Commit
•
4909500
1
Parent(s):
0821a9a
Add training and evaluation results
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- added_tokens.json +109 -0
- all_results.json +14 -0
- checkpoint-1000/config.json +42 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/preprocessor_config.json +0 -0
- checkpoint-1000/pytorch_model.bin +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scaler.pt +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/trainer_state.json +265 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-2000/config.json +42 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/preprocessor_config.json +0 -0
- checkpoint-2000/pytorch_model.bin +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scaler.pt +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/trainer_state.json +514 -0
- checkpoint-2000/training_args.bin +3 -0
- checkpoint-3000/config.json +42 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/preprocessor_config.json +0 -0
- checkpoint-3000/pytorch_model.bin +3 -0
- checkpoint-3000/rng_state.pth +3 -0
- checkpoint-3000/scaler.pt +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/trainer_state.json +763 -0
- checkpoint-3000/training_args.bin +3 -0
- checkpoint-4000/config.json +42 -0
- checkpoint-4000/optimizer.pt +3 -0
- checkpoint-4000/preprocessor_config.json +0 -0
- checkpoint-4000/pytorch_model.bin +3 -0
- checkpoint-4000/rng_state.pth +3 -0
- checkpoint-4000/scaler.pt +3 -0
- checkpoint-4000/scheduler.pt +3 -0
- checkpoint-4000/trainer_state.json +1012 -0
- checkpoint-4000/training_args.bin +3 -0
- checkpoint-5000/config.json +42 -0
- checkpoint-5000/optimizer.pt +3 -0
- checkpoint-5000/preprocessor_config.json +0 -0
- checkpoint-5000/pytorch_model.bin +3 -0
- checkpoint-5000/rng_state.pth +3 -0
- checkpoint-5000/scaler.pt +3 -0
- checkpoint-5000/scheduler.pt +3 -0
- checkpoint-5000/trainer_state.json +1261 -0
- checkpoint-5000/training_args.bin +3 -0
- config.json +42 -0
- eval_results.json +8 -0
- merges.txt +0 -0
added_tokens.json
ADDED
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|af|>": 50327,
|
3 |
+
"<|am|>": 50334,
|
4 |
+
"<|ar|>": 50272,
|
5 |
+
"<|as|>": 50350,
|
6 |
+
"<|az|>": 50304,
|
7 |
+
"<|ba|>": 50355,
|
8 |
+
"<|be|>": 50330,
|
9 |
+
"<|bg|>": 50292,
|
10 |
+
"<|bn|>": 50302,
|
11 |
+
"<|bo|>": 50347,
|
12 |
+
"<|br|>": 50309,
|
13 |
+
"<|bs|>": 50315,
|
14 |
+
"<|ca|>": 50270,
|
15 |
+
"<|cs|>": 50283,
|
16 |
+
"<|cy|>": 50297,
|
17 |
+
"<|da|>": 50285,
|
18 |
+
"<|de|>": 50261,
|
19 |
+
"<|el|>": 50281,
|
20 |
+
"<|endoftext|>": 50257,
|
21 |
+
"<|en|>": 50259,
|
22 |
+
"<|es|>": 50262,
|
23 |
+
"<|et|>": 50307,
|
24 |
+
"<|eu|>": 50310,
|
25 |
+
"<|fa|>": 50300,
|
26 |
+
"<|fi|>": 50277,
|
27 |
+
"<|fo|>": 50338,
|
28 |
+
"<|fr|>": 50265,
|
29 |
+
"<|gl|>": 50319,
|
30 |
+
"<|gu|>": 50333,
|
31 |
+
"<|haw|>": 50352,
|
32 |
+
"<|ha|>": 50354,
|
33 |
+
"<|hi|>": 50276,
|
34 |
+
"<|hr|>": 50291,
|
35 |
+
"<|ht|>": 50339,
|
36 |
+
"<|hu|>": 50286,
|
37 |
+
"<|hy|>": 50312,
|
38 |
+
"<|id|>": 50275,
|
39 |
+
"<|is|>": 50311,
|
40 |
+
"<|it|>": 50274,
|
41 |
+
"<|iw|>": 50279,
|
42 |
+
"<|ja|>": 50266,
|
43 |
+
"<|jw|>": 50356,
|
44 |
+
"<|ka|>": 50329,
|
45 |
+
"<|kk|>": 50316,
|
46 |
+
"<|km|>": 50323,
|
47 |
+
"<|kn|>": 50306,
|
48 |
+
"<|ko|>": 50264,
|
49 |
+
"<|la|>": 50294,
|
50 |
+
"<|lb|>": 50345,
|
51 |
+
"<|ln|>": 50353,
|
52 |
+
"<|lo|>": 50336,
|
53 |
+
"<|lt|>": 50293,
|
54 |
+
"<|lv|>": 50301,
|
55 |
+
"<|mg|>": 50349,
|
56 |
+
"<|mi|>": 50295,
|
57 |
+
"<|mk|>": 50308,
|
58 |
+
"<|ml|>": 50296,
|
59 |
+
"<|mn|>": 50314,
|
60 |
+
"<|mr|>": 50320,
|
61 |
+
"<|ms|>": 50282,
|
62 |
+
"<|mt|>": 50343,
|
63 |
+
"<|my|>": 50346,
|
64 |
+
"<|ne|>": 50313,
|
65 |
+
"<|nl|>": 50271,
|
66 |
+
"<|nn|>": 50342,
|
67 |
+
"<|nocaptions|>": 50362,
|
68 |
+
"<|notimestamps|>": 50363,
|
69 |
+
"<|no|>": 50288,
|
70 |
+
"<|oc|>": 50328,
|
71 |
+
"<|pa|>": 50321,
|
72 |
+
"<|pl|>": 50269,
|
73 |
+
"<|ps|>": 50340,
|
74 |
+
"<|pt|>": 50267,
|
75 |
+
"<|ro|>": 50284,
|
76 |
+
"<|ru|>": 50263,
|
77 |
+
"<|sa|>": 50344,
|
78 |
+
"<|sd|>": 50332,
|
79 |
+
"<|si|>": 50322,
|
80 |
+
"<|sk|>": 50298,
|
81 |
+
"<|sl|>": 50305,
|
82 |
+
"<|sn|>": 50324,
|
83 |
+
"<|so|>": 50326,
|
84 |
+
"<|sq|>": 50317,
|
85 |
+
"<|sr|>": 50303,
|
86 |
+
"<|startoflm|>": 50360,
|
87 |
+
"<|startofprev|>": 50361,
|
88 |
+
"<|startoftranscript|>": 50258,
|
89 |
+
"<|su|>": 50357,
|
90 |
+
"<|sv|>": 50273,
|
91 |
+
"<|sw|>": 50318,
|
92 |
+
"<|ta|>": 50287,
|
93 |
+
"<|te|>": 50299,
|
94 |
+
"<|tg|>": 50331,
|
95 |
+
"<|th|>": 50289,
|
96 |
+
"<|tk|>": 50341,
|
97 |
+
"<|tl|>": 50348,
|
98 |
+
"<|transcribe|>": 50359,
|
99 |
+
"<|translate|>": 50358,
|
100 |
+
"<|tr|>": 50268,
|
101 |
+
"<|tt|>": 50351,
|
102 |
+
"<|uk|>": 50280,
|
103 |
+
"<|ur|>": 50290,
|
104 |
+
"<|uz|>": 50337,
|
105 |
+
"<|vi|>": 50278,
|
106 |
+
"<|yi|>": 50335,
|
107 |
+
"<|yo|>": 50325,
|
108 |
+
"<|zh|>": 50260
|
109 |
+
}
|
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 12.22,
|
3 |
+
"eval_loss": 0.4679179787635803,
|
4 |
+
"eval_runtime": 796.5779,
|
5 |
+
"eval_samples": 2894,
|
6 |
+
"eval_samples_per_second": 3.633,
|
7 |
+
"eval_steps_per_second": 0.227,
|
8 |
+
"eval_wer": 0.3209599593667993,
|
9 |
+
"train_loss": 0.06323314253110439,
|
10 |
+
"train_runtime": 18790.9423,
|
11 |
+
"train_samples": 6540,
|
12 |
+
"train_samples_per_second": 4.257,
|
13 |
+
"train_steps_per_second": 0.266
|
14 |
+
}
|
checkpoint-1000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 768,
|
15 |
+
"decoder_attention_heads": 12,
|
16 |
+
"decoder_ffn_dim": 3072,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 12,
|
22 |
+
"encoder_ffn_dim": 3072,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 12,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 12,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.24.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-1000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:403973e42d66853177c4cf5c97f822ccecfb2fd012e695841ac63928dc3b726a
|
3 |
+
size 1934158597
|
checkpoint-1000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-1000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bc9c4b5b667d96cc6e731bb368be69fce6ccb6d1a2441e3ad8454c028541a19c
|
3 |
+
size 967099139
|
checkpoint-1000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d251f16cc31f8a875a0930f5cb64eb5148fd07bd80921642c3c950df9216f781
|
3 |
+
size 14439
|
checkpoint-1000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9138267cab7020bb27f7ec0cbdf00702bc12067a0b377da6c68cf6dfd8468bb8
|
3 |
+
size 559
|
checkpoint-1000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9c78134860a2d7706519a27394ea9ac224151a062004ffdc714abfaa057d194
|
3 |
+
size 623
|
checkpoint-1000/trainer_state.json
ADDED
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.444987775061125,
|
5 |
+
"global_step": 1000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.06,
|
12 |
+
"learning_rate": 4.6000000000000004e-07,
|
13 |
+
"loss": 0.8917,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.12,
|
18 |
+
"learning_rate": 9.400000000000001e-07,
|
19 |
+
"loss": 0.8196,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.18,
|
24 |
+
"learning_rate": 1.44e-06,
|
25 |
+
"loss": 0.6476,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.24,
|
30 |
+
"learning_rate": 1.94e-06,
|
31 |
+
"loss": 0.4701,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.31,
|
36 |
+
"learning_rate": 2.4400000000000004e-06,
|
37 |
+
"loss": 0.5104,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.37,
|
42 |
+
"learning_rate": 2.9400000000000002e-06,
|
43 |
+
"loss": 0.3547,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.43,
|
48 |
+
"learning_rate": 3.44e-06,
|
49 |
+
"loss": 0.4427,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.49,
|
54 |
+
"learning_rate": 3.94e-06,
|
55 |
+
"loss": 0.3209,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.55,
|
60 |
+
"learning_rate": 4.440000000000001e-06,
|
61 |
+
"loss": 0.3846,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.61,
|
66 |
+
"learning_rate": 4.94e-06,
|
67 |
+
"loss": 0.252,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.67,
|
72 |
+
"learning_rate": 5.4400000000000004e-06,
|
73 |
+
"loss": 0.3344,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.73,
|
78 |
+
"learning_rate": 5.94e-06,
|
79 |
+
"loss": 0.2421,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.79,
|
84 |
+
"learning_rate": 6.440000000000001e-06,
|
85 |
+
"loss": 0.3204,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.86,
|
90 |
+
"learning_rate": 6.9400000000000005e-06,
|
91 |
+
"loss": 0.2263,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.92,
|
96 |
+
"learning_rate": 7.440000000000001e-06,
|
97 |
+
"loss": 0.2931,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.98,
|
102 |
+
"learning_rate": 7.94e-06,
|
103 |
+
"loss": 0.1939,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.04,
|
108 |
+
"learning_rate": 8.44e-06,
|
109 |
+
"loss": 0.2428,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.1,
|
114 |
+
"learning_rate": 8.94e-06,
|
115 |
+
"loss": 0.1915,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.16,
|
120 |
+
"learning_rate": 9.440000000000001e-06,
|
121 |
+
"loss": 0.1955,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.22,
|
126 |
+
"learning_rate": 9.940000000000001e-06,
|
127 |
+
"loss": 0.1694,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.28,
|
132 |
+
"learning_rate": 9.951111111111111e-06,
|
133 |
+
"loss": 0.1946,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.34,
|
138 |
+
"learning_rate": 9.895555555555557e-06,
|
139 |
+
"loss": 0.1883,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.41,
|
144 |
+
"learning_rate": 9.84e-06,
|
145 |
+
"loss": 0.1889,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.47,
|
150 |
+
"learning_rate": 9.784444444444445e-06,
|
151 |
+
"loss": 0.1598,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.53,
|
156 |
+
"learning_rate": 9.72888888888889e-06,
|
157 |
+
"loss": 0.1764,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.59,
|
162 |
+
"learning_rate": 9.673333333333334e-06,
|
163 |
+
"loss": 0.1652,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.65,
|
168 |
+
"learning_rate": 9.617777777777778e-06,
|
169 |
+
"loss": 0.1718,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.71,
|
174 |
+
"learning_rate": 9.562222222222223e-06,
|
175 |
+
"loss": 0.157,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.77,
|
180 |
+
"learning_rate": 9.506666666666667e-06,
|
181 |
+
"loss": 0.1685,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.83,
|
186 |
+
"learning_rate": 9.451111111111112e-06,
|
187 |
+
"loss": 0.1597,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.89,
|
192 |
+
"learning_rate": 9.395555555555556e-06,
|
193 |
+
"loss": 0.1641,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.96,
|
198 |
+
"learning_rate": 9.340000000000002e-06,
|
199 |
+
"loss": 0.1446,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.02,
|
204 |
+
"learning_rate": 9.284444444444444e-06,
|
205 |
+
"loss": 0.1322,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.08,
|
210 |
+
"learning_rate": 9.22888888888889e-06,
|
211 |
+
"loss": 0.0919,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.14,
|
216 |
+
"learning_rate": 9.173333333333334e-06,
|
217 |
+
"loss": 0.0799,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.2,
|
222 |
+
"learning_rate": 9.117777777777778e-06,
|
223 |
+
"loss": 0.0925,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.26,
|
228 |
+
"learning_rate": 9.062222222222224e-06,
|
229 |
+
"loss": 0.0725,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.32,
|
234 |
+
"learning_rate": 9.006666666666666e-06,
|
235 |
+
"loss": 0.1034,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.38,
|
240 |
+
"learning_rate": 8.951111111111112e-06,
|
241 |
+
"loss": 0.082,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.44,
|
246 |
+
"learning_rate": 8.895555555555556e-06,
|
247 |
+
"loss": 0.1011,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.44,
|
252 |
+
"eval_loss": 0.30747494101524353,
|
253 |
+
"eval_runtime": 799.5663,
|
254 |
+
"eval_samples_per_second": 3.619,
|
255 |
+
"eval_steps_per_second": 0.226,
|
256 |
+
"eval_wer": 0.3463133835604842,
|
257 |
+
"step": 1000
|
258 |
+
}
|
259 |
+
],
|
260 |
+
"max_steps": 5000,
|
261 |
+
"num_train_epochs": 13,
|
262 |
+
"total_flos": 4.61505772191744e+18,
|
263 |
+
"trial_name": null,
|
264 |
+
"trial_params": null
|
265 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fec8c5571ff240cb312ed7e0a1c73cd7df398177a18e7338f2a77971adf3a16c
|
3 |
+
size 3503
|
checkpoint-2000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 768,
|
15 |
+
"decoder_attention_heads": 12,
|
16 |
+
"decoder_ffn_dim": 3072,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 12,
|
22 |
+
"encoder_ffn_dim": 3072,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 12,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 12,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.24.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-2000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a3500652c2fae322fff2d07317d8d0fa8f61c29c04072adc77c018fb991ff0c0
