Upload trainer_state.json
Browse files- trainer_state.json +1486 -0
trainer_state.json
ADDED
@@ -0,0 +1,1486 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 32.30769230769231,
|
5 |
+
"global_step": 2100,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.15,
|
12 |
+
"learning_rate": 9e-07,
|
13 |
+
"loss": 18.8043,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.31,
|
18 |
+
"learning_rate": 1.8e-06,
|
19 |
+
"loss": 23.3938,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.46,
|
24 |
+
"learning_rate": 2.8000000000000003e-06,
|
25 |
+
"loss": 20.8738,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.62,
|
30 |
+
"learning_rate": 3.8e-06,
|
31 |
+
"loss": 19.9257,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.77,
|
36 |
+
"learning_rate": 4.800000000000001e-06,
|
37 |
+
"loss": 22.7793,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.92,
|
42 |
+
"learning_rate": 5.8e-06,
|
43 |
+
"loss": 18.2911,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.08,
|
48 |
+
"learning_rate": 6.800000000000001e-06,
|
49 |
+
"loss": 20.4742,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 1.23,
|
54 |
+
"learning_rate": 7.7e-06,
|
55 |
+
"loss": 19.7014,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 1.38,
|
60 |
+
"learning_rate": 8.7e-06,
|
61 |
+
"loss": 15.6078,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 1.54,
|
66 |
+
"learning_rate": 9.7e-06,
|
67 |
+
"loss": 15.1693,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 1.54,
|
72 |
+
"eval_cer": 0.9953039832285115,
|
73 |
+
"eval_loss": 12.956756591796875,
|
74 |
+
"eval_runtime": 111.4724,
|
75 |
+
"eval_samples_per_second": 3.059,
|
76 |
+
"eval_steps_per_second": 0.386,
|
77 |
+
"eval_wer": 1.0,
|
78 |
+
"step": 100
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 1.69,
|
82 |
+
"learning_rate": 1.0700000000000001e-05,
|
83 |
+
"loss": 12.7027,
|
84 |
+
"step": 110
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 1.85,
|
88 |
+
"learning_rate": 1.1700000000000001e-05,
|
89 |
+
"loss": 10.9656,
|
90 |
+
"step": 120
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 2.0,
|
94 |
+
"learning_rate": 1.27e-05,
|
95 |
+
"loss": 10.6914,
|
96 |
+
"step": 130
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 2.15,
|
100 |
+
"learning_rate": 1.3700000000000001e-05,
|
101 |
+
"loss": 7.6051,
|
102 |
+
"step": 140
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 2.31,
|
106 |
+
"learning_rate": 1.47e-05,
|
107 |
+
"loss": 8.7814,
|
108 |
+
"step": 150
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 2.46,
|
112 |
+
"learning_rate": 1.5700000000000002e-05,
|
113 |
+
"loss": 7.5898,
|
114 |
+
"step": 160
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 2.62,
|
118 |
+
"learning_rate": 1.6700000000000003e-05,
|
119 |
+
"loss": 7.0232,
|
120 |
+
"step": 170
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 2.77,
|
124 |
+
"learning_rate": 1.77e-05,
|
125 |
+
"loss": 6.8873,
|
126 |
+
"step": 180
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 2.92,
|
130 |
+
"learning_rate": 1.87e-05,
|
131 |
+
"loss": 5.7658,
|
132 |
+
"step": 190
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 3.08,
|
136 |
+
"learning_rate": 1.97e-05,
|
137 |
+
"loss": 6.434,
|
138 |
+
"step": 200
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 3.08,
|
142 |
+
"eval_cer": 0.9953039832285115,
|
143 |
+
"eval_loss": 5.401930809020996,
|
144 |
+
"eval_runtime": 111.63,
|
145 |
+
"eval_samples_per_second": 3.055,
|
146 |
+
"eval_steps_per_second": 0.385,
|
147 |
+
"eval_wer": 1.0,
|
148 |
+
"step": 200
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 3.23,
|
152 |
+
