,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 2.39,
|
459 |
+
"learning_rate": 6.1542857142857145e-06,
|
460 |
+
"loss": 0.2016,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 2.43,
|
465 |
+
"learning_rate": 6.0828571428571435e-06,
|
466 |
+
"loss": 0.2029,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 2.46,
|
471 |
+
"learning_rate": 6.011428571428572e-06,
|
472 |
+
"loss": 0.1911,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 2.49,
|
477 |
+
"learning_rate": 5.94e-06,
|
478 |
+
"loss": 0.1904,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 2.52,
|
483 |
+
"learning_rate": 5.868571428571429e-06,
|
484 |
+
"loss": 0.1853,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 2.55,
|
489 |
+
"learning_rate": 5.797142857142858e-06,
|
490 |
+
"loss": 0.195,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 2.59,
|
495 |
+
"learning_rate": 5.725714285714287e-06,
|
496 |
+
"loss": 0.196,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 2.59,
|
501 |
+
"eval_loss": 0.4576089084148407,
|
502 |
+
"eval_runtime": 1112.3017,
|
503 |
+
"eval_samples_per_second": 4.557,
|
504 |
+
"eval_steps_per_second": 0.57,
|
505 |
+
"eval_wer": 34.55136995509666,
|
506 |
+
"step": 2000
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 2.62,
|
510 |
+
"learning_rate": 5.654285714285714e-06,
|
511 |
+
"loss": 0.198,
|
512 |
+
"step": 2025
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 2.65,
|
516 |
+
"learning_rate": 5.582857142857143e-06,
|
517 |
+
"loss": 0.1952,
|
518 |
+
"step": 2050
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 2.68,
|
522 |
+
"learning_rate": 5.511428571428572e-06,
|
523 |
+
"loss": 0.192,
|
524 |
+
"step": 2075
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 2.72,
|
528 |
+
"learning_rate": 5.4400000000000004e-06,
|
529 |
+
"loss": 0.1991,
|
530 |
+
"step": 2100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 2.75,
|
534 |
+
"learning_rate": 5.368571428571429e-06,
|
535 |
+
"loss": 0.2055,
|
536 |
+
"step": 2125
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.78,
|
540 |
+
"learning_rate": 5.297142857142858e-06,
|
541 |
+
"loss": 0.2006,
|
542 |
+
"step": 2150
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 2.81,
|
546 |
+
"learning_rate": 5.225714285714286e-06,
|
547 |
+
"loss": 0.1853,
|
548 |
+
"step": 2175
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 2.85,
|
552 |
+
"learning_rate": 5.154285714285715e-06,
|
553 |
+
"loss": 0.1883,
|
554 |
+
"step": 2200
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 2.88,
|
558 |
+
"learning_rate": 5.082857142857144e-06,
|
559 |
+
"loss": 0.1924,
|
560 |
+
"step": 2225
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 2.91,
|
564 |
+
"learning_rate": 5.011428571428571e-06,
|
565 |
+
"loss": 0.1911,
|
566 |
+
"step": 2250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 2.94,
|
570 |
+
"learning_rate": 4.94e-06,
|
571 |
+
"loss": 0.1937,
|
572 |
+
"step": 2275
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 2.98,
|
576 |
+
"learning_rate": 4.868571428571429e-06,
|
577 |
+
"loss": 0.1884,
|
578 |
+
"step": 2300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 3.01,
|
582 |
+
"learning_rate": 4.797142857142857e-06,
|
583 |
+
"loss": 0.1812,
|
584 |
+
"step": 2325
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 3.04,
|
588 |
+
"learning_rate": 4.725714285714286e-06,
|
589 |
+
"loss": 0.1154,
|
590 |
+
"step": 2350
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 3.07,
|
594 |
+
"learning_rate": 4.6542857142857145e-06,
|
595 |
+
"loss": 0.1184,
|
596 |
+
"step": 2375
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 3.1,
|
600 |
+
"learning_rate": 4.5828571428571435e-06,