|
3 |
+
size 1934158597
|
checkpoint-2000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-2000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:13990204bf0607e0cfb994d4431b729262f3d5b416a2f2e43eb9ed2db1917196
|
3 |
+
size 967099139
|
checkpoint-2000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:94ec4917e5e353cf5b3660cbdbfa0964af5ac4b48854bff5a516fdad649e9f62
|
3 |
+
size 14503
|
checkpoint-2000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dbfc503f2a238174eb8f397003c75c42d174d28561238697426efef457e97107
|
3 |
+
size 559
|
checkpoint-2000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8000b12dd8c86f786c29746bd14d823324e4964eddaa5dd7f79998b24b5438f6
|
3 |
+
size 623
|
checkpoint-2000/trainer_state.json
ADDED
@@ -0,0 +1,514 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 4.88997555012225,
|
5 |
+
"global_step": 2000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.06,
|
12 |
+
"learning_rate": 4.6000000000000004e-07,
|
13 |
+
"loss": 0.8917,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.12,
|
18 |
+
"learning_rate": 9.400000000000001e-07,
|
19 |
+
"loss": 0.8196,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.18,
|
24 |
+
"learning_rate": 1.44e-06,
|
25 |
+
"loss": 0.6476,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.24,
|
30 |
+
"learning_rate": 1.94e-06,
|
31 |
+
"loss": 0.4701,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.31,
|
36 |
+
"learning_rate": 2.4400000000000004e-06,
|
37 |
+
"loss": 0.5104,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.37,
|
42 |
+
"learning_rate": 2.9400000000000002e-06,
|
43 |
+
"loss": 0.3547,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.43,
|
48 |
+
"learning_rate": 3.44e-06,
|
49 |
+
"loss": 0.4427,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.49,
|
54 |
+
"learning_rate": 3.94e-06,
|
55 |
+
"loss": 0.3209,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.55,
|
60 |
+
"learning_rate": 4.440000000000001e-06,
|
61 |
+
"loss": 0.3846,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.61,
|
66 |
+
"learning_rate": 4.94e-06,
|
67 |
+
"loss": 0.252,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.67,
|
72 |
+
"learning_rate": 5.4400000000000004e-06,
|
73 |
+
"loss": 0.3344,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.73,
|
78 |
+
"learning_rate": 5.94e-06,
|
79 |
+
"loss": 0.2421,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.79,
|
84 |
+
"learning_rate": 6.440000000000001e-06,
|
85 |
+
"loss": 0.3204,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.86,
|
90 |
+
"learning_rate": 6.9400000000000005e-06,
|
91 |
+
"loss": 0.2263,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.92,
|
96 |
+
"learning_rate": 7.440000000000001e-06,
|
97 |
+
"loss": 0.2931,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.98,
|
102 |
+
"learning_rate": 7.94e-06,
|
103 |
+
"loss": 0.1939,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.04,
|
108 |
+
"learning_rate": 8.44e-06,
|
109 |
+
"loss": 0.2428,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.1,
|
114 |
+
"learning_rate": 8.94e-06,
|
115 |
+
"loss": 0.1915,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.16,
|
120 |
+
"learning_rate": 9.440000000000001e-06,
|
121 |
+
"loss": 0.1955,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.22,
|
126 |
+
"learning_rate": 9.940000000000001e-06,
|
127 |
+
"loss": 0.1694,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.28,
|
132 |
+
"learning_rate": 9.951111111111111e-06,
|
133 |
+
"loss": 0.1946,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.34,
|
138 |
+
"learning_rate": 9.895555555555557e-06,
|
139 |
+
"loss": 0.1883,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.41,
|
144 |
+
"learning_rate": 9.84e-06,
|
145 |
+
"loss": 0.1889,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.47,
|
150 |
+
"learning_rate": 9.784444444444445e-06,
|
151 |
+
"loss": 0.1598,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.53,
|
156 |
+
"learning_rate": 9.72888888888889e-06,
|
157 |
+
"loss": 0.1764,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.59,
|
162 |
+
"learning_rate": 9.673333333333334e-06,
|
163 |
+
"loss": 0.1652,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.65,
|
168 |
+
"learning_rate": 9.617777777777778e-06,
|
169 |
+
"loss": 0.1718,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.71,
|
174 |
+
"learning_rate": 9.562222222222223e-06,
|
175 |
+
"loss": 0.157,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.77,
|
180 |
+
"learning_rate": 9.506666666666667e-06,
|
181 |
+
"loss": 0.1685,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.83,
|
186 |
+
"learning_rate": 9.451111111111112e-06,
|
187 |
+
"loss": 0.1597,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.89,
|
192 |
+
"learning_rate": 9.395555555555556e-06,
|
193 |
+
"loss": 0.1641,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.96,
|
198 |
+
"learning_rate": 9.340000000000002e-06,
|
199 |
+
"loss": 0.1446,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.02,
|
204 |
+
"learning_rate": 9.284444444444444e-06,
|
205 |
+
"loss": 0.1322,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.08,
|
210 |
+
"learning_rate": 9.22888888888889e-06,
|
211 |
+
"loss": 0.0919,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.14,
|
216 |
+
"learning_rate": 9.173333333333334e-06,
|
217 |
+
"loss": 0.0799,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.2,
|
222 |
+
"learning_rate": 9.117777777777778e-06,
|
223 |
+
"loss": 0.0925,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.26,
|
228 |
+
"learning_rate": 9.062222222222224e-06,
|
229 |
+
"loss": 0.0725,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.32,
|
234 |
+
"learning_rate": 9.006666666666666e-06,
|
235 |
+
"loss": 0.1034,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.38,
|
240 |
+
"learning_rate": 8.951111111111112e-06,
|
241 |
+
"loss": 0.082,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.44,
|
246 |
+
"learning_rate": 8.895555555555556e-06,
|
247 |
+
"loss": 0.1011,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.44,
|
252 |
+
"eval_loss": 0.30747494101524353,
|
253 |
+
"eval_runtime": 799.5663,
|
254 |
+
"eval_samples_per_second": 3.619,
|
255 |
+
"eval_steps_per_second": 0.226,
|
256 |
+
"eval_wer": 0.3463133835604842,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 2.51,
|
261 |
+
"learning_rate": 8.84e-06,
|
262 |
+
"loss": 0.0762,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 2.57,
|
267 |
+
"learning_rate": 8.784444444444446e-06,
|
268 |
+
"loss": 0.1048,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 2.63,
|
273 |
+
"learning_rate": 8.72888888888889e-06,
|
274 |
+
"loss": 0.0804,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 2.69,
|
279 |
+
"learning_rate": 8.673333333333334e-06,
|
280 |
+
"loss": 0.1036,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 2.75,
|
285 |
+
"learning_rate": 8.617777777777778e-06,
|
286 |
+
"loss": 0.0789,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 2.81,
|
291 |
+
"learning_rate": 8.562222222222224e-06,
|
292 |
+
"loss": 0.0976,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 2.87,
|
297 |
+
"learning_rate": 8.506666666666668e-06,
|
298 |
+
"loss": 0.0822,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.93,
|
303 |
+
"learning_rate": 8.451111111111112e-06,
|
304 |
+
"loss": 0.0981,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 3.0,
|
309 |
+
"learning_rate": 8.395555555555557e-06,
|
310 |
+
"loss": 0.0591,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.06,
|
315 |
+
"learning_rate": 8.34e-06,
|
316 |
+
"loss": 0.0594,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 3.12,
|
321 |
+
"learning_rate": 8.284444444444446e-06,
|
322 |
+
"loss": 0.0348,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 3.18,
|
327 |
+
"learning_rate": 8.22888888888889e-06,
|
328 |
+
"loss": 0.062,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 3.24,
|
333 |
+
"learning_rate": 8.173333333333334e-06,
|
334 |
+
"loss": 0.035,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 3.3,
|
339 |
+
"learning_rate": 8.11777777777778e-06,
|
340 |
+
"loss": 0.0612,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 3.36,
|
345 |
+
"learning_rate": 8.062222222222222e-06,
|
346 |
+
"loss": 0.0328,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 3.42,
|
351 |
+
"learning_rate": 8.006666666666667e-06,
|
352 |
+
"loss": 0.0643,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 3.48,
|
357 |
+
"learning_rate": 7.951111111111111e-06,
|
358 |
+
"loss": 0.0306,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 3.55,
|
363 |
+
"learning_rate": 7.895555555555557e-06,
|
364 |
+
"loss": 0.0601,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 3.61,
|
369 |
+
"learning_rate": 7.840000000000001e-06,
|
370 |
+
"loss": 0.0304,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.67,
|
375 |
+
"learning_rate": 7.784444444444445e-06,
|
376 |
+
"loss": 0.0582,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.73,
|
381 |
+
"learning_rate": 7.72888888888889e-06,
|
382 |
+
"loss": 0.0341,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.79,
|
387 |
+
"learning_rate": 7.673333333333333e-06,
|
388 |
+
"loss": 0.0625,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.85,
|
393 |
+
"learning_rate": 7.617777777777778e-06,
|
394 |
+
"loss": 0.0365,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.91,
|
399 |
+
"learning_rate": 7.562222222222223e-06,
|
400 |
+
"loss": 0.0622,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 3.97,
|
405 |
+
"learning_rate": 7.506666666666668e-06,
|
406 |
+
"loss": 0.032,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 4.03,
|
411 |
+
"learning_rate": 7.451111111111111e-06,
|
412 |
+
"loss": 0.041,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 4.1,
|
417 |
+
"learning_rate": 7.395555555555556e-06,
|
418 |
+
"loss": 0.0217,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 4.16,
|
423 |
+
"learning_rate": 7.340000000000001e-06,
|
424 |
+
"loss": 0.0267,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 4.22,
|
429 |
+
"learning_rate": 7.284444444444445e-06,
|
430 |
+
"loss": 0.0241,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 4.28,
|
435 |
+
"learning_rate": 7.22888888888889e-06,
|
436 |
+
"loss": 0.0292,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 4.34,
|
441 |
+
"learning_rate": 7.173333333333335e-06,
|
442 |
+
"loss": 0.0204,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 4.4,
|
447 |
+
"learning_rate": 7.117777777777778e-06,
|
448 |
+
"loss": 0.0234,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 4.46,
|
453 |
+
"learning_rate": 7.062222222222223e-06,
|
454 |
+
"loss": 0.0211,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 4.52,
|
459 |
+
"learning_rate": 7.006666666666667e-06,
|
460 |
+
"loss": 0.031,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 4.58,
|
465 |
+
"learning_rate": 6.951111111111112e-06,
|
466 |
+
"loss": 0.0196,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 4.65,
|
471 |
+
"learning_rate": 6.8955555555555565e-06,
|
472 |
+
"loss": 0.0306,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 4.71,
|
477 |
+
"learning_rate": 6.8400000000000014e-06,
|
478 |
+
"loss": 0.0212,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 4.77,
|
483 |
+
"learning_rate": 6.784444444444445e-06,
|
484 |
+
"loss": 0.0271,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 4.83,
|
489 |
+
"learning_rate": 6.7288888888888895e-06,
|
490 |
+
"loss": 0.0224,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 4.89,
|
495 |
+
"learning_rate": 6.6733333333333335e-06,
|
496 |
+
"loss": 0.0264,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 4.89,
|
501 |
+
"eval_loss": 0.355797678232193,
|
502 |
+
"eval_runtime": 796.3076,
|
503 |
+
"eval_samples_per_second": 3.634,
|
504 |
+
"eval_steps_per_second": 0.227,
|
505 |
+
"eval_wer": 0.3312875645475324,
|
506 |
+
"step": 2000
|
507 |
+
}
|
508 |
+
],
|
509 |
+
"max_steps": 5000,
|
510 |
+
"num_train_epochs": 13,
|
511 |
+
"total_flos": 9.23011544383488e+18,
|
512 |
+
"trial_name": null,
|
513 |
+
"trial_params": null
|
514 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fec8c5571ff240cb312ed7e0a1c73cd7df398177a18e7338f2a77971adf3a16c
|
3 |
+
size 3503
|
checkpoint-3000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 768,
|
15 |
+
"decoder_attention_heads": 12,
|
16 |
+
"decoder_ffn_dim": 3072,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 12,
|
22 |
+
"encoder_ffn_dim": 3072,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 12,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 12,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.24.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-3000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e5affe8936eb235fa8f1566cc10830ad86ede93f1e5e2a264e5e8ed8b2639dbc
|
3 |
+
size 1934158597
|
checkpoint-3000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3aa5fd61a0202802f01ce2e6fdb26b9865b5c6f8d071929a02ae4c14f25b7e89