"learning_rate": 2.07e-05,
|
153 |
+
"loss": 5.5544,
|
154 |
+
"step": 210
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 3.38,
|
158 |
+
"learning_rate": 2.1700000000000002e-05,
|
159 |
+
"loss": 5.2728,
|
160 |
+
"step": 220
|
161 |
+
},
|
162 |
+
{
|
163 |
+
"epoch": 3.54,
|
164 |
+
"learning_rate": 2.2700000000000003e-05,
|
165 |
+
"loss": 5.3269,
|
166 |
+
"step": 230
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 3.69,
|
170 |
+
"learning_rate": 2.37e-05,
|
171 |
+
"loss": 4.8495,
|
172 |
+
"step": 240
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 3.85,
|
176 |
+
"learning_rate": 2.47e-05,
|
177 |
+
"loss": 4.7196,
|
178 |
+
"step": 250
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 4.0,
|
182 |
+
"learning_rate": 2.57e-05,
|
183 |
+
"loss": 4.7143,
|
184 |
+
"step": 260
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 4.15,
|
188 |
+
"learning_rate": 2.6700000000000002e-05,
|
189 |
+
"loss": 4.0896,
|
190 |
+
"step": 270
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 4.31,
|
194 |
+
"learning_rate": 2.7700000000000002e-05,
|
195 |
+
"loss": 4.342,
|
196 |
+
"step": 280
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 4.46,
|
200 |
+
"learning_rate": 2.87e-05,
|
201 |
+
"loss": 3.9603,
|
202 |
+
"step": 290
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 4.62,
|
206 |
+
"learning_rate": 2.97e-05,
|
207 |
+
"loss": 3.9273,
|
208 |
+
"step": 300
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 4.62,
|
212 |
+
"eval_cer": 0.9953039832285115,
|
213 |
+
"eval_loss": 3.848849058151245,
|
214 |
+
"eval_runtime": 111.5668,
|
215 |
+
"eval_samples_per_second": 3.056,
|
216 |
+
"eval_steps_per_second": 0.385,
|
217 |
+
"eval_wer": 1.0,
|
218 |
+
"step": 300
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 4.77,
|
222 |
+
"learning_rate": 3.07e-05,
|
223 |
+
"loss": 3.9211,
|
224 |
+
"step": 310
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 4.92,
|
228 |
+
"learning_rate": 3.1700000000000005e-05,
|
229 |
+
"loss": 3.6815,
|
230 |
+
"step": 320
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 5.08,
|
234 |
+
"learning_rate": 3.27e-05,
|
235 |
+
"loss": 3.7151,
|
236 |
+
"step": 330
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 5.23,
|
240 |
+
"learning_rate": 3.3700000000000006e-05,
|
241 |
+
"loss": 3.6599,
|
242 |
+
"step": 340
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 5.38,
|
246 |
+
"learning_rate": 3.4699999999999996e-05,
|
247 |
+
"loss": 3.4484,
|
248 |
+
"step": 350
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 5.54,
|
252 |
+
"learning_rate": 3.57e-05,
|
253 |
+
"loss": 3.4587,
|
254 |
+
"step": 360
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 5.69,
|
258 |
+
"learning_rate": 3.6700000000000004e-05,
|
259 |
+
"loss": 3.3518,
|
260 |
+
"step": 370
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 5.85,
|
264 |
+
"learning_rate": 3.77e-05,
|
265 |
+
"loss": 3.3416,
|
266 |
+
"step": 380
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 6.0,
|
270 |
+
"learning_rate": 3.8700000000000006e-05,
|
271 |
+
"loss": 3.3674,
|
272 |
+
"step": 390
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 6.15,
|
276 |
+
"learning_rate": 3.97e-05,
|
277 |
+
"loss": 3.1977,
|
278 |
+
"step": 400
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 6.15,
|
282 |
+
"eval_cer": 0.9953039832285115,
|
283 |
+
"eval_loss": 3.2796881198883057,
|
284 |
+
"eval_runtime": 111.0722,
|
285 |
+
"eval_samples_per_second": 3.07,
|
286 |
+
"eval_steps_per_second": 0.387,
|
287 |
+
"eval_wer": 1.0,
|
288 |
+
"step": 400
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 6.31,
|
292 |
+
