Commit
•
4c62af7
1
Parent(s):
bd7da12
Upload 5 files
Browse files- preprocessor_config.json +14 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +149 -0
- training_args.bin +3 -0
preprocessor_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_length": 30,
|
3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
4 |
+
"feature_size": 80,
|
5 |
+
"hop_length": 160,
|
6 |
+
"n_fft": 400,
|
7 |
+
"n_samples": 480000,
|
8 |
+
"nb_max_frames": 3000,
|
9 |
+
"padding_side": "right",
|
10 |
+
"padding_value": 0.0,
|
11 |
+
"processor_class": "WhisperProcessor",
|
12 |
+
"return_attention_mask": false,
|
13 |
+
"sampling_rate": 16000
|
14 |
+
}
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:adc7d1854cf50edd5c7f2ffc715d879a73af1faae1fd790b0cfb43f77391619e
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4972de9db3800457d03520ac8cfa919776659b10a4cce66a859f0f168b6cc103
|
3 |
+
size 1064
|
trainer_state.json
ADDED
@@ -0,0 +1,149 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 26.972262247838618,
|
3 |
+
"best_model_checkpoint": "./whisper-medium-hy-large-data\\checkpoint-3090",
|
4 |
+
"epoch": 8.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 3090,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.9993527508090615,
|
13 |
+
"grad_norm": 2.3409805297851562,
|
14 |
+
"learning_rate": 9.60099492175355e-06,
|
15 |
+
"loss": 0.0122,
|
16 |
+
"step": 386
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.9993527508090615,
|
20 |
+
"eval_loss": 0.19175899028778076,
|
21 |
+
"eval_runtime": 42548.1103,
|
22 |
+
"eval_samples_per_second": 0.101,
|
23 |
+
"eval_steps_per_second": 0.05,
|
24 |
+
"eval_wer": 28.368155619596543,
|
25 |
+
"step": 386
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"epoch": 1.998705501618123,
|
29 |
+
"grad_norm": 1.1300331354141235,
|
30 |
+
"learning_rate": 9.200953466680485e-06,
|
31 |
+
"loss": 0.0095,
|
32 |
+
"step": 772
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.998705501618123,
|
36 |
+
"eval_loss": 0.19632793962955475,
|
37 |
+
"eval_runtime": 45633.7169,
|
38 |
+
"eval_samples_per_second": 0.094,
|
39 |
+
"eval_steps_per_second": 0.047,
|
40 |
+
"eval_wer": 30.382444764649374,
|
41 |
+
"step": 772
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 2.9980582524271844,
|
45 |
+
"grad_norm": 0.9480321407318115,
|
46 |
+
"learning_rate": 8.800912011607421e-06,
|
47 |
+
"loss": 0.0076,
|
48 |
+
"step": 1158
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 2.9980582524271844,
|
52 |
+
"eval_loss": 0.19936133921146393,
|
53 |
+
"eval_runtime": 45880.7942,
|
54 |
+
"eval_samples_per_second": 0.093,
|
55 |
+
"eval_steps_per_second": 0.047,
|
56 |
+
"eval_wer": 27.785782901056677,
|
57 |
+
"step": 1158
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 4.0,
|
61 |
+
"grad_norm": 4.70139217376709,
|
62 |
+
"learning_rate": 8.399834179707742e-06,
|
63 |
+
"loss": 0.0057,
|
64 |
+
"step": 1545
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 4.0,
|
68 |
+
"eval_loss": 0.20888063311576843,
|
69 |
+
"eval_runtime": 32746.3235,
|
70 |
+
"eval_samples_per_second": 0.131,
|
71 |
+
"eval_steps_per_second": 0.065,
|
72 |
+
"eval_wer": 27.668707973102784,
|
73 |
+
"step": 1545
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 4.999352750809061,
|
77 |
+
"grad_norm": 1.9372472763061523,
|
78 |
+
"learning_rate": 7.999792724634678e-06,
|
79 |
+
"loss": 0.0046,
|
80 |
+
"step": 1931
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 4.999352750809061,
|
84 |
+
"eval_loss": 0.2178862988948822,
|
85 |
+
"eval_runtime": 46002.5276,
|
86 |
+
"eval_samples_per_second": 0.093,
|
87 |
+
"eval_steps_per_second": 0.047,
|
88 |
+
"eval_wer": 28.542267050912585,
|
89 |
+
"step": 1931
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 5.998705501618123,
|
93 |
+
"grad_norm": 1.5416231155395508,
|
94 |
+
"learning_rate": 7.599751269561612e-06,
|
95 |
+
"loss": 0.0036,
|
96 |
+
"step": 2317
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 5.998705501618123,
|
100 |
+
"eval_loss": 0.21689267456531525,
|
101 |
+
"eval_runtime": 45863.2972,
|
102 |
+
"eval_samples_per_second": 0.093,
|
103 |
+
"eval_steps_per_second": 0.047,
|
104 |
+
"eval_wer": 27.554634966378483,
|
105 |
+
"step": 2317
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 6.998058252427184,
|
109 |
+
"grad_norm": 0.7119281888008118,
|
110 |
+
"learning_rate": 7.199709814488549e-06,
|
111 |
+
"loss": 0.0028,
|
112 |
+
"step": 2703
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 6.998058252427184,
|
116 |
+
"eval_loss": 0.21707293391227722,
|
117 |
+
"eval_runtime": 45784.2835,
|
118 |
+
"eval_samples_per_second": 0.094,
|
119 |
+
"eval_steps_per_second": 0.047,
|
120 |
+
"eval_wer": 27.251440922190202,
|
121 |
+
"step": 2703
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 8.0,
|
125 |
+
"grad_norm": 0.8986366391181946,
|
126 |
+
"learning_rate": 6.798631982588871e-06,
|
127 |
+
"loss": 0.0028,
|
128 |
+
"step": 3090
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 8.0,
|
132 |
+
"eval_loss": 0.2156277447938919,
|
133 |
+
"eval_runtime": 32539.3621,
|
134 |
+
"eval_samples_per_second": 0.132,
|
135 |
+
"eval_steps_per_second": 0.066,
|
136 |
+
"eval_wer": 26.972262247838618,
|
137 |
+
"step": 3090
|
138 |
+
}
|
139 |
+
],
|
140 |
+
"logging_steps": 500,
|
141 |
+
"max_steps": 9650,
|
142 |
+
"num_input_tokens_seen": 0,
|
143 |
+
"num_train_epochs": 25,
|
144 |
+
"save_steps": 500,
|
145 |
+
"total_flos": 5.04587066277888e+19,
|
146 |
+
"train_batch_size": 4,
|
147 |
+
"trial_name": null,
|
148 |
+
"trial_params": null
|
149 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:599de08dcff27715891ec9094552ea779ebb27763652418f3c2243d7a859ded1
|
3 |
+
size 5176
|