|
3 |
+
size 967099139
|
checkpoint-3000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:220f7ac8b4d5013a1f4dd429827376f14b5b5e2ef0cf5813128b94e3fefdba3c
|
3 |
+
size 14503
|
checkpoint-3000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c7af27207bb917aa5a1b5906fc66666e6b462688e81b41c6a40080fb84b3d7e2
|
3 |
+
size 559
|
checkpoint-3000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:af693833583db059e5cb294c9b5b90b3ab94346bc50b382ea5dac69799e82f35
|
3 |
+
size 623
|
checkpoint-3000/trainer_state.json
ADDED
@@ -0,0 +1,763 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 7.334963325183374,
|
5 |
+
"global_step": 3000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.06,
|
12 |
+
"learning_rate": 4.6000000000000004e-07,
|
13 |
+
"loss": 0.8917,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.12,
|
18 |
+
"learning_rate": 9.400000000000001e-07,
|
19 |
+
"loss": 0.8196,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.18,
|
24 |
+
"learning_rate": 1.44e-06,
|
25 |
+
"loss": 0.6476,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.24,
|
30 |
+
"learning_rate": 1.94e-06,
|
31 |
+
"loss": 0.4701,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.31,
|
36 |
+
"learning_rate": 2.4400000000000004e-06,
|
37 |
+
"loss": 0.5104,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.37,
|
42 |
+
"learning_rate": 2.9400000000000002e-06,
|
43 |
+
"loss": 0.3547,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.43,
|
48 |
+
"learning_rate": 3.44e-06,
|
49 |
+
"loss": 0.4427,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.49,
|
54 |
+
"learning_rate": 3.94e-06,
|
55 |
+
"loss": 0.3209,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.55,
|
60 |
+
"learning_rate": 4.440000000000001e-06,
|
61 |
+
"loss": 0.3846,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.61,
|
66 |
+
"learning_rate": 4.94e-06,
|
67 |
+
"loss": 0.252,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.67,
|
72 |
+
"learning_rate": 5.4400000000000004e-06,
|
73 |
+
"loss": 0.3344,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.73,
|
78 |
+
"learning_rate": 5.94e-06,
|
79 |
+
"loss": 0.2421,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.79,
|
84 |
+
"learning_rate": 6.440000000000001e-06,
|
85 |
+
"loss": 0.3204,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.86,
|
90 |
+
"learning_rate": 6.9400000000000005e-06,
|
91 |
+
"loss": 0.2263,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.92,
|
96 |
+
"learning_rate": 7.440000000000001e-06,
|
97 |
+
"loss": 0.2931,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.98,
|
102 |
+
"learning_rate": 7.94e-06,
|
103 |
+
"loss": 0.1939,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.04,
|
108 |
+
"learning_rate": 8.44e-06,
|
109 |
+
"loss": 0.2428,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.1,
|
114 |
+
"learning_rate": 8.94e-06,
|
115 |
+
"loss": 0.1915,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.16,
|
120 |
+
"learning_rate": 9.440000000000001e-06,
|
121 |
+
"loss": 0.1955,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.22,
|
126 |
+
"learning_rate": 9.940000000000001e-06,
|
127 |
+
"loss": 0.1694,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.28,
|
132 |
+
"learning_rate": 9.951111111111111e-06,
|
133 |
+
"loss": 0.1946,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.34,
|
138 |
+
"learning_rate": 9.895555555555557e-06,
|
139 |
+
"loss": 0.1883,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.41,
|
144 |
+
"learning_rate": 9.84e-06,
|
145 |
+
"loss": 0.1889,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.47,
|
150 |
+
"learning_rate": 9.784444444444445e-06,
|
151 |
+
"loss": 0.1598,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.53,
|
156 |
+
"learning_rate": 9.72888888888889e-06,
|
157 |
+
"loss": 0.1764,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.59,
|
162 |
+
"learning_rate": 9.673333333333334e-06,
|
163 |
+
"loss": 0.1652,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.65,
|
168 |
+
"learning_rate": 9.617777777777778e-06,
|
169 |
+
"loss": 0.1718,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.71,
|
174 |
+
"learning_rate": 9.562222222222223e-06,
|
175 |
+
"loss": 0.157,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.77,
|
180 |
+
"learning_rate": 9.506666666666667e-06,
|
181 |
+
"loss": 0.1685,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.83,
|
186 |
+
"learning_rate": 9.451111111111112e-06,
|
187 |
+
"loss": 0.1597,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.89,
|
192 |
+
"learning_rate": 9.395555555555556e-06,
|
193 |
+
"loss": 0.1641,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.96,
|
198 |
+
"learning_rate": 9.340000000000002e-06,
|
199 |
+
"loss": 0.1446,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.02,
|
204 |
+
"learning_rate": 9.284444444444444e-06,
|
205 |
+
"loss": 0.1322,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.08,
|
210 |
+
"learning_rate": 9.22888888888889e-06,
|
211 |
+
"loss": 0.0919,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.14,
|
216 |
+
"learning_rate": 9.173333333333334e-06,
|
217 |
+
"loss": 0.0799,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.2,
|
222 |
+
"learning_rate": 9.117777777777778e-06,
|
223 |
+
"loss": 0.0925,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.26,
|
228 |
+
"learning_rate": 9.062222222222224e-06,
|
229 |
+
"loss": 0.0725,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.32,
|
234 |
+
"learning_rate": 9.006666666666666e-06,
|
235 |
+
"loss": 0.1034,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.38,
|
240 |
+
"learning_rate": 8.951111111111112e-06,
|
241 |
+
"loss": 0.082,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.44,
|
246 |
+
"learning_rate": 8.895555555555556e-06,
|
247 |
+
"loss": 0.1011,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.44,
|
252 |
+
"eval_loss": 0.30747494101524353,
|
253 |
+
"eval_runtime": 799.5663,
|
254 |
+
"eval_samples_per_second": 3.619,
|
255 |
+
"eval_steps_per_second": 0.226,
|
256 |
+
"eval_wer": 0.3463133835604842,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 2.51,
|
261 |
+
"learning_rate": 8.84e-06,
|
262 |
+
"loss": 0.0762,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 2.57,
|
267 |
+
"learning_rate": 8.784444444444446e-06,
|
268 |
+
"loss": 0.1048,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 2.63,
|
273 |
+
"learning_rate": 8.72888888888889e-06,
|
274 |
+
"loss": 0.0804,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 2.69,
|
279 |
+
"learning_rate": 8.673333333333334e-06,
|
280 |
+
"loss": 0.1036,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 2.75,
|
285 |
+
"learning_rate": 8.617777777777778e-06,
|
286 |
+
"loss": 0.0789,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 2.81,
|
291 |
+
"learning_rate": 8.562222222222224e-06,
|
292 |
+
"loss": 0.0976,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 2.87,
|
297 |
+
"learning_rate": 8.506666666666668e-06,
|
298 |
+
"loss": 0.0822,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.93,
|
303 |
+
"learning_rate": 8.451111111111112e-06,
|
304 |
+
"loss": 0.0981,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 3.0,
|
309 |
+
"learning_rate": 8.395555555555557e-06,
|
310 |
+
"loss": 0.0591,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.06,
|
315 |
+
"learning_rate": 8.34e-06,
|
316 |
+
"loss": 0.0594,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 3.12,
|
321 |
+
"learning_rate": 8.284444444444446e-06,
|
322 |
+
"loss": 0.0348,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 3.18,
|
327 |
+
"learning_rate": 8.22888888888889e-06,
|
328 |
+
"loss": 0.062,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 3.24,
|
333 |
+
"learning_rate": 8.173333333333334e-06,
|
334 |
+
"loss": 0.035,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 3.3,
|
339 |
+
"learning_rate": 8.11777777777778e-06,
|
340 |
+
"loss": 0.0612,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 3.36,
|
345 |
+
"learning_rate": 8.062222222222222e-06,
|
346 |
+
"loss": 0.0328,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 3.42,
|
351 |
+
"learning_rate": 8.006666666666667e-06,
|
352 |
+
"loss": 0.0643,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 3.48,
|
357 |
+
"learning_rate": 7.951111111111111e-06,
|
358 |
+
"loss": 0.0306,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 3.55,
|
363 |
+
"learning_rate": 7.895555555555557e-06,
|
364 |
+
"loss": 0.0601,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 3.61,
|
369 |
+
"learning_rate": 7.840000000000001e-06,
|
370 |
+
"loss": 0.0304,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.67,
|
375 |
+
"learning_rate": 7.784444444444445e-06,
|
376 |
+
"loss": 0.0582,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.73,
|
381 |
+
"learning_rate": 7.72888888888889e-06,
|
382 |
+
"loss": 0.0341,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.79,
|
387 |
+
"learning_rate": 7.673333333333333e-06,
|
388 |
+
"loss": 0.0625,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.85,
|
393 |
+
"learning_rate": 7.617777777777778e-06,
|
394 |
+
"loss": 0.0365,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.91,
|
399 |
+
"learning_rate": 7.562222222222223e-06,
|
400 |
+
"loss": 0.0622,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 3.97,
|
405 |
+
"learning_rate": 7.506666666666668e-06,
|
406 |
+
"loss": 0.032,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 4.03,
|
411 |
+
"learning_rate": 7.451111111111111e-06,
|
412 |
+
"loss": 0.041,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 4.1,
|
417 |
+
"learning_rate": 7.395555555555556e-06,
|
418 |
+
"loss": 0.0217,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 4.16,
|
423 |
+
"learning_rate": 7.340000000000001e-06,
|
424 |
+
"loss": 0.0267,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 4.22,
|
429 |
+
"learning_rate": 7.284444444444445e-06,
|
430 |
+
"loss": 0.0241,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 4.28,
|
435 |
+
"learning_rate": 7.22888888888889e-06,
|
436 |
+
"loss": 0.0292,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 4.34,
|
441 |
+
"learning_rate": 7.173333333333335e-06,
|
442 |
+
"loss": 0.0204,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 4.4,
|
447 |
+
"learning_rate": 7.117777777777778e-06,
|
448 |
+
"loss": 0.0234,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 4.46,
|
453 |
+
"learning_rate": 7.062222222222223e-06,
|
454 |
+
"loss": 0.0211,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 4.52,
|
459 |
+
"learning_rate": 7.006666666666667e-06,
|
460 |
+
"loss": 0.031,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 4.58,
|
465 |
+
"learning_rate": 6.951111111111112e-06,
|
466 |
+
"loss": 0.0196,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 4.65,
|
471 |
+
"learning_rate": 6.8955555555555565e-06,
|
472 |
+
"loss": 0.0306,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 4.71,
|
477 |
+
"learning_rate": 6.8400000000000014e-06,
|
478 |
+
"loss": 0.0212,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 4.77,
|
483 |
+
"learning_rate": 6.784444444444445e-06,
|
484 |
+
"loss": 0.0271,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 4.83,
|
489 |
+
"learning_rate": 6.7288888888888895e-06,
|
490 |
+
"loss": 0.0224,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 4.89,
|
495 |
+
"learning_rate": 6.6733333333333335e-06,
|
496 |
+
"loss": 0.0264,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 4.89,
|
501 |
+
"eval_loss": 0.355797678232193,
|
502 |
+
"eval_runtime": 796.3076,
|
503 |
+
"eval_samples_per_second": 3.634,
|
504 |
+
"eval_steps_per_second": 0.227,
|
505 |
+
"eval_wer": 0.3312875645475324,
|
506 |
+
"step": 2000
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 4.95,
|
510 |
+
"learning_rate": 6.617777777777778e-06,
|
511 |
+
"loss": 0.0229,
|
512 |
+
"step": 2025
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 5.01,
|
516 |
+
"learning_rate": 6.562222222222223e-06,
|
517 |
+
"loss": 0.0245,
|
518 |
+
"step": 2050
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 5.07,
|
522 |
+
"learning_rate": 6.5066666666666665e-06,
|
523 |
+
"loss": 0.0149,
|
524 |
+
"step": 2075
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 5.13,
|
528 |
+
"learning_rate": 6.451111111111111e-06,
|
529 |
+
"loss": 0.0114,
|
530 |
+
"step": 2100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 5.2,
|
534 |
+
"learning_rate": 6.395555555555556e-06,
|
535 |
+
"loss": 0.0126,
|
536 |
+
"step": 2125
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 5.26,
|
540 |
+
"learning_rate": 6.34e-06,
|
541 |
+
"loss": 0.0093,
|
542 |
+
"step": 2150
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 5.32,
|
546 |
+
"learning_rate": 6.284444444444445e-06,
|
547 |
+
"loss": 0.014,
|
548 |
+
"step": 2175
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 5.38,
|
552 |
+
"learning_rate": 6.22888888888889e-06,
|
553 |
+
"loss": 0.0099,
|
554 |
+
"step": 2200
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 5.44,
|
558 |
+