"learning_rate": 4.07e-05,
|
293 |
+
"loss": 3.3328,
|
294 |
+
"step": 410
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 6.46,
|
298 |
+
"learning_rate": 4.17e-05,
|
299 |
+
"loss": 3.2257,
|
300 |
+
"step": 420
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 6.62,
|
304 |
+
"learning_rate": 4.27e-05,
|
305 |
+
"loss": 3.2563,
|
306 |
+
"step": 430
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 6.77,
|
310 |
+
"learning_rate": 4.3700000000000005e-05,
|
311 |
+
"loss": 3.264,
|
312 |
+
"step": 440
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 6.92,
|
316 |
+
"learning_rate": 4.47e-05,
|
317 |
+
"loss": 3.166,
|
318 |
+
"step": 450
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 7.08,
|
322 |
+
"learning_rate": 4.5700000000000006e-05,
|
323 |
+
"loss": 3.223,
|
324 |
+
"step": 460
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 7.23,
|
328 |
+
"learning_rate": 4.6700000000000003e-05,
|
329 |
+
"loss": 3.1885,
|
330 |
+
"step": 470
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 7.38,
|
334 |
+
"learning_rate": 4.77e-05,
|
335 |
+
"loss": 3.182,
|
336 |
+
"step": 480
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 7.54,
|
340 |
+
"learning_rate": 4.87e-05,
|
341 |
+
"loss": 3.2332,
|
342 |
+
"step": 490
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 7.69,
|
346 |
+
"learning_rate": 4.97e-05,
|
347 |
+
"loss": 3.1592,
|
348 |
+
"step": 500
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 7.69,
|
352 |
+
"eval_cer": 0.9953039832285115,
|
353 |
+
"eval_loss": 3.201068162918091,
|
354 |
+
"eval_runtime": 111.8171,
|
355 |
+
"eval_samples_per_second": 3.05,
|
356 |
+
"eval_steps_per_second": 0.385,
|
357 |
+
"eval_wer": 1.0,
|
358 |
+
"step": 500
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 7.85,
|
362 |
+
"learning_rate": 5.0700000000000006e-05,
|
363 |
+
"loss": 3.1996,
|
364 |
+
"step": 510
|
365 |
+
},
|
366 |
+
{
|
367 |
+
"epoch": 8.0,
|
368 |
+
"learning_rate": 5.17e-05,
|
369 |
+
"loss": 3.2977,
|
370 |
+
"step": 520
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 8.15,
|
374 |
+
"learning_rate": 5.270000000000001e-05,
|
375 |
+
"loss": 3.1425,
|
376 |
+
"step": 530
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 8.31,
|
380 |
+
"learning_rate": 5.3700000000000004e-05,
|
381 |
+
"loss": 3.2105,
|
382 |
+
"step": 540
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 8.46,
|
386 |
+
"learning_rate": 5.470000000000001e-05,
|
387 |
+
"loss": 3.1593,
|
388 |
+
"step": 550
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 8.62,
|
392 |
+
"learning_rate": 5.5700000000000005e-05,
|
393 |
+
"loss": 3.1665,
|
394 |
+
"step": 560
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 8.77,
|
398 |
+
"learning_rate": 5.6699999999999996e-05,
|
399 |
+
"loss": 3.2244,
|
400 |
+
"step": 570
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 8.92,
|
404 |
+
"learning_rate": 5.77e-05,
|
405 |
+
"loss": 3.1578,
|
406 |
+
"step": 580
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 9.08,
|
410 |
+
"learning_rate": 5.87e-05,
|
411 |
+
"loss": 3.1924,
|
412 |
+
"step": 590
|
413 |
+
},
|
414 |
+
{
|
415 |
+
"epoch": 9.23,
|
416 |
+
"learning_rate": 5.97e-05,
|
417 |
+
"loss": 3.1667,
|
418 |
+
"step": 600
|
419 |
+
},
|
420 |
+
{
|
421 |
+
"epoch": 9.23,
|
422 |
+
"eval_cer": 0.9953039832285115,
|
423 |
+
"eval_loss": 3.139559030532837,
|
424 |
+
"eval_runtime": 109.7588,
|
425 |
+
"eval_samples_per_second": 3.107,
|
426 |
+
"eval_steps_per_second": 0.392,
|
427 |
+
"eval_wer": 1.0,
|
428 |
+
"step": 600
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 9.38,
|
432 |
+
"learning_rate": 6.07e-05,
|