|
601 |
+
"loss": 0.1205,
|
602 |
+
"step": 2400
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 3.14,
|
606 |
+
"learning_rate": 4.511428571428572e-06,
|
607 |
+
"loss": 0.1257,
|
608 |
+
"step": 2425
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 3.17,
|
612 |
+
"learning_rate": 4.440000000000001e-06,
|
613 |
+
"loss": 0.1304,
|
614 |
+
"step": 2450
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 3.2,
|
618 |
+
"learning_rate": 4.368571428571429e-06,
|
619 |
+
"loss": 0.1235,
|
620 |
+
"step": 2475
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 3.23,
|
624 |
+
"learning_rate": 4.297142857142858e-06,
|
625 |
+
"loss": 0.1222,
|
626 |
+
"step": 2500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 3.27,
|
630 |
+
"learning_rate": 4.225714285714286e-06,
|
631 |
+
"loss": 0.1226,
|
632 |
+
"step": 2525
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 3.3,
|
636 |
+
"learning_rate": 4.154285714285714e-06,
|
637 |
+
"loss": 0.1228,
|
638 |
+
"step": 2550
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 3.33,
|
642 |
+
"learning_rate": 4.082857142857143e-06,
|
643 |
+
"loss": 0.1273,
|
644 |
+
"step": 2575
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 3.36,
|
648 |
+
"learning_rate": 4.011428571428571e-06,
|
649 |
+
"loss": 0.1185,
|
650 |
+
"step": 2600
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 3.4,
|
654 |
+
"learning_rate": 3.94e-06,
|
655 |
+
"loss": 0.1225,
|
656 |
+
"step": 2625
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 3.43,
|
660 |
+
"learning_rate": 3.8685714285714286e-06,
|
661 |
+
"loss": 0.1248,
|
662 |
+
"step": 2650
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 3.46,
|
666 |
+
"learning_rate": 3.7971428571428576e-06,
|
667 |
+
"loss": 0.122,
|
668 |
+
"step": 2675
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 3.49,
|
672 |
+
"learning_rate": 3.7257142857142857e-06,
|
673 |
+
"loss": 0.1126,
|
674 |
+
"step": 2700
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 3.53,
|
678 |
+
"learning_rate": 3.6542857142857148e-06,
|
679 |
+
"loss": 0.1233,
|
680 |
+
"step": 2725
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 3.56,
|
684 |
+
"learning_rate": 3.582857142857143e-06,
|
685 |
+
"loss": 0.1137,
|
686 |
+
"step": 2750
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 3.59,
|
690 |
+
"learning_rate": 3.511428571428572e-06,
|
691 |
+
"loss": 0.1209,
|
692 |
+
"step": 2775
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 3.62,
|
696 |
+
"learning_rate": 3.44e-06,
|
697 |
+
"loss": 0.1202,
|
698 |
+
"step": 2800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 3.65,
|
702 |
+
"learning_rate": 3.3685714285714287e-06,
|
703 |
+
"loss": 0.1257,
|
704 |
+
"step": 2825
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 3.69,
|
708 |
+
"learning_rate": 3.2971428571428577e-06,
|
709 |
+
"loss": 0.1217,
|
710 |
+
"step": 2850
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 3.72,
|
714 |
+
"learning_rate": 3.225714285714286e-06,
|
715 |
+
"loss": 0.118,
|
716 |
+
"step": 2875
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 3.75,
|
720 |
+
"learning_rate": 3.154285714285715e-06,
|
721 |
+
"loss": 0.1194,
|
722 |
+
"step": 2900
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 3.78,
|
726 |
+
"learning_rate": 3.082857142857143e-06,
|
727 |
+
"loss": 0.1237,
|
728 |
+
"step": 2925
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 3.82,
|
732 |
+
"learning_rate": 3.0114285714285716e-06,
|
733 |
+
"loss": 0.1247,
|
734 |
+
"step": 2950
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 3.85,
|
738 |
+
"learning_rate": 2.9400000000000002e-06,
|
739 |
+
"loss": 0.117,
|
740 |
+
"step": 2975