"learning_rate": 6.173333333333333e-06,
|
559 |
+
"loss": 0.0121,
|
560 |
+
"step": 2225
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 5.5,
|
564 |
+
"learning_rate": 6.117777777777778e-06,
|
565 |
+
"loss": 0.008,
|
566 |
+
"step": 2250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 5.56,
|
570 |
+
"learning_rate": 6.062222222222223e-06,
|
571 |
+
"loss": 0.0156,
|
572 |
+
"step": 2275
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 5.62,
|
576 |
+
"learning_rate": 6.006666666666667e-06,
|
577 |
+
"loss": 0.0132,
|
578 |
+
"step": 2300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 5.68,
|
582 |
+
"learning_rate": 5.951111111111112e-06,
|
583 |
+
"loss": 0.0132,
|
584 |
+
"step": 2325
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 5.75,
|
588 |
+
"learning_rate": 5.895555555555557e-06,
|
589 |
+
"loss": 0.0093,
|
590 |
+
"step": 2350
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 5.81,
|
594 |
+
"learning_rate": 5.84e-06,
|
595 |
+
"loss": 0.0161,
|
596 |
+
"step": 2375
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 5.87,
|
600 |
+
"learning_rate": 5.784444444444445e-06,
|
601 |
+
"loss": 0.0114,
|
602 |
+
"step": 2400
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 5.93,
|
606 |
+
"learning_rate": 5.72888888888889e-06,
|
607 |
+
"loss": 0.016,
|
608 |
+
"step": 2425
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 5.99,
|
612 |
+
"learning_rate": 5.673333333333334e-06,
|
613 |
+
"loss": 0.0084,
|
614 |
+
"step": 2450
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 6.05,
|
618 |
+
"learning_rate": 5.617777777777779e-06,
|
619 |
+
"loss": 0.0114,
|
620 |
+
"step": 2475
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 6.11,
|
624 |
+
"learning_rate": 5.562222222222222e-06,
|
625 |
+
"loss": 0.0044,
|
626 |
+
"step": 2500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 6.17,
|
630 |
+
"learning_rate": 5.506666666666667e-06,
|
631 |
+
"loss": 0.007,
|
632 |
+
"step": 2525
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 6.23,
|
636 |
+
"learning_rate": 5.451111111111112e-06,
|
637 |
+
"loss": 0.0035,
|
638 |
+
"step": 2550
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 6.3,
|
642 |
+
"learning_rate": 5.3955555555555565e-06,
|
643 |
+
"loss": 0.009,
|
644 |
+
"step": 2575
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 6.36,
|
648 |
+
"learning_rate": 5.3400000000000005e-06,
|
649 |
+
"loss": 0.0038,
|
650 |
+
"step": 2600
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 6.42,
|
654 |
+
"learning_rate": 5.2844444444444454e-06,
|
655 |
+
"loss": 0.008,
|
656 |
+
"step": 2625
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 6.48,
|
660 |
+
"learning_rate": 5.228888888888889e-06,
|
661 |
+
"loss": 0.0035,
|
662 |
+
"step": 2650
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 6.54,
|
666 |
+
"learning_rate": 5.1733333333333335e-06,
|
667 |
+
"loss": 0.0079,
|
668 |
+
"step": 2675
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 6.6,
|
672 |
+
"learning_rate": 5.117777777777778e-06,
|
673 |
+
"loss": 0.0041,
|
674 |
+
"step": 2700
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 6.66,
|
678 |
+
"learning_rate": 5.062222222222222e-06,
|
679 |
+
"loss": 0.0077,
|
680 |
+
"step": 2725
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 6.72,
|
684 |
+
"learning_rate": 5.006666666666667e-06,
|
685 |
+
"loss": 0.004,
|
686 |
+
"step": 2750
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 6.78,
|
690 |
+
"learning_rate": 4.951111111111111e-06,
|
691 |
+
"loss": 0.0091,
|
692 |
+
"step": 2775
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 6.85,
|
696 |
+
"learning_rate": 4.895555555555556e-06,
|
697 |
+
"loss": 0.0041,
|
698 |
+
"step": 2800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 6.91,
|
702 |
+
"learning_rate": 4.84e-06,
|
703 |
+
"loss": 0.0072,
|
704 |
+
"step": 2825
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 6.97,
|
708 |
+
"learning_rate": 4.784444444444445e-06,
|
709 |
+
"loss": 0.0042,
|
710 |
+
"step": 2850
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 7.03,
|
714 |
+
"learning_rate": 4.728888888888889e-06,
|
715 |
+
"loss": 0.0053,
|
716 |
+
"step": 2875
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 7.09,
|
720 |
+
"learning_rate": 4.673333333333333e-06,
|
721 |
+
"loss": 0.0033,
|
722 |
+
"step": 2900
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 7.15,
|
726 |
+
"learning_rate": 4.617777777777778e-06,
|
727 |
+
"loss": 0.0022,
|
728 |
+
"step": 2925
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 7.21,
|
732 |
+
"learning_rate": 4.562222222222222e-06,
|
733 |
+
"loss": 0.003,
|
734 |
+
"step": 2950
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 7.27,
|
738 |
+
"learning_rate": 4.506666666666667e-06,
|
739 |
+
"loss": 0.003,
|
740 |
+
"step": 2975
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 7.33,
|
744 |
+
"learning_rate": 4.451111111111112e-06,
|
745 |
+
"loss": 0.0025,
|
746 |
+
"step": 3000
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 7.33,
|
750 |
+
"eval_loss": 0.42144426703453064,
|
751 |
+
"eval_runtime": 791.1382,
|
752 |
+
"eval_samples_per_second": 3.658,
|
753 |
+
"eval_steps_per_second": 0.229,
|
754 |
+
"eval_wer": 0.3259121307034623,
|
755 |
+
"step": 3000
|
756 |
+
}
|
757 |
+
],
|
758 |
+
"max_steps": 5000,
|
759 |
+
"num_train_epochs": 13,
|
760 |
+
"total_flos": 1.384401882415104e+19,
|
761 |
+
"trial_name": null,
|
762 |
+
"trial_params": null
|
763 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fec8c5571ff240cb312ed7e0a1c73cd7df398177a18e7338f2a77971adf3a16c
|
3 |
+
size 3503
|
checkpoint-4000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 768,
|
15 |
+
"decoder_attention_heads": 12,
|
16 |
+
"decoder_ffn_dim": 3072,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 12,
|
22 |
+
"encoder_ffn_dim": 3072,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 12,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 12,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.24.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-4000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b6d37d22fe7603a412e0e65705fe8b0f483db1d935df05b1b6a95363cf4c960f
|
3 |
+
size 1934158597
|
checkpoint-4000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-4000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d040ad0a00b620e290f2624fba2349439514b2d46bdcdcec19c260b64cc9b31
|
3 |
+
size 967099139
|
checkpoint-4000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7f08035b190fdc80f2676a5fbe1d345a930da42051bc5543f9101e776a277d53
|
3 |
+
size 14503
|
checkpoint-4000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:007bd789dd0a2ab6c4de81d9c6a2dfc613c01405c7450d479ad0571f50727c96
|
3 |
+
size 559
|
checkpoint-4000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:937ebf4b8852053d18f5cf1290182a44c88ac326bb951d800b50d17da440d805
|
3 |
+
size 623
|
checkpoint-4000/trainer_state.json
ADDED
@@ -0,0 +1,1012 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 9.7799511002445,
|
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": 0.06,
|
12 |
+
"learning_rate": 4.6000000000000004e-07,
|
13 |
+
"loss": 0.8917,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.12,
|
18 |
+
"learning_rate": 9.400000000000001e-07,
|
19 |
+
"loss": 0.8196,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.18,
|
24 |
+
"learning_rate": 1.44e-06,
|
25 |
+
"loss": 0.6476,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.24,
|
30 |
+
"learning_rate": 1.94e-06,
|
31 |
+
"loss": 0.4701,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.31,
|
36 |
+
"learning_rate": 2.4400000000000004e-06,
|
37 |
+
"loss": 0.5104,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.37,
|
42 |
+
"learning_rate": 2.9400000000000002e-06,
|
43 |
+
"loss": 0.3547,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.43,
|
48 |
+
"learning_rate": 3.44e-06,
|
49 |
+
"loss": 0.4427,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.49,
|
54 |
+
"learning_rate": 3.94e-06,
|
55 |
+
"loss": 0.3209,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.55,
|
60 |
+
"learning_rate": 4.440000000000001e-06,
|
61 |
+
"loss": 0.3846,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.61,
|
66 |
+
"learning_rate": 4.94e-06,
|
67 |
+
"loss": 0.252,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.67,
|
72 |
+
"learning_rate": 5.4400000000000004e-06,
|
73 |
+
"loss": 0.3344,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.73,
|
78 |
+
"learning_rate": 5.94e-06,
|
79 |
+
"loss": 0.2421,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.79,
|
84 |
+
"learning_rate": 6.440000000000001e-06,
|
85 |
+
"loss": 0.3204,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.86,
|
90 |
+
"learning_rate": 6.9400000000000005e-06,
|
91 |
+
"loss": 0.2263,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.92,
|
96 |
+
"learning_rate": 7.440000000000001e-06,
|
97 |
+
"loss": 0.2931,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.98,
|
102 |
+
"learning_rate": 7.94e-06,
|
103 |
+
"loss": 0.1939,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.04,
|
108 |
+
"learning_rate": 8.44e-06,
|
109 |
+
"loss": 0.2428,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.1,
|
114 |
+
"learning_rate": 8.94e-06,
|
115 |
+
"loss": 0.1915,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.16,
|
120 |
+
"learning_rate": 9.440000000000001e-06,
|
121 |
+
"loss": 0.1955,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.22,
|
126 |
+
"learning_rate": 9.940000000000001e-06,
|
127 |
+
"loss": 0.1694,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.28,
|
132 |
+
"learning_rate": 9.951111111111111e-06,
|
133 |
+
"loss": 0.1946,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.34,
|
138 |
+
"learning_rate": 9.895555555555557e-06,
|
139 |
+
"loss": 0.1883,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.41,
|
144 |
+
"learning_rate": 9.84e-06,
|
145 |
+
"loss": 0.1889,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.47,
|
150 |
+
"learning_rate": 9.784444444444445e-06,
|
151 |
+
"loss": 0.1598,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.53,
|
156 |
+
"learning_rate": 9.72888888888889e-06,
|
157 |
+
"loss": 0.1764,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.59,
|
162 |
+
"learning_rate": 9.673333333333334e-06,
|
163 |
+
"loss": 0.1652,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.65,
|
168 |
+
"learning_rate": 9.617777777777778e-06,
|
169 |
+
"loss": 0.1718,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.71,
|
174 |
+
"learning_rate": 9.562222222222223e-06,
|
175 |
+
"loss": 0.157,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.77,
|
180 |
+
"learning_rate": 9.506666666666667e-06,
|
181 |
+
"loss": 0.1685,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.83,
|
186 |
+
"learning_rate": 9.451111111111112e-06,
|
187 |
+
"loss": 0.1597,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.89,
|
192 |
+
"learning_rate": 9.395555555555556e-06,
|
193 |
+
"loss": 0.1641,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.96,
|
198 |
+
"learning_rate": 9.340000000000002e-06,
|
199 |
+
"loss": 0.1446,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.02,
|
204 |
+
"learning_rate": 9.284444444444444e-06,
|
205 |
+
"loss": 0.1322,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.08,
|
210 |
+
"learning_rate": 9.22888888888889e-06,
|
211 |
+
"loss": 0.0919,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.14,
|
216 |
+
"learning_rate": 9.173333333333334e-06,
|
217 |
+
"loss": 0.0799,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.2,
|
222 |
+
"learning_rate": 9.117777777777778e-06,
|
223 |
+
"loss": 0.0925,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.26,
|
228 |
+
"learning_rate": 9.062222222222224e-06,
|
229 |
+
"loss": 0.0725,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.32,
|
234 |
+
"learning_rate": 9.006666666666666e-06,
|
235 |
+
"loss": 0.1034,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.38,
|
240 |
+
"learning_rate": 8.951111111111112e-06,
|
241 |
+
"loss": 0.082,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.44,
|
246 |
+
"learning_rate": 8.895555555555556e-06,
|
247 |
+
"loss": 0.1011,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.44,
|
252 |
+
"eval_loss": 0.30747494101524353,
|
253 |
+
"eval_runtime": 799.5663,
|
254 |
+
"eval_samples_per_second": 3.619,
|
255 |
+
"eval_steps_per_second": 0.226,
|
256 |
+
"eval_wer": 0.3463133835604842,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 2.51,
|
261 |
+
"learning_rate": 8.84e-06,
|
262 |
+
"loss": 0.0762,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 2.57,
|
267 |
+
"learning_rate": 8.784444444444446e-06,
|
268 |
+
"loss": 0.1048,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 2.63,
|
273 |
+
"learning_rate": 8.72888888888889e-06,
|
274 |
+
"loss": 0.0804,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 2.69,
|
279 |
+
"learning_rate": 8.673333333333334e-06,
|
280 |
+
"loss": 0.1036,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 2.75,
|
285 |
+
"learning_rate": 8.617777777777778e-06,
|