433 |
+
"loss": 3.1127,
|
434 |
+
"step": 610
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 9.54,
|
438 |
+
"learning_rate": 6.170000000000001e-05,
|
439 |
+
"loss": 3.1775,
|
440 |
+
"step": 620
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 9.69,
|
444 |
+
"learning_rate": 6.27e-05,
|
445 |
+
"loss": 3.1132,
|
446 |
+
"step": 630
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 9.85,
|
450 |
+
"learning_rate": 6.37e-05,
|
451 |
+
"loss": 3.1704,
|
452 |
+
"step": 640
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 10.0,
|
456 |
+
"learning_rate": 6.47e-05,
|
457 |
+
"loss": 3.1306,
|
458 |
+
"step": 650
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 10.15,
|
462 |
+
"learning_rate": 6.570000000000001e-05,
|
463 |
+
"loss": 3.0967,
|
464 |
+
"step": 660
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 10.31,
|
468 |
+
"learning_rate": 6.670000000000001e-05,
|
469 |
+
"loss": 3.1375,
|
470 |
+
"step": 670
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 10.46,
|
474 |
+
"learning_rate": 6.77e-05,
|
475 |
+
"loss": 3.1139,
|
476 |
+
"step": 680
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 10.62,
|
480 |
+
"learning_rate": 6.87e-05,
|
481 |
+
"loss": 3.098,
|
482 |
+
"step": 690
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 10.77,
|
486 |
+
"learning_rate": 6.97e-05,
|
487 |
+
"loss": 3.1111,
|
488 |
+
"step": 700
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 10.77,
|
492 |
+
"eval_cer": 0.9953039832285115,
|
493 |
+
"eval_loss": 3.0903170108795166,
|
494 |
+
"eval_runtime": 109.1662,
|
495 |
+
"eval_samples_per_second": 3.124,
|
496 |
+
"eval_steps_per_second": 0.394,
|
497 |
+
"eval_wer": 1.0,
|
498 |
+
"step": 700
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 10.92,
|
502 |
+
"learning_rate": 7.07e-05,
|
503 |
+
"loss": 3.0775,
|
504 |
+
"step": 710
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 11.08,
|
508 |
+
"learning_rate": 7.17e-05,
|
509 |
+
"loss": 3.1294,
|
510 |
+
"step": 720
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 11.23,
|
514 |
+
"learning_rate": 7.27e-05,
|
515 |
+
"loss": 3.0943,
|
516 |
+
"step": 730
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 11.38,
|
520 |
+
"learning_rate": 7.37e-05,
|
521 |
+
"loss": 3.0812,
|
522 |
+
"step": 740
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 11.54,
|
526 |
+
"learning_rate": 7.47e-05,
|
527 |
+
"loss": 3.0843,
|
528 |
+
"step": 750
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 11.69,
|
532 |
+
"learning_rate": 7.570000000000001e-05,
|
533 |
+
"loss": 3.0668,
|
534 |
+
"step": 760
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 11.85,
|
538 |
+
"learning_rate": 7.670000000000001e-05,
|
539 |
+
"loss": 3.0617,
|
540 |
+
"step": 770
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 12.0,
|
544 |
+
"learning_rate": 7.77e-05,
|
545 |
+
"loss": 3.0469,
|
546 |
+
"step": 780
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 12.15,
|
550 |
+
"learning_rate": 7.87e-05,
|
551 |
+
"loss": 2.9867,
|
552 |
+
"step": 790
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 12.31,
|
556 |
+
"learning_rate": 7.970000000000001e-05,
|
557 |
+
"loss": 2.9955,
|
558 |
+
"step": 800
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 12.31,
|
562 |
+
"eval_cer": 0.9953039832285115,
|
563 |
+
"eval_loss": 2.895956039428711,
|
564 |
+
"eval_runtime": 109.2356,
|
565 |
+
"eval_samples_per_second": 3.122,
|
566 |
+
"eval_steps_per_second": 0.394,
|
567 |
+
"eval_wer": 1.0,
|
568 |
+
"step": 800
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 12.46,
|
572 |
+
"learning_rate": 8.070000000000001e-05,
|
573 |