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 3.88,
|
744 |
+
"learning_rate": 2.868571428571429e-06,
|
745 |
+
"loss": 0.122,
|
746 |
+
"step": 3000
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 3.88,
|
750 |
+
"eval_loss": 0.4576736092567444,
|
751 |
+
"eval_runtime": 1103.9323,
|
752 |
+
"eval_samples_per_second": 4.592,
|
753 |
+
"eval_steps_per_second": 0.574,
|
754 |
+
"eval_wer": 34.49490468124009,
|
755 |
+
"step": 3000
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 3.91,
|
759 |
+
"learning_rate": 2.797142857142857e-06,
|
760 |
+
"loss": 0.1202,
|
761 |
+
"step": 3025
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 3.95,
|
765 |
+
"learning_rate": 2.725714285714286e-06,
|
766 |
+
"loss": 0.1245,
|
767 |
+
"step": 3050
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 3.98,
|
771 |
+
"learning_rate": 2.654285714285714e-06,
|
772 |
+
"loss": 0.1344,
|
773 |
+
"step": 3075
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 4.01,
|
777 |
+
"learning_rate": 2.582857142857143e-06,
|
778 |
+
"loss": 0.1073,
|
779 |
+
"step": 3100
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 4.04,
|
783 |
+
"learning_rate": 2.5114285714285718e-06,
|
784 |
+
"loss": 0.093,
|
785 |
+
"step": 3125
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 4.08,
|
789 |
+
"learning_rate": 2.4400000000000004e-06,
|
790 |
+
"loss": 0.0861,
|
791 |
+
"step": 3150
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 4.11,
|
795 |
+
"learning_rate": 2.3685714285714285e-06,
|
796 |
+
"loss": 0.085,
|
797 |
+
"step": 3175
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 4.14,
|
801 |
+
"learning_rate": 2.297142857142857e-06,
|
802 |
+
"loss": 0.0791,
|
803 |
+
"step": 3200
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 4.17,
|
807 |
+
"learning_rate": 2.2257142857142857e-06,
|
808 |
+
"loss": 0.0799,
|
809 |
+
"step": 3225
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 4.2,
|
813 |
+
"learning_rate": 2.1542857142857147e-06,
|
814 |
+
"loss": 0.0823,
|
815 |
+
"step": 3250
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 4.24,
|
819 |
+
"learning_rate": 2.0828571428571433e-06,
|
820 |
+
"loss": 0.0836,
|
821 |
+
"step": 3275
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 4.27,
|
825 |
+
"learning_rate": 2.0114285714285715e-06,
|
826 |
+
"loss": 0.0818,
|
827 |
+
"step": 3300
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 4.3,
|
831 |
+
"learning_rate": 1.94e-06,
|
832 |
+
"loss": 0.0875,
|
833 |
+
"step": 3325
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 4.33,
|
837 |
+
"learning_rate": 1.8685714285714289e-06,
|
838 |
+
"loss": 0.0809,
|
839 |
+
"step": 3350
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 4.37,
|
843 |
+
"learning_rate": 1.7971428571428572e-06,
|
844 |
+
"loss": 0.0884,
|
845 |
+
"step": 3375
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 4.4,
|
849 |
+
"learning_rate": 1.7257142857142858e-06,
|
850 |
+
"loss": 0.0816,
|
851 |
+
"step": 3400
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 4.43,
|
855 |
+
"learning_rate": 1.6542857142857144e-06,
|
856 |
+
"loss": 0.0787,
|
857 |
+
"step": 3425
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 4.46,
|
861 |
+
"learning_rate": 1.582857142857143e-06,
|
862 |
+
"loss": 0.0759,
|
863 |
+
"step": 3450
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 4.5,
|
867 |
+
"learning_rate": 1.5114285714285714e-06,
|
868 |
+
"loss": 0.0889,
|
869 |
+
"step": 3475
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 4.53,
|
873 |
+
"learning_rate": 1.44e-06,
|
874 |
+
"loss": 0.0833,
|
875 |
+
"step": 3500
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 4.56,
|
879 |
+
"learning_rate": 1.3685714285714286e-06,