286 |
+
"loss": 0.0789,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 2.81,
|
291 |
+
"learning_rate": 8.562222222222224e-06,
|
292 |
+
"loss": 0.0976,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 2.87,
|
297 |
+
"learning_rate": 8.506666666666668e-06,
|
298 |
+
"loss": 0.0822,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.93,
|
303 |
+
"learning_rate": 8.451111111111112e-06,
|
304 |
+
"loss": 0.0981,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 3.0,
|
309 |
+
"learning_rate": 8.395555555555557e-06,
|
310 |
+
"loss": 0.0591,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.06,
|
315 |
+
"learning_rate": 8.34e-06,
|
316 |
+
"loss": 0.0594,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 3.12,
|
321 |
+
"learning_rate": 8.284444444444446e-06,
|
322 |
+
"loss": 0.0348,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 3.18,
|
327 |
+
"learning_rate": 8.22888888888889e-06,
|
328 |
+
"loss": 0.062,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 3.24,
|
333 |
+
"learning_rate": 8.173333333333334e-06,
|
334 |
+
"loss": 0.035,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 3.3,
|
339 |
+
"learning_rate": 8.11777777777778e-06,
|
340 |
+
"loss": 0.0612,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 3.36,
|
345 |
+
"learning_rate": 8.062222222222222e-06,
|
346 |
+
"loss": 0.0328,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 3.42,
|
351 |
+
"learning_rate": 8.006666666666667e-06,
|
352 |
+
"loss": 0.0643,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 3.48,
|
357 |
+
"learning_rate": 7.951111111111111e-06,
|
358 |
+
"loss": 0.0306,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 3.55,
|
363 |
+
"learning_rate": 7.895555555555557e-06,
|
364 |
+
"loss": 0.0601,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 3.61,
|
369 |
+
"learning_rate": 7.840000000000001e-06,
|
370 |
+
"loss": 0.0304,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.67,
|
375 |
+
"learning_rate": 7.784444444444445e-06,
|
376 |
+
"loss": 0.0582,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.73,
|
381 |
+
"learning_rate": 7.72888888888889e-06,
|
382 |
+
"loss": 0.0341,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.79,
|
387 |
+
"learning_rate": 7.673333333333333e-06,
|
388 |
+
"loss": 0.0625,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.85,
|
393 |
+
"learning_rate": 7.617777777777778e-06,
|
394 |
+
"loss": 0.0365,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.91,
|
399 |
+
"learning_rate": 7.562222222222223e-06,
|
400 |
+
"loss": 0.0622,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 3.97,
|
405 |
+
"learning_rate": 7.506666666666668e-06,
|
406 |
+
"loss": 0.032,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 4.03,
|
411 |
+
"learning_rate": 7.451111111111111e-06,
|
412 |
+
"loss": 0.041,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 4.1,
|
417 |
+
"learning_rate": 7.395555555555556e-06,
|
418 |
+
"loss": 0.0217,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 4.16,
|
423 |
+
"learning_rate": 7.340000000000001e-06,
|
424 |
+
"loss": 0.0267,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 4.22,
|
429 |
+
"learning_rate": 7.284444444444445e-06,
|
430 |
+
"loss": 0.0241,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 4.28,
|
435 |
+
"learning_rate": 7.22888888888889e-06,
|
436 |
+
"loss": 0.0292,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 4.34,
|
441 |
+
"learning_rate": 7.173333333333335e-06,
|
442 |
+
"loss": 0.0204,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 4.4,
|
447 |
+
"learning_rate": 7.117777777777778e-06,
|
448 |
+
"loss": 0.0234,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 4.46,
|
453 |
+
"learning_rate": 7.062222222222223e-06,
|
454 |
+
"loss": 0.0211,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 4.52,
|
459 |
+
"learning_rate": 7.006666666666667e-06,
|
460 |
+
"loss": 0.031,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 4.58,
|
465 |
+
"learning_rate": 6.951111111111112e-06,
|
466 |
+
"loss": 0.0196,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 4.65,
|
471 |
+
"learning_rate": 6.8955555555555565e-06,
|
472 |
+
"loss": 0.0306,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 4.71,
|
477 |
+
"learning_rate": 6.8400000000000014e-06,
|
478 |
+
"loss": 0.0212,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 4.77,
|
483 |
+
"learning_rate": 6.784444444444445e-06,
|
484 |
+
"loss": 0.0271,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 4.83,
|
489 |
+
"learning_rate": 6.7288888888888895e-06,
|
490 |
+
"loss": 0.0224,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 4.89,
|
495 |
+
"learning_rate": 6.6733333333333335e-06,
|
496 |
+
"loss": 0.0264,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 4.89,
|
501 |
+
"eval_loss": 0.355797678232193,
|
502 |
+
"eval_runtime": 796.3076,
|
503 |
+
"eval_samples_per_second": 3.634,
|
504 |
+
"eval_steps_per_second": 0.227,
|
505 |
+
"eval_wer": 0.3312875645475324,
|
506 |
+
"step": 2000
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 4.95,
|
510 |
+
"learning_rate": 6.617777777777778e-06,
|
511 |
+
"loss": 0.0229,
|
512 |
+
"step": 2025
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 5.01,
|
516 |
+
"learning_rate": 6.562222222222223e-06,
|
517 |
+
"loss": 0.0245,
|
518 |
+
"step": 2050
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 5.07,
|
522 |
+
"learning_rate": 6.5066666666666665e-06,
|
523 |
+
"loss": 0.0149,
|
524 |
+
"step": 2075
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 5.13,
|
528 |
+
"learning_rate": 6.451111111111111e-06,
|
529 |
+
"loss": 0.0114,
|
530 |
+
"step": 2100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 5.2,
|
534 |
+
"learning_rate": 6.395555555555556e-06,
|
535 |
+
"loss": 0.0126,
|
536 |
+
"step": 2125
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 5.26,
|
540 |
+
"learning_rate": 6.34e-06,
|
541 |
+
"loss": 0.0093,
|
542 |
+
"step": 2150
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 5.32,
|
546 |
+
"learning_rate": 6.284444444444445e-06,
|
547 |
+
"loss": 0.014,
|
548 |
+
"step": 2175
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 5.38,
|
552 |
+
"learning_rate": 6.22888888888889e-06,
|
553 |
+
"loss": 0.0099,
|
554 |
+
"step": 2200
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 5.44,
|
558 |
+
"learning_rate": 6.173333333333333e-06,
|
559 |
+
"loss": 0.0121,
|
560 |
+
"step": 2225
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 5.5,
|
564 |
+
"learning_rate": 6.117777777777778e-06,
|
565 |
+
"loss": 0.008,
|
566 |
+
"step": 2250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 5.56,
|
570 |
+
"learning_rate": 6.062222222222223e-06,
|
571 |
+
"loss": 0.0156,
|
572 |
+
"step": 2275
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 5.62,
|
576 |
+
"learning_rate": 6.006666666666667e-06,
|
577 |
+
"loss": 0.0132,
|
578 |
+
"step": 2300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 5.68,
|
582 |
+
"learning_rate": 5.951111111111112e-06,
|
583 |
+
"loss": 0.0132,
|
584 |
+
"step": 2325
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 5.75,
|
588 |
+
"learning_rate": 5.895555555555557e-06,
|
589 |
+
"loss": 0.0093,
|
590 |
+
"step": 2350
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 5.81,
|
594 |
+
"learning_rate": 5.84e-06,
|
595 |
+
"loss": 0.0161,
|
596 |
+
"step": 2375
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 5.87,
|
600 |
+
"learning_rate": 5.784444444444445e-06,
|
601 |
+
"loss": 0.0114,
|
602 |
+
"step": 2400
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 5.93,
|
606 |
+
"learning_rate": 5.72888888888889e-06,
|
607 |
+
"loss": 0.016,
|
608 |
+
"step": 2425
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 5.99,
|
612 |
+
"learning_rate": 5.673333333333334e-06,
|
613 |
+
"loss": 0.0084,
|
614 |
+
"step": 2450
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 6.05,
|
618 |
+
"learning_rate": 5.617777777777779e-06,
|
619 |
+
"loss": 0.0114,
|
620 |
+
"step": 2475
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 6.11,
|
624 |
+
"learning_rate": 5.562222222222222e-06,
|
625 |
+
"loss": 0.0044,
|
626 |
+
"step": 2500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 6.17,
|
630 |
+
"learning_rate": 5.506666666666667e-06,
|
631 |
+
"loss": 0.007,
|
632 |
+
"step": 2525
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 6.23,
|
636 |
+
"learning_rate": 5.451111111111112e-06,
|
637 |
+
"loss": 0.0035,
|
638 |
+
"step": 2550
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 6.3,
|
642 |
+
"learning_rate": 5.3955555555555565e-06,
|
643 |
+
"loss": 0.009,
|
644 |
+
"step": 2575
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 6.36,
|
648 |
+
"learning_rate": 5.3400000000000005e-06,
|
649 |
+
"loss": 0.0038,
|
650 |
+
"step": 2600
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 6.42,
|
654 |
+
"learning_rate": 5.2844444444444454e-06,
|
655 |
+
"loss": 0.008,
|
656 |
+
"step": 2625
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 6.48,
|
660 |
+
"learning_rate": 5.228888888888889e-06,
|
661 |
+
"loss": 0.0035,
|
662 |
+
"step": 2650
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 6.54,
|
666 |
+
"learning_rate": 5.1733333333333335e-06,
|
667 |
+
"loss": 0.0079,
|
668 |
+
"step": 2675
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 6.6,
|
672 |
+
"learning_rate": 5.117777777777778e-06,
|
673 |
+
"loss": 0.0041,
|
674 |
+
"step": 2700
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 6.66,
|
678 |
+
"learning_rate": 5.062222222222222e-06,
|
679 |
+
"loss": 0.0077,
|
680 |
+
"step": 2725
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 6.72,
|
684 |
+
"learning_rate": 5.006666666666667e-06,
|
685 |
+
"loss": 0.004,
|
686 |
+
"step": 2750
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 6.78,
|
690 |
+
"learning_rate": 4.951111111111111e-06,
|
691 |
+
"loss": 0.0091,
|
692 |
+
"step": 2775
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 6.85,
|
696 |
+
"learning_rate": 4.895555555555556e-06,
|
697 |
+
"loss": 0.0041,
|
698 |
+
"step": 2800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 6.91,
|
702 |
+
"learning_rate": 4.84e-06,
|
703 |
+
"loss": 0.0072,
|
704 |
+
"step": 2825
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 6.97,
|
708 |
+
"learning_rate": 4.784444444444445e-06,
|
709 |
+
"loss": 0.0042,
|
710 |
+
"step": 2850
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 7.03,
|
714 |
+
"learning_rate": 4.728888888888889e-06,
|
715 |
+
"loss": 0.0053,
|
716 |
+
"step": 2875
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 7.09,
|
720 |
+
"learning_rate": 4.673333333333333e-06,
|
721 |
+
"loss": 0.0033,
|
722 |
+
"step": 2900
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 7.15,
|
726 |
+
"learning_rate": 4.617777777777778e-06,
|
727 |
+
"loss": 0.0022,
|
728 |
+
"step": 2925
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 7.21,
|
732 |
+
"learning_rate": 4.562222222222222e-06,
|
733 |
+
"loss": 0.003,
|
734 |
+
"step": 2950
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 7.27,
|
738 |
+
"learning_rate": 4.506666666666667e-06,
|
739 |
+
"loss": 0.003,
|
740 |
+
"step": 2975
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 7.33,
|
744 |
+
"learning_rate": 4.451111111111112e-06,
|
745 |
+
"loss": 0.0025,
|
746 |
+
"step": 3000
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 7.33,
|
750 |
+
"eval_loss": 0.42144426703453064,
|
751 |
+
"eval_runtime": 791.1382,
|
752 |
+
"eval_samples_per_second": 3.658,
|
753 |
+
"eval_steps_per_second": 0.229,
|
754 |
+
"eval_wer": 0.3259121307034623,
|
755 |
+
"step": 3000
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 7.4,
|
759 |
+
"learning_rate": 4.395555555555556e-06,
|
760 |
+
"loss": 0.003,
|
761 |
+
"step": 3025
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 7.46,
|
765 |
+
"learning_rate": 4.34e-06,
|
766 |
+
"loss": 0.0019,
|
767 |
+
"step": 3050
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 7.52,
|
771 |
+
"learning_rate": 4.284444444444445e-06,
|
772 |
+
"loss": 0.0027,
|
773 |
+
"step": 3075
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 7.58,
|
777 |
+
"learning_rate": 4.228888888888889e-06,
|
778 |
+
"loss": 0.0035,
|
779 |
+
"step": 3100
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 7.64,
|
783 |
+
"learning_rate": 4.173333333333334e-06,
|
784 |
+
"loss": 0.0028,
|
785 |
+
"step": 3125
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 7.7,
|
789 |
+
"learning_rate": 4.117777777777779e-06,
|
790 |
+
"loss": 0.0018,
|
791 |
+
"step": 3150
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 7.76,
|
795 |
+
"learning_rate": 4.062222222222223e-06,
|
796 |
+
"loss": 0.0029,
|
797 |
+
"step": 3175
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 7.82,
|
801 |
+
"learning_rate": 4.006666666666667e-06,
|
802 |
+
"loss": 0.003,
|
803 |
+
"step": 3200
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 7.89,
|
807 |
+
"learning_rate": 3.951111111111112e-06,
|
808 |
+
"loss": 0.0028,
|
809 |
+
"step": 3225
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 7.95,
|
813 |
+