+
"loss": 2.9201,
|
574 |
+
"step": 810
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 12.62,
|
578 |
+
"learning_rate": 8.17e-05,
|
579 |
+
"loss": 2.8683,
|
580 |
+
"step": 820
|
581 |
+
},
|
582 |
+
{
|
583 |
+
"epoch": 12.77,
|
584 |
+
"learning_rate": 8.27e-05,
|
585 |
+
"loss": 2.8432,
|
586 |
+
"step": 830
|
587 |
+
},
|
588 |
+
{
|
589 |
+
"epoch": 12.92,
|
590 |
+
"learning_rate": 8.37e-05,
|
591 |
+
"loss": 2.6776,
|
592 |
+
"step": 840
|
593 |
+
},
|
594 |
+
{
|
595 |
+
"epoch": 13.08,
|
596 |
+
"learning_rate": 8.47e-05,
|
597 |
+
"loss": 2.5611,
|
598 |
+
"step": 850
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 13.23,
|
602 |
+
"learning_rate": 8.57e-05,
|
603 |
+
"loss": 2.4031,
|
604 |
+
"step": 860
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 13.38,
|
608 |
+
"learning_rate": 8.67e-05,
|
609 |
+
"loss": 2.2939,
|
610 |
+
"step": 870
|
611 |
+
},
|
612 |
+
{
|
613 |
+
"epoch": 13.54,
|
614 |
+
"learning_rate": 8.77e-05,
|
615 |
+
"loss": 2.0835,
|
616 |
+
"step": 880
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 13.69,
|
620 |
+
"learning_rate": 8.87e-05,
|
621 |
+
"loss": 1.8785,
|
622 |
+
"step": 890
|
623 |
+
},
|
624 |
+
{
|
625 |
+
"epoch": 13.85,
|
626 |
+
"learning_rate": 8.970000000000001e-05,
|
627 |
+
"loss": 1.784,
|
628 |
+
"step": 900
|
629 |
+
},
|
630 |
+
{
|
631 |
+
"epoch": 13.85,
|
632 |
+
"eval_cer": 0.33366876310272536,
|
633 |
+
"eval_loss": 1.4985727071762085,
|
634 |
+
"eval_runtime": 109.5182,
|
635 |
+
"eval_samples_per_second": 3.114,
|
636 |
+
"eval_steps_per_second": 0.393,
|
637 |
+
"eval_wer": 0.8279052553663953,
|
638 |
+
"step": 900
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 14.0,
|
642 |
+
"learning_rate": 9.070000000000001e-05,
|
643 |
+
"loss": 1.6083,
|
644 |
+
"step": 910
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 14.15,
|
648 |
+
"learning_rate": 9.17e-05,
|
649 |
+
"loss": 1.5164,
|
650 |
+
"step": 920
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 14.31,
|
654 |
+
"learning_rate": 9.27e-05,
|
655 |
+
"loss": 1.4019,
|
656 |
+
"step": 930
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 14.46,
|
660 |
+
"learning_rate": 9.370000000000001e-05,
|
661 |
+
"loss": 1.3325,
|
662 |
+
"step": 940
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 14.62,
|
666 |
+
"learning_rate": 9.47e-05,
|
667 |
+
"loss": 1.3251,
|
668 |
+
"step": 950
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 14.77,
|
672 |
+
"learning_rate": 9.57e-05,
|
673 |
+
"loss": 1.2811,
|
674 |
+
"step": 960
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 14.92,
|
678 |
+
"learning_rate": 9.67e-05,
|
679 |
+
"loss": 1.1927,
|
680 |
+
"step": 970
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 15.08,
|
684 |
+
"learning_rate": 9.77e-05,
|
685 |
+
"loss": 1.2017,
|
686 |
+
"step": 980
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 15.23,
|
690 |
+
"learning_rate": 9.87e-05,
|
691 |
+
"loss": 1.0452,
|
692 |
+
"step": 990
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 15.38,
|
696 |
+
"learning_rate": 9.970000000000001e-05,
|
697 |
+
"loss": 1.1511,
|
698 |
+
"step": 1000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 15.38,
|
702 |
+
"eval_cer": 0.21970649895178196,
|
703 |
+
"eval_loss": 0.8053019046783447,
|
704 |
+
"eval_runtime": 109.46,
|
705 |
+
"eval_samples_per_second": 3.115,
|
706 |
+
"eval_steps_per_second": 0.393,
|
707 |
+
"eval_wer": 0.6402664692820134,
|
708 |
+
"step": 1000
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 15.54,
|
712 |
+
"learning_rate": 9.994166666666667e-05,
|
713 |
+
"loss": 5.2926,
|
714 |
+
"step": 1010