|
880 |
+
"loss": 0.0879,
|
881 |
+
"step": 3525
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 4.59,
|
885 |
+
"learning_rate": 1.2971428571428574e-06,
|
886 |
+
"loss": 0.0847,
|
887 |
+
"step": 3550
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 4.62,
|
891 |
+
"learning_rate": 1.2257142857142857e-06,
|
892 |
+
"loss": 0.0856,
|
893 |
+
"step": 3575
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 4.66,
|
897 |
+
"learning_rate": 1.1542857142857143e-06,
|
898 |
+
"loss": 0.077,
|
899 |
+
"step": 3600
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 4.69,
|
903 |
+
"learning_rate": 1.082857142857143e-06,
|
904 |
+
"loss": 0.077,
|
905 |
+
"step": 3625
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 4.72,
|
909 |
+
"learning_rate": 1.0114285714285715e-06,
|
910 |
+
"loss": 0.0868,
|
911 |
+
"step": 3650
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 4.75,
|
915 |
+
"learning_rate": 9.400000000000001e-07,
|
916 |
+
"loss": 0.0818,
|
917 |
+
"step": 3675
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 4.79,
|
921 |
+
"learning_rate": 8.685714285714286e-07,
|
922 |
+
"loss": 0.0845,
|
923 |
+
"step": 3700
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 4.82,
|
927 |
+
"learning_rate": 7.971428571428572e-07,
|
928 |
+
"loss": 0.0787,
|
929 |
+
"step": 3725
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 4.85,
|
933 |
+
"learning_rate": 7.257142857142857e-07,
|
934 |
+
"loss": 0.0848,
|
935 |
+
"step": 3750
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 4.88,
|
939 |
+
"learning_rate": 6.542857142857144e-07,
|
940 |
+
"loss": 0.0872,
|
941 |
+
"step": 3775
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 4.92,
|
945 |
+
"learning_rate": 5.82857142857143e-07,
|
946 |
+
"loss": 0.0814,
|
947 |
+
"step": 3800
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 4.95,
|
951 |
+
"learning_rate": 5.114285714285714e-07,
|
952 |
+
"loss": 0.0778,
|
953 |
+
"step": 3825
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 4.98,
|
957 |
+
"learning_rate": 4.4e-07,
|
958 |
+
"loss": 0.0857,
|
959 |
+
"step": 3850
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 5.01,
|
963 |
+
"learning_rate": 3.685714285714286e-07,
|
964 |
+
"loss": 0.0704,
|
965 |
+
"step": 3875
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 5.05,
|
969 |
+
"learning_rate": 2.9714285714285715e-07,
|
970 |
+
"loss": 0.0661,
|
971 |
+
"step": 3900
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 5.08,
|
975 |
+
"learning_rate": 2.2571428571428574e-07,
|
976 |
+
"loss": 0.0652,
|
977 |
+
"step": 3925
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 5.11,
|
981 |
+
"learning_rate": 1.542857142857143e-07,
|
982 |
+
"loss": 0.0627,
|
983 |
+
"step": 3950
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 5.14,
|
987 |
+
"learning_rate": 8.285714285714285e-08,
|
988 |
+
"loss": 0.0685,
|
989 |
+
"step": 3975
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 5.17,
|
993 |
+
"learning_rate": 1.142857142857143e-08,
|
994 |
+
"loss": 0.0671,
|
995 |
+
"step": 4000
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 5.17,
|
999 |
+
"eval_loss": 0.4669732451438904,
|
1000 |
+
"eval_runtime": 1101.5548,
|
1001 |
+
"eval_samples_per_second": 4.602,
|
1002 |
+
"eval_steps_per_second": 0.576,
|
1003 |
+
"eval_wer": 34.12384716732543,
|
1004 |
+
"step": 4000
|
1005 |
+
}
|
1006 |
+
],
|
1007 |
+
"max_steps": 4000,
|
1008 |
+
"num_train_epochs": 6,
|
1009 |
+
"total_flos": 4.1484453543936e+18,
|
1010 |
+
"trial_name": null,
|
1011 |
+
"trial_params": null
|
1012 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1204f22f411e02ebee9987424125bab562ee81339590adb85462abfd871441b5
|
3 |
+
size 3567
|