"learning_rate": 3.895555555555556e-06,
|
814 |
+
"loss": 0.0022,
|
815 |
+
"step": 3250
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 8.01,
|
819 |
+
"learning_rate": 3.8400000000000005e-06,
|
820 |
+
"loss": 0.0015,
|
821 |
+
"step": 3275
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 8.07,
|
825 |
+
"learning_rate": 3.784444444444445e-06,
|
826 |
+
"loss": 0.0018,
|
827 |
+
"step": 3300
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 8.13,
|
831 |
+
"learning_rate": 3.728888888888889e-06,
|
832 |
+
"loss": 0.001,
|
833 |
+
"step": 3325
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 8.19,
|
837 |
+
"learning_rate": 3.673333333333334e-06,
|
838 |
+
"loss": 0.0011,
|
839 |
+
"step": 3350
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 8.25,
|
843 |
+
"learning_rate": 3.617777777777778e-06,
|
844 |
+
"loss": 0.001,
|
845 |
+
"step": 3375
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 8.31,
|
849 |
+
"learning_rate": 3.5622222222222224e-06,
|
850 |
+
"loss": 0.0016,
|
851 |
+
"step": 3400
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 8.37,
|
855 |
+
"learning_rate": 3.5066666666666673e-06,
|
856 |
+
"loss": 0.0006,
|
857 |
+
"step": 3425
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 8.44,
|
861 |
+
"learning_rate": 3.4511111111111113e-06,
|
862 |
+
"loss": 0.0017,
|
863 |
+
"step": 3450
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 8.5,
|
867 |
+
"learning_rate": 3.3955555555555558e-06,
|
868 |
+
"loss": 0.0012,
|
869 |
+
"step": 3475
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 8.56,
|
873 |
+
"learning_rate": 3.3400000000000006e-06,
|
874 |
+
"loss": 0.0012,
|
875 |
+
"step": 3500
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 8.62,
|
879 |
+
"learning_rate": 3.2844444444444447e-06,
|
880 |
+
"loss": 0.0009,
|
881 |
+
"step": 3525
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 8.68,
|
885 |
+
"learning_rate": 3.228888888888889e-06,
|
886 |
+
"loss": 0.0009,
|
887 |
+
"step": 3550
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 8.74,
|
891 |
+
"learning_rate": 3.173333333333334e-06,
|
892 |
+
"loss": 0.0006,
|
893 |
+
"step": 3575
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 8.8,
|
897 |
+
"learning_rate": 3.117777777777778e-06,
|
898 |
+
"loss": 0.0011,
|
899 |
+
"step": 3600
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 8.86,
|
903 |
+
"learning_rate": 3.0622222222222225e-06,
|
904 |
+
"loss": 0.0008,
|
905 |
+
"step": 3625
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 8.92,
|
909 |
+
"learning_rate": 3.0066666666666674e-06,
|
910 |
+
"loss": 0.0017,
|
911 |
+
"step": 3650
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 8.99,
|
915 |
+
"learning_rate": 2.9511111111111114e-06,
|
916 |
+
"loss": 0.0006,
|
917 |
+
"step": 3675
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 9.05,
|
921 |
+
"learning_rate": 2.895555555555556e-06,
|
922 |
+
"loss": 0.0008,
|
923 |
+
"step": 3700
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 9.11,
|
927 |
+
"learning_rate": 2.84e-06,
|
928 |
+
"loss": 0.0003,
|
929 |
+
"step": 3725
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 9.17,
|
933 |
+
"learning_rate": 2.784444444444445e-06,
|
934 |
+
"loss": 0.0005,
|
935 |
+
"step": 3750
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 9.23,
|
939 |
+
"learning_rate": 2.7288888888888893e-06,
|
940 |
+
"loss": 0.0003,
|
941 |
+
"step": 3775
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 9.29,
|
945 |
+
"learning_rate": 2.6733333333333333e-06,
|
946 |
+
"loss": 0.0006,
|
947 |
+
"step": 3800
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 9.35,
|
951 |
+
"learning_rate": 2.617777777777778e-06,
|
952 |
+
"loss": 0.0005,
|
953 |
+
"step": 3825
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 9.41,
|
957 |
+
"learning_rate": 2.5622222222222226e-06,
|
958 |
+
"loss": 0.0011,
|
959 |
+
"step": 3850
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 9.47,
|
963 |
+
"learning_rate": 2.5066666666666667e-06,
|
964 |
+
"loss": 0.0006,
|
965 |
+
"step": 3875
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 9.54,
|
969 |
+
"learning_rate": 2.451111111111111e-06,
|
970 |
+
"loss": 0.0004,
|
971 |
+
"step": 3900
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 9.6,
|
975 |
+
"learning_rate": 2.3955555555555556e-06,
|
976 |
+
"loss": 0.0003,
|
977 |
+
"step": 3925
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 9.66,
|
981 |
+
"learning_rate": 2.3400000000000005e-06,
|
982 |
+
"loss": 0.0008,
|
983 |
+
"step": 3950
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 9.72,
|
987 |
+
"learning_rate": 2.2844444444444445e-06,
|
988 |
+
"loss": 0.0003,
|
989 |
+
"step": 3975
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 9.78,
|
993 |
+
"learning_rate": 2.228888888888889e-06,
|
994 |
+
"loss": 0.0006,
|
995 |
+
"step": 4000
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 9.78,
|
999 |
+
"eval_loss": 0.45190635323524475,
|
1000 |
+
"eval_runtime": 793.814,
|
1001 |
+
"eval_samples_per_second": 3.646,
|
1002 |
+
"eval_steps_per_second": 0.228,
|
1003 |
+
"eval_wer": 0.3201134343519851,
|
1004 |
+
"step": 4000
|
1005 |
+
}
|
1006 |
+
],
|
1007 |
+
"max_steps": 5000,
|
1008 |
+
"num_train_epochs": 13,
|
1009 |
+
"total_flos": 1.845907654606848e+19,
|
1010 |
+
"trial_name": null,
|
1011 |
+
"trial_params": null
|
1012 |
+
}
|
checkpoint-4000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fec8c5571ff240cb312ed7e0a1c73cd7df398177a18e7338f2a77971adf3a16c
|
3 |
+
size 3503
|
checkpoint-5000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 768,
|
15 |
+
"decoder_attention_heads": 12,
|
16 |
+
"decoder_ffn_dim": 3072,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 12,
|
22 |
+
"encoder_ffn_dim": 3072,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 12,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 12,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.24.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-5000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6574ede8828ce7e8b76c56637e9a656a8d1321f768b1149e8ea124c49839bac0
|
3 |
+
size 1934158597
|
checkpoint-5000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-5000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8bb504464d8d35ad8712bb8dcff25d8f8753ff6a38dcafb9c515a6616b0051ea
|
3 |
+
size 967099139
|
checkpoint-5000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:87d6c0ed97f113f0037435c35278b1ca2235636c1192dd168b0eb7b677525465
|
3 |
+
size 14503
|
checkpoint-5000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d0bef3d8a635f779582dad8d3b271e5dfc9158126327e81d920a1cf48f2c212c
|
3 |
+
size 559
|
checkpoint-5000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e46e8af358b59b47de3f1218cb3502e031f9c6e823cca92a4f00f302a8c8d999
|
3 |
+
size 623
|
checkpoint-5000/trainer_state.json
ADDED
@@ -0,0 +1,1261 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 12.224938875305623,
|
5 |
+
"global_step": 5000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.06,
|
12 |
+
"learning_rate": 4.6000000000000004e-07,
|
13 |
+
"loss": 0.8917,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.12,
|
18 |
+
"learning_rate": 9.400000000000001e-07,
|
19 |
+
"loss": 0.8196,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.18,
|
24 |
+
"learning_rate": 1.44e-06,
|
25 |
+
"loss": 0.6476,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.24,
|
30 |
+
"learning_rate": 1.94e-06,
|
31 |
+
"loss": 0.4701,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.31,
|
36 |
+
"learning_rate": 2.4400000000000004e-06,
|
37 |
+
"loss": 0.5104,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.37,
|
42 |
+
"learning_rate": 2.9400000000000002e-06,
|
43 |
+
"loss": 0.3547,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.43,
|
48 |
+
"learning_rate": 3.44e-06,
|
49 |
+
"loss": 0.4427,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.49,
|
54 |
+
"learning_rate": 3.94e-06,
|
55 |
+
"loss": 0.3209,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.55,
|
60 |
+
"learning_rate": 4.440000000000001e-06,
|
61 |
+
"loss": 0.3846,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.61,
|
66 |
+
"learning_rate": 4.94e-06,
|
67 |
+
"loss": 0.252,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.67,
|
72 |
+
"learning_rate": 5.4400000000000004e-06,
|
73 |
+
"loss": 0.3344,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.73,
|
78 |
+
"learning_rate": 5.94e-06,
|
79 |
+
"loss": 0.2421,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.79,
|
84 |
+
"learning_rate": 6.440000000000001e-06,
|
85 |
+
"loss": 0.3204,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.86,
|
90 |
+
"learning_rate": 6.9400000000000005e-06,
|
91 |
+
"loss": 0.2263,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.92,
|
96 |
+
"learning_rate": 7.440000000000001e-06,
|
97 |
+
"loss": 0.2931,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.98,
|
102 |
+
"learning_rate": 7.94e-06,
|
103 |
+
"loss": 0.1939,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.04,
|
108 |
+
"learning_rate": 8.44e-06,
|
109 |
+
"loss": 0.2428,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.1,
|
114 |
+
"learning_rate": 8.94e-06,
|
115 |
+
"loss": 0.1915,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.16,
|
120 |
+
"learning_rate": 9.440000000000001e-06,
|
121 |
+
"loss": 0.1955,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.22,
|
126 |
+
"learning_rate": 9.940000000000001e-06,
|
127 |
+
"loss": 0.1694,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.28,
|
132 |
+
"learning_rate": 9.951111111111111e-06,
|
133 |
+
"loss": 0.1946,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.34,
|
138 |
+
"learning_rate": 9.895555555555557e-06,
|
139 |
+
"loss": 0.1883,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.41,
|
144 |
+
"learning_rate": 9.84e-06,
|
145 |
+
"loss": 0.1889,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.47,
|
150 |
+
"learning_rate": 9.784444444444445e-06,
|
151 |
+
"loss": 0.1598,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.53,
|
156 |
+
"learning_rate": 9.72888888888889e-06,
|
157 |
+
"loss": 0.1764,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.59,
|
162 |
+
"learning_rate": 9.673333333333334e-06,
|
163 |
+
"loss": 0.1652,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.65,
|
168 |
+
"learning_rate": 9.617777777777778e-06,
|
169 |
+
"loss": 0.1718,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.71,
|
174 |
+
"learning_rate": 9.562222222222223e-06,
|
175 |
+
"loss": 0.157,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.77,
|
180 |
+
"learning_rate": 9.506666666666667e-06,
|
181 |
+
"loss": 0.1685,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.83,
|
186 |
+
"learning_rate": 9.451111111111112e-06,
|
187 |
+
"loss": 0.1597,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 1.89,
|
192 |
+
"learning_rate": 9.395555555555556e-06,
|
193 |
+
"loss": 0.1641,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.96,
|
198 |
+
"learning_rate": 9.340000000000002e-06,
|
199 |
+
"loss": 0.1446,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.02,
|
204 |
+
"learning_rate": 9.284444444444444e-06,
|
205 |
+
"loss": 0.1322,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.08,
|
210 |
+
"learning_rate": 9.22888888888889e-06,
|
211 |
+
"loss": 0.0919,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.14,
|
216 |
+
"learning_rate": 9.173333333333334e-06,
|
217 |
+
"loss": 0.0799,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.2,
|
222 |
+
"learning_rate": 9.117777777777778e-06,
|
223 |
+
"loss": 0.0925,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.26,
|
228 |
+
"learning_rate": 9.062222222222224e-06,
|
229 |
+
"loss": 0.0725,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.32,
|
234 |
+
"learning_rate": 9.006666666666666e-06,
|
235 |
+
"loss": 0.1034,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.38,
|
240 |
+
"learning_rate": 8.951111111111112e-06,
|
241 |
+
"loss": 0.082,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.44,
|
246 |
+
"learning_rate": 8.895555555555556e-06,
|
247 |
+
"loss": 0.1011,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.44,
|
252 |
+
"eval_loss": 0.30747494101524353,
|
253 |
+
"eval_runtime": 799.5663,
|
254 |
+
"eval_samples_per_second": 3.619,
|
255 |
+
"eval_steps_per_second": 0.226,
|
256 |
+
"eval_wer": 0.3463133835604842,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 2.51,
|
261 |
+
"learning_rate": 8.84e-06,
|
262 |
+
"loss": 0.0762,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 2.57,
|
267 |
+
"learning_rate": 8.784444444444446e-06,
|
268 |
+
"loss": 0.1048,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 2.63,
|
273 |
+
"learning_rate": 8.72888888888889e-06,
|
274 |
+
"loss": 0.0804,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 2.69,
|
279 |
+
"learning_rate": 8.673333333333334e-06,
|
280 |
+
"loss": 0.1036,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 2.75,
|
285 |
+
"learning_rate": 8.617777777777778e-06,
|
286 |
+
"loss": 0.0789,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 2.81,