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 15.69,
|
718 |
+
"learning_rate": 9.985833333333334e-05,
|
719 |
+
"loss": 3.4895,
|
720 |
+
"step": 1020
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 15.85,
|
724 |
+
"learning_rate": 9.977500000000001e-05,
|
725 |
+
"loss": 3.3029,
|
726 |
+
"step": 1030
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 16.0,
|
730 |
+
"learning_rate": 9.969166666666667e-05,
|
731 |
+
"loss": 3.2695,
|
732 |
+
"step": 1040
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 16.15,
|
736 |
+
"learning_rate": 9.960833333333333e-05,
|
737 |
+
"loss": 3.1054,
|
738 |
+
"step": 1050
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 16.31,
|
742 |
+
"learning_rate": 9.952500000000001e-05,
|
743 |
+
"loss": 3.0923,
|
744 |
+
"step": 1060
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 16.46,
|
748 |
+
"learning_rate": 9.944166666666667e-05,
|
749 |
+
"loss": 2.9955,
|
750 |
+
"step": 1070
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 16.62,
|
754 |
+
"learning_rate": 9.935833333333334e-05,
|
755 |
+
"loss": 2.9343,
|
756 |
+
"step": 1080
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 16.77,
|
760 |
+
"learning_rate": 9.9275e-05,
|
761 |
+
"loss": 2.7871,
|
762 |
+
"step": 1090
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 16.92,
|
766 |
+
"learning_rate": 9.919166666666667e-05,
|
767 |
+
"loss": 2.3674,
|
768 |
+
"step": 1100
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 16.92,
|
772 |
+
"eval_cer": 0.4807547169811321,
|
773 |
+
"eval_loss": 1.9613749980926514,
|
774 |
+
"eval_runtime": 26.9012,
|
775 |
+
"eval_samples_per_second": 12.676,
|
776 |
+
"eval_steps_per_second": 1.598,
|
777 |
+
"eval_wer": 0.9925980754996299,
|
778 |
+
"step": 1100
|
779 |
+
},
|
780 |
+
{
|
781 |
+
"epoch": 17.08,
|
782 |
+
"learning_rate": 9.910833333333333e-05,
|
783 |
+
"loss": 2.1319,
|
784 |
+
"step": 1110
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 17.23,
|
788 |
+
"learning_rate": 9.9025e-05,
|
789 |
+
"loss": 1.6758,
|
790 |
+
"step": 1120
|
791 |
+
},
|
792 |
+
{
|
793 |
+
"epoch": 17.38,
|
794 |
+
"learning_rate": 9.894166666666668e-05,
|
795 |
+
"loss": 1.5306,
|
796 |
+
"step": 1130
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 17.54,
|
800 |
+
"learning_rate": 9.885833333333334e-05,
|
801 |
+
"loss": 1.3654,
|
802 |
+
"step": 1140
|
803 |
+
},
|
804 |
+
{
|
805 |
+
"epoch": 17.69,
|
806 |
+
"learning_rate": 9.8775e-05,
|
807 |
+
"loss": 1.2197,
|
808 |
+
"step": 1150
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 17.85,
|
812 |
+
"learning_rate": 9.869166666666668e-05,
|
813 |
+
"loss": 1.1739,
|
814 |
+
"step": 1160
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 18.0,
|
818 |
+
"learning_rate": 9.860833333333334e-05,
|
819 |
+
"loss": 1.0703,
|
820 |
+
"step": 1170
|
821 |
+
},
|
822 |
+
{
|
823 |
+
"epoch": 18.15,
|
824 |
+
"learning_rate": 9.8525e-05,
|
825 |
+
"loss": 1.064,
|
826 |
+
"step": 1180
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 18.31,
|
830 |
+
"learning_rate": 9.844166666666667e-05,
|
831 |
+
"loss": 1.0071,
|
832 |
+
"step": 1190
|
833 |
+
},
|
834 |
+
{
|
835 |
+
"epoch": 18.46,
|
836 |
+
"learning_rate": 9.835833333333334e-05,
|
837 |
+
"loss": 0.8716,
|
838 |
+
"step": 1200
|
839 |
+
},
|
840 |
+
{
|
841 |
+
"epoch": 18.46,
|
842 |
+
"eval_cer": 0.20771488469601676,
|
843 |
+
"eval_loss": 0.7716531753540039,
|
844 |
+
"eval_runtime": 26.7363,
|
845 |
+
"eval_samples_per_second": 12.754,
|
846 |
+
"eval_steps_per_second": 1.608,
|
847 |
+
"eval_wer": 0.609178386380459,
|
848 |
+
"step": 1200
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 18.62,