|
291 |
+
"learning_rate": 8.562222222222224e-06,
|
292 |
+
"loss": 0.0976,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 2.87,
|
297 |
+
"learning_rate": 8.506666666666668e-06,
|
298 |
+
"loss": 0.0822,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.93,
|
303 |
+
"learning_rate": 8.451111111111112e-06,
|
304 |
+
"loss": 0.0981,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 3.0,
|
309 |
+
"learning_rate": 8.395555555555557e-06,
|
310 |
+
"loss": 0.0591,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.06,
|
315 |
+
"learning_rate": 8.34e-06,
|
316 |
+
"loss": 0.0594,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 3.12,
|
321 |
+
"learning_rate": 8.284444444444446e-06,
|
322 |
+
"loss": 0.0348,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 3.18,
|
327 |
+
"learning_rate": 8.22888888888889e-06,
|
328 |
+
"loss": 0.062,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 3.24,
|
333 |
+
"learning_rate": 8.173333333333334e-06,
|
334 |
+
"loss": 0.035,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 3.3,
|
339 |
+
"learning_rate": 8.11777777777778e-06,
|
340 |
+
"loss": 0.0612,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 3.36,
|
345 |
+
"learning_rate": 8.062222222222222e-06,
|
346 |
+
"loss": 0.0328,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 3.42,
|
351 |
+
"learning_rate": 8.006666666666667e-06,
|
352 |
+
"loss": 0.0643,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 3.48,
|
357 |
+
"learning_rate": 7.951111111111111e-06,
|
358 |
+
"loss": 0.0306,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 3.55,
|
363 |
+
"learning_rate": 7.895555555555557e-06,
|
364 |
+
"loss": 0.0601,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 3.61,
|
369 |
+
"learning_rate": 7.840000000000001e-06,
|
370 |
+
"loss": 0.0304,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.67,
|
375 |
+
"learning_rate": 7.784444444444445e-06,
|
376 |
+
"loss": 0.0582,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.73,
|
381 |
+
"learning_rate": 7.72888888888889e-06,
|
382 |
+
"loss": 0.0341,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.79,
|
387 |
+
"learning_rate": 7.673333333333333e-06,
|
388 |
+
"loss": 0.0625,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.85,
|
393 |
+
"learning_rate": 7.617777777777778e-06,
|
394 |
+
"loss": 0.0365,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.91,
|
399 |
+
"learning_rate": 7.562222222222223e-06,
|
400 |
+
"loss": 0.0622,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 3.97,
|
405 |
+
"learning_rate": 7.506666666666668e-06,
|
406 |
+
"loss": 0.032,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 4.03,
|
411 |
+
"learning_rate": 7.451111111111111e-06,
|
412 |
+
"loss": 0.041,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 4.1,
|
417 |
+
"learning_rate": 7.395555555555556e-06,
|
418 |
+
"loss": 0.0217,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 4.16,
|
423 |
+
"learning_rate": 7.340000000000001e-06,
|
424 |
+
"loss": 0.0267,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 4.22,
|
429 |
+
"learning_rate": 7.284444444444445e-06,
|
430 |
+
"loss": 0.0241,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 4.28,
|
435 |
+
"learning_rate": 7.22888888888889e-06,
|
436 |
+
"loss": 0.0292,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 4.34,
|
441 |
+
"learning_rate": 7.173333333333335e-06,
|
442 |
+
"loss": 0.0204,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 4.4,
|
447 |
+
"learning_rate": 7.117777777777778e-06,
|
448 |
+
"loss": 0.0234,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 4.46,
|
453 |
+
"learning_rate": 7.062222222222223e-06,
|
454 |
+
"loss": 0.0211,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 4.52,
|
459 |
+
"learning_rate": 7.006666666666667e-06,
|
460 |
+
"loss": 0.031,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 4.58,
|
465 |
+
"learning_rate": 6.951111111111112e-06,
|
466 |
+
"loss": 0.0196,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 4.65,
|
471 |
+
"learning_rate": 6.8955555555555565e-06,
|
472 |
+
"loss": 0.0306,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 4.71,
|
477 |
+
"learning_rate": 6.8400000000000014e-06,
|
478 |
+
"loss": 0.0212,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 4.77,
|
483 |
+
"learning_rate": 6.784444444444445e-06,
|
484 |
+
"loss": 0.0271,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 4.83,
|
489 |
+
"learning_rate": 6.7288888888888895e-06,
|
490 |
+
"loss": 0.0224,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 4.89,
|
495 |
+
"learning_rate": 6.6733333333333335e-06,
|
496 |
+
"loss": 0.0264,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 4.89,
|
501 |
+
"eval_loss": 0.355797678232193,
|
502 |
+
"eval_runtime": 796.3076,
|
503 |
+
"eval_samples_per_second": 3.634,
|
504 |
+
"eval_steps_per_second": 0.227,
|
505 |
+
"eval_wer": 0.3312875645475324,
|
506 |
+
"step": 2000
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 4.95,
|
510 |
+
"learning_rate": 6.617777777777778e-06,
|
511 |
+
"loss": 0.0229,
|
512 |
+
"step": 2025
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 5.01,
|
516 |
+
"learning_rate": 6.562222222222223e-06,
|
517 |
+
"loss": 0.0245,
|
518 |
+
"step": 2050
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 5.07,
|
522 |
+
"learning_rate": 6.5066666666666665e-06,
|
523 |
+
"loss": 0.0149,
|
524 |
+
"step": 2075
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 5.13,
|
528 |
+
"learning_rate": 6.451111111111111e-06,
|
529 |
+
"loss": 0.0114,
|
530 |
+
"step": 2100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 5.2,
|
534 |
+
"learning_rate": 6.395555555555556e-06,
|
535 |
+
"loss": 0.0126,
|
536 |
+
"step": 2125
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 5.26,
|
540 |
+
"learning_rate": 6.34e-06,
|
541 |
+
"loss": 0.0093,
|
542 |
+
"step": 2150
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 5.32,
|
546 |
+
"learning_rate": 6.284444444444445e-06,
|
547 |
+
"loss": 0.014,
|
548 |
+
"step": 2175
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 5.38,
|
552 |
+
"learning_rate": 6.22888888888889e-06,
|
553 |
+
"loss": 0.0099,
|
554 |
+
"step": 2200
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 5.44,
|
558 |
+
"learning_rate": 6.173333333333333e-06,
|
559 |
+
"loss": 0.0121,
|
560 |
+
"step": 2225
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 5.5,
|
564 |
+
"learning_rate": 6.117777777777778e-06,
|
565 |
+
"loss": 0.008,
|
566 |
+
"step": 2250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 5.56,
|
570 |
+
"learning_rate": 6.062222222222223e-06,
|
571 |
+
"loss": 0.0156,
|
572 |
+
"step": 2275
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 5.62,
|
576 |
+
"learning_rate": 6.006666666666667e-06,
|
577 |
+
"loss": 0.0132,
|
578 |
+
"step": 2300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 5.68,
|
582 |
+
"learning_rate": 5.951111111111112e-06,
|
583 |
+
"loss": 0.0132,
|
584 |
+
"step": 2325
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 5.75,
|
588 |
+
"learning_rate": 5.895555555555557e-06,
|
589 |
+
"loss": 0.0093,
|
590 |
+
"step": 2350
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 5.81,
|
594 |
+
"learning_rate": 5.84e-06,
|
595 |
+
"loss": 0.0161,
|
596 |
+
"step": 2375
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 5.87,
|
600 |
+
"learning_rate": 5.784444444444445e-06,
|
601 |
+
"loss": 0.0114,
|
602 |
+
"step": 2400
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 5.93,
|
606 |
+
"learning_rate": 5.72888888888889e-06,
|
607 |
+
"loss": 0.016,
|
608 |
+
"step": 2425
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 5.99,
|
612 |
+
"learning_rate": 5.673333333333334e-06,
|
613 |
+
"loss": 0.0084,
|
614 |
+
"step": 2450
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 6.05,
|
618 |
+
"learning_rate": 5.617777777777779e-06,
|
619 |
+
"loss": 0.0114,
|
620 |
+
"step": 2475
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 6.11,
|
624 |
+
"learning_rate": 5.562222222222222e-06,
|
625 |
+
"loss": 0.0044,
|
626 |
+
"step": 2500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 6.17,
|
630 |
+
"learning_rate": 5.506666666666667e-06,
|
631 |
+
"loss": 0.007,
|
632 |
+
"step": 2525
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 6.23,
|
636 |
+
"learning_rate": 5.451111111111112e-06,
|
637 |
+
"loss": 0.0035,
|
638 |
+
"step": 2550
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 6.3,
|
642 |
+
"learning_rate": 5.3955555555555565e-06,
|
643 |
+
"loss": 0.009,
|
644 |
+
"step": 2575
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 6.36,
|
648 |
+
"learning_rate": 5.3400000000000005e-06,
|
649 |
+
"loss": 0.0038,
|
650 |
+
"step": 2600
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 6.42,
|
654 |
+
"learning_rate": 5.2844444444444454e-06,
|
655 |
+
"loss": 0.008,
|
656 |
+
"step": 2625
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 6.48,
|
660 |
+
"learning_rate": 5.228888888888889e-06,
|
661 |
+
"loss": 0.0035,
|
662 |
+
"step": 2650
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 6.54,
|
666 |
+
"learning_rate": 5.1733333333333335e-06,
|
667 |
+
"loss": 0.0079,
|
668 |
+
"step": 2675
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 6.6,
|
672 |
+
"learning_rate": 5.117777777777778e-06,
|
673 |
+
"loss": 0.0041,
|
674 |
+
"step": 2700
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 6.66,
|
678 |
+
"learning_rate": 5.062222222222222e-06,
|
679 |
+
"loss": 0.0077,
|
680 |
+
"step": 2725
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 6.72,
|
684 |
+
"learning_rate": 5.006666666666667e-06,
|
685 |
+
"loss": 0.004,
|
686 |
+
"step": 2750
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 6.78,
|
690 |
+
"learning_rate": 4.951111111111111e-06,
|
691 |
+
"loss": 0.0091,
|
692 |
+
"step": 2775
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 6.85,
|
696 |
+
"learning_rate": 4.895555555555556e-06,
|
697 |
+
"loss": 0.0041,
|
698 |
+
"step": 2800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 6.91,
|
702 |
+
"learning_rate": 4.84e-06,
|
703 |
+
"loss": 0.0072,
|
704 |
+
"step": 2825
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 6.97,
|
708 |
+
"learning_rate": 4.784444444444445e-06,
|
709 |
+
"loss": 0.0042,
|
710 |
+
"step": 2850
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 7.03,
|
714 |
+
"learning_rate": 4.728888888888889e-06,
|
715 |
+
"loss": 0.0053,
|
716 |
+
"step": 2875
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 7.09,
|
720 |
+
"learning_rate": 4.673333333333333e-06,
|
721 |
+
"loss": 0.0033,
|
722 |
+
"step": 2900
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 7.15,
|
726 |
+
"learning_rate": 4.617777777777778e-06,
|
727 |
+
"loss": 0.0022,
|
728 |
+
"step": 2925
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 7.21,
|
732 |
+
"learning_rate": 4.562222222222222e-06,
|
733 |
+
"loss": 0.003,
|
734 |
+
"step": 2950
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 7.27,
|
738 |
+
"learning_rate": 4.506666666666667e-06,
|
739 |
+
"loss": 0.003,
|
740 |
+
"step": 2975
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 7.33,
|
744 |
+
"learning_rate": 4.451111111111112e-06,
|
745 |
+
"loss": 0.0025,
|
746 |
+
"step": 3000
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 7.33,
|
750 |
+
"eval_loss": 0.42144426703453064,
|
751 |
+
"eval_runtime": 791.1382,
|
752 |
+
"eval_samples_per_second": 3.658,
|
753 |
+
"eval_steps_per_second": 0.229,
|
754 |
+
"eval_wer": 0.3259121307034623,
|
755 |
+
"step": 3000
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 7.4,
|
759 |
+
"learning_rate": 4.395555555555556e-06,
|
760 |
+
"loss": 0.003,
|
761 |
+
"step": 3025
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 7.46,
|
765 |
+
"learning_rate": 4.34e-06,
|
766 |
+
"loss": 0.0019,
|
767 |
+
"step": 3050
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 7.52,
|
771 |
+
"learning_rate": 4.284444444444445e-06,
|
772 |
+
"loss": 0.0027,
|
773 |
+
"step": 3075
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 7.58,
|
777 |
+
"learning_rate": 4.228888888888889e-06,
|
778 |
+
"loss": 0.0035,
|
779 |
+
"step": 3100
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 7.64,
|
783 |
+
"learning_rate": 4.173333333333334e-06,
|
784 |
+
"loss": 0.0028,
|
785 |
+
"step": 3125
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 7.7,
|
789 |
+
"learning_rate": 4.117777777777779e-06,
|
790 |
+
"loss": 0.0018,
|
791 |
+
"step": 3150
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 7.76,
|
795 |
+
"learning_rate": 4.062222222222223e-06,
|
796 |
+
"loss": 0.0029,
|
797 |
+
"step": 3175
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 7.82,
|
801 |
+
"learning_rate": 4.006666666666667e-06,
|
802 |
+
"loss": 0.003,
|
803 |
+
"step": 3200
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 7.89,
|
807 |
+
"learning_rate": 3.951111111111112e-06,
|
808 |
+
"loss": 0.0028,
|
809 |
+
"step": 3225
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 7.95,
|
813 |
+
"learning_rate": 3.895555555555556e-06,
|
814 |
+
"loss": 0.0022,
|
815 |
+
"step": 3250
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 8.01,
|
819 |
+
"learning_rate": 3.8400000000000005e-06,
|
820 |
+
"loss": 0.0015,
|
821 |
+
"step": 3275
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 8.07,
|
825 |
+
"learning_rate": 3.784444444444445e-06,
|
826 |
+
"loss": 0.0018,
|
827 |
+
"step": 3300
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 8.13,
|
831 |
+
"learning_rate": 3.728888888888889e-06,
|
832 |
+
"loss": 0.001,
|
833 |
+
"step": 3325
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 8.19,
|
837 |
+
"learning_rate": 3.673333333333334e-06,
|
838 |
+
"loss": 0.0011,
|
839 |
+
"step": 3350
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 8.25,
|
843 |
+
"learning_rate": 3.617777777777778e-06,
|
844 |
+
"loss": 0.001,
|
845 |
+
"step": 3375
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 8.31,
|
849 |
+
"learning_rate": 3.5622222222222224e-06,
|
850 |
+
"loss": 0.0016,
|
851 |
+
"step": 3400
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 8.37,
|
855 |
+
"learning_rate": 3.5066666666666673e-06,
|
856 |
+
"loss": 0.0006,
|
857 |
+
"step": 3425
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 8.44,
|
861 |
+
"learning_rate": 3.4511111111111113e-06,
|
862 |
+
"loss": 0.0017,
|
863 |
+
"step": 3450
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 8.5,
|
867 |
+
"learning_rate": 3.3955555555555558e-06,
|
868 |
+
"loss": 0.0012,
|
869 |
+
"step": 3475
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 8.56,
|
873 |
+
"learning_rate": 3.3400000000000006e-06,
|
874 |
+
"loss": 0.0012,
|
875 |
+
"step": 3500
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 8.62,
|
879 |
+
"learning_rate": 3.2844444444444447e-06,
|
880 |
+
"loss": 0.0009,
|
881 |
+
"step": 3525
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 8.68,
|
885 |
+
"learning_rate": 3.228888888888889e-06,
|
886 |
+
"loss": 0.0009,
|
887 |
+
"step": 3550
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 8.74,
|
891 |
+
"learning_rate": 3.173333333333334e-06,
|
892 |
+
"loss": 0.0006,
|
893 |
+
"step": 3575
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 8.8,
|
897 |
+
"learning_rate": 3.117777777777778e-06,
|
898 |
+
"loss": 0.0011,
|
899 |
+
"step": 3600
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 8.86,
|
903 |
+
"learning_rate": 3.0622222222222225e-06,
|
904 |
+
"loss": 0.0008,
|
905 |
+
"step": 3625
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 8.92,
|
909 |
+
"learning_rate": 3.0066666666666674e-06,
|
910 |
+
"loss": 0.0017,
|
911 |
+
"step": 3650
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 8.99,
|
915 |
+
"learning_rate": 2.9511111111111114e-06,
|
916 |
+
"loss": 0.0006,
|
917 |
+
"step": 3675
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 9.05,
|
921 |
+
"learning_rate": 2.895555555555556e-06,
|
922 |
+
"loss": 0.0008,
|
923 |
+
"step": 3700
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 9.11,
|
927 |
+
"learning_rate": 2.84e-06,
|
928 |
+
"loss": 0.0003,
|
929 |
+
"step": 3725
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 9.17,
|
933 |
+
"learning_rate": 2.784444444444445e-06,
|
934 |
+
"loss": 0.0005,
|
935 |
+
"step": 3750
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 9.23,
|
939 |
+
"learning_rate": 2.7288888888888893e-06,
|
940 |
+
"loss": 0.0003,
|
941 |
+
"step": 3775
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 9.29,
|
945 |
+
"learning_rate": 2.6733333333333333e-06,
|
946 |
+
"loss": 0.0006,
|
947 |
+
"step": 3800
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 9.35,
|
951 |
+
"learning_rate": 2.617777777777778e-06,
|
952 |
+
"loss": 0.0005,
|
953 |
+
"step": 3825
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 9.41,
|
957 |
+
"learning_rate": 2.5622222222222226e-06,
|
958 |
+
"loss": 0.0011,
|
959 |
+
"step": 3850
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 9.47,
|
963 |
+
"learning_rate": 2.5066666666666667e-06,
|
964 |
+
"loss": 0.0006,
|
965 |
+
"step": 3875
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 9.54,
|
969 |
+
"learning_rate": 2.451111111111111e-06,
|
970 |
+
"loss": 0.0004,
|
971 |
+
"step": 3900
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 9.6,
|
975 |
+
"learning_rate": 2.3955555555555556e-06,
|
976 |
+
"loss": 0.0003,
|
977 |
+
"step": 3925
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 9.66,
|
981 |
+
"learning_rate": 2.3400000000000005e-06,
|
982 |
+
"loss": 0.0008,
|
983 |
+
"step": 3950
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 9.72,
|
987 |
+
"learning_rate": 2.2844444444444445e-06,
|
988 |
+
"loss": 0.0003,
|
989 |
+
"step": 3975
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 9.78,
|
993 |
+
"learning_rate": 2.228888888888889e-06,
|
994 |
+
"loss": 0.0006,
|
995 |
+
"step": 4000
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 9.78,
|
999 |
+
"eval_loss": 0.45190635323524475,
|
1000 |
+
"eval_runtime": 793.814,
|
1001 |
+
"eval_samples_per_second": 3.646,
|
1002 |
+
"eval_steps_per_second": 0.228,
|
1003 |
+
"eval_wer": 0.3201134343519851,
|
1004 |
+
"step": 4000
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 9.84,
|
1008 |
+
"learning_rate": 2.1733333333333334e-06,
|
1009 |
+
"loss": 0.0003,
|
1010 |
+
"step": 4025
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 9.9,
|
1014 |
+
"learning_rate": 2.117777777777778e-06,
|
1015 |
+
"loss": 0.0006,
|
1016 |
+
"step": 4050
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 9.96,
|
1020 |
+
"learning_rate": 2.0622222222222223e-06,
|
1021 |
+
"loss": 0.0003,
|
1022 |
+
"step": 4075
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 10.02,
|
1026 |
+
"learning_rate": 2.006666666666667e-06,
|
1027 |
+
"loss": 0.0004,
|
1028 |
+
"step": 4100
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 10.09,
|
1032 |
+
"learning_rate": 1.9511111111111113e-06,
|
1033 |
+
"loss": 0.0003,
|
1034 |
+
"step": 4125
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 10.15,
|
1038 |
+
"learning_rate": 1.8955555555555557e-06,
|
1039 |
+
"loss": 0.0003,
|
1040 |
+
"step": 4150
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 10.21,
|
1044 |
+
"learning_rate": 1.8400000000000002e-06,
|
1045 |
+
"loss": 0.0003,
|
1046 |
+
"step": 4175
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 10.27,
|
1050 |
+
"learning_rate": 1.7844444444444444e-06,
|
1051 |
+
"loss": 0.0003,
|
1052 |
+
"step": 4200
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 10.33,
|
1056 |
+
"learning_rate": 1.728888888888889e-06,
|
1057 |
+
"loss": 0.0003,
|
1058 |
+
"step": 4225
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 10.39,
|
1062 |
+
"learning_rate": 1.6733333333333335e-06,
|
1063 |
+
"loss": 0.0003,
|
1064 |
+
"step": 4250
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 10.45,
|
1068 |
+
"learning_rate": 1.6177777777777778e-06,
|
1069 |
+
"loss": 0.0003,
|
1070 |
+
"step": 4275
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 10.51,
|
1074 |
+
"learning_rate": 1.5622222222222225e-06,
|
1075 |
+
"loss": 0.0003,
|
1076 |
+
"step": 4300
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 10.57,
|
1080 |
+
"learning_rate": 1.506666666666667e-06,
|
1081 |
+
"loss": 0.0003,
|
1082 |
+
"step": 4325
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 10.64,
|
1086 |
+
"learning_rate": 1.4511111111111112e-06,
|
1087 |
+
"loss": 0.0002,
|
1088 |
+
"step": 4350
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 10.7,
|
1092 |
+
"learning_rate": 1.3955555555555556e-06,
|
1093 |
+
"loss": 0.0008,
|
1094 |
+
"step": 4375
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 10.76,
|
1098 |
+
"learning_rate": 1.34e-06,
|
1099 |
+
"loss": 0.0003,
|
1100 |
+
"step": 4400
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 10.82,
|
1104 |
+
"learning_rate": 1.2844444444444445e-06,
|
1105 |
+
"loss": 0.0007,
|
1106 |
+
"step": 4425
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 10.88,
|
1110 |
+
"learning_rate": 1.228888888888889e-06,
|
1111 |
+
"loss": 0.0004,
|
1112 |
+
"step": 4450
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 10.94,
|
1116 |
+
"learning_rate": 1.1733333333333335e-06,
|
1117 |
+
"loss": 0.0003,
|
1118 |
+
"step": 4475
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 11.0,
|
1122 |
+
"learning_rate": 1.117777777777778e-06,
|
1123 |
+
"loss": 0.0002,
|
1124 |
+
"step": 4500
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 11.06,
|
1128 |
+
"learning_rate": 1.0622222222222222e-06,
|
1129 |
+
"loss": 0.0003,
|
1130 |
+
"step": 4525
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 11.12,
|
1134 |
+
"learning_rate": 1.0066666666666668e-06,
|
1135 |
+
"loss": 0.0002,
|
1136 |
+
"step": 4550
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 11.19,
|
1140 |
+
"learning_rate": 9.511111111111111e-07,
|
1141 |
+
"loss": 0.0004,
|
1142 |
+
"step": 4575
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 11.25,
|
1146 |
+
"learning_rate": 8.955555555555557e-07,
|
1147 |
+
"loss": 0.0002,
|
1148 |
+
"step": 4600
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 11.31,
|
1152 |
+
"learning_rate": 8.400000000000001e-07,
|
1153 |
+
"loss": 0.0003,
|
1154 |
+
"step": 4625
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 11.37,
|
1158 |
+
"learning_rate": 7.844444444444445e-07,
|
1159 |
+
"loss": 0.0002,
|
1160 |
+
"step": 4650
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 11.43,
|
1164 |
+
"learning_rate": 7.28888888888889e-07,
|
1165 |
+
"loss": 0.0003,
|
1166 |
+
"step": 4675
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 11.49,
|
1170 |
+
"learning_rate": 6.733333333333334e-07,
|
1171 |
+
"loss": 0.0002,
|
1172 |
+
"step": 4700
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 11.55,
|
1176 |
+
"learning_rate": 6.177777777777778e-07,
|
1177 |
+
"loss": 0.0003,
|
1178 |
+
"step": 4725
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 11.61,
|
1182 |
+
"learning_rate": 5.622222222222223e-07,
|
1183 |
+
"loss": 0.0002,
|
1184 |
+
"step": 4750
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 11.67,
|
1188 |
+
"learning_rate": 5.066666666666667e-07,
|
1189 |
+
"loss": 0.0004,
|
1190 |
+
"step": 4775
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 11.74,
|
1194 |
+
"learning_rate": 4.511111111111111e-07,
|
1195 |
+
"loss": 0.0002,
|
1196 |
+
"step": 4800
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 11.8,
|
1200 |
+
"learning_rate": 3.9555555555555557e-07,
|
1201 |
+
"loss": 0.0003,
|
1202 |
+
"step": 4825
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 11.86,
|
1206 |
+
"learning_rate": 3.4000000000000003e-07,
|
1207 |
+
"loss": 0.0001,
|
1208 |
+
"step": 4850
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 11.92,
|
1212 |
+
"learning_rate": 2.844444444444445e-07,
|
1213 |
+
"loss": 0.0003,
|
1214 |
+
"step": 4875
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 11.98,
|
1218 |
+
"learning_rate": 2.2888888888888892e-07,
|
1219 |
+
"loss": 0.0001,
|
1220 |
+
"step": 4900
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 12.04,
|
1224 |
+
"learning_rate": 1.7333333333333335e-07,
|
1225 |
+
"loss": 0.0003,
|
1226 |
+
"step": 4925
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 12.1,
|
1230 |
+
"learning_rate": 1.1777777777777778e-07,
|
1231 |
+
"loss": 0.0002,
|
1232 |
+
"step": 4950
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 12.16,
|
1236 |
+
"learning_rate": 6.222222222222223e-08,
|
1237 |
+
"loss": 0.0002,
|
1238 |
+
"step": 4975
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 12.22,
|
1242 |
+
"learning_rate": 6.666666666666667e-09,
|
1243 |
+
"loss": 0.0002,
|
1244 |
+
"step": 5000
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 12.22,
|
1248 |
+
"eval_loss": 0.4679179787635803,
|
1249 |
+
"eval_runtime": 788.4223,
|
1250 |
+
"eval_samples_per_second": 3.671,
|
1251 |
+
"eval_steps_per_second": 0.23,
|
1252 |
+
"eval_wer": 0.3209599593667993,
|
1253 |
+
"step": 5000
|
1254 |
+
}
|
1255 |
+
],
|
1256 |
+
"max_steps": 5000,
|
1257 |
+
"num_train_epochs": 13,
|
1258 |
+
"total_flos": 2.307297992638464e+19,
|
1259 |
+
"trial_name": null,
|
1260 |
+
"trial_params": null
|
1261 |
+
}
|
checkpoint-5000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fec8c5571ff240cb312ed7e0a1c73cd7df398177a18e7338f2a77971adf3a16c
|
3 |
+
size 3503
|
config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-small",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 768,
|
15 |
+
"decoder_attention_heads": 12,
|
16 |
+
"decoder_ffn_dim": 3072,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 12,
|
22 |
+
"encoder_ffn_dim": 3072,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 12,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 12,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.24.0.dev0",
|
40 |
+
"use_cache": true,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eval_loss": 0.4679179787635803,
|
3 |
+
"eval_runtime": 796.5779,
|
4 |
+
"eval_samples": 2894,
|
5 |
+
"eval_samples_per_second": 3.633,
|
6 |
+
"eval_steps_per_second": 0.227,
|
7 |
+
"eval_wer": 0.3209599593667993
|
8 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|