|
852 |
+
"learning_rate": 9.8275e-05,
|
853 |
+
"loss": 0.9567,
|
854 |
+
"step": 1210
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 18.77,
|
858 |
+
"learning_rate": 9.819166666666668e-05,
|
859 |
+
"loss": 0.9223,
|
860 |
+
"step": 1220
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 18.92,
|
864 |
+
"learning_rate": 9.810833333333334e-05,
|
865 |
+
"loss": 0.8467,
|
866 |
+
"step": 1230
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 19.08,
|
870 |
+
"learning_rate": 9.8025e-05,
|
871 |
+
"loss": 0.8997,
|
872 |
+
"step": 1240
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 19.23,
|
876 |
+
"learning_rate": 9.794166666666667e-05,
|
877 |
+
"loss": 0.7379,
|
878 |
+
"step": 1250
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 19.38,
|
882 |
+
"learning_rate": 9.785833333333334e-05,
|
883 |
+
"loss": 0.7997,
|
884 |
+
"step": 1260
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 19.54,
|
888 |
+
"learning_rate": 9.7775e-05,
|
889 |
+
"loss": 0.8364,
|
890 |
+
"step": 1270
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 19.69,
|
894 |
+
"learning_rate": 9.769166666666667e-05,
|
895 |
+
"loss": 0.6978,
|
896 |
+
"step": 1280
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 19.85,
|
900 |
+
"learning_rate": 9.760833333333334e-05,
|
901 |
+
"loss": 0.7572,
|
902 |
+
"step": 1290
|
903 |
+
},
|
904 |
+
{
|
905 |
+
"epoch": 20.0,
|
906 |
+
"learning_rate": 9.7525e-05,
|
907 |
+
"loss": 0.7108,
|
908 |
+
"step": 1300
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 20.0,
|
912 |
+
"eval_cer": 0.16545073375262054,
|
913 |
+
"eval_loss": 0.5405445098876953,
|
914 |
+
"eval_runtime": 26.8237,
|
915 |
+
"eval_samples_per_second": 12.713,
|
916 |
+
"eval_steps_per_second": 1.603,
|
917 |
+
"eval_wer": 0.5159141376757957,
|
918 |
+
"step": 1300
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 20.15,
|
922 |
+
"learning_rate": 9.744166666666667e-05,
|
923 |
+
"loss": 0.7364,
|
924 |
+
"step": 1310
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 20.31,
|
928 |
+
"learning_rate": 9.735833333333335e-05,
|
929 |
+
"loss": 0.6907,
|
930 |
+
"step": 1320
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 20.46,
|
934 |
+
"learning_rate": 9.7275e-05,
|
935 |
+
"loss": 0.6201,
|
936 |
+
"step": 1330
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 20.62,
|
940 |
+
"learning_rate": 9.719166666666667e-05,
|
941 |
+
"loss": 0.6772,
|
942 |
+
"step": 1340
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 20.77,
|
946 |
+
"learning_rate": 9.710833333333334e-05,
|
947 |
+
"loss": 0.6928,
|
948 |
+
"step": 1350
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 20.92,
|
952 |
+
"learning_rate": 9.7025e-05,
|
953 |
+
"loss": 0.6312,
|
954 |
+
"step": 1360
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 21.08,
|
958 |
+
"learning_rate": 9.694166666666667e-05,
|
959 |
+
"loss": 0.6452,
|
960 |
+
"step": 1370
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 21.23,
|
964 |
+
"learning_rate": 9.685833333333333e-05,
|
965 |
+
"loss": 0.551,
|
966 |
+
"step": 1380
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 21.38,
|
970 |
+
"learning_rate": 9.677500000000001e-05,
|
971 |
+
"loss": 0.6457,
|
972 |
+
"step": 1390
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 21.54,
|
976 |
+
"learning_rate": 9.669166666666667e-05,
|
977 |
+
"loss": 0.612,
|
978 |
+
"step": 1400
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 21.54,
|
982 |
+
"eval_cer": 0.140041928721174,
|
983 |
+
"eval_loss": 0.42258089780807495,
|
984 |
+
"eval_runtime": 27.0022,
|
985 |
+
"eval_samples_per_second": 12.629,
|
986 |
+
"eval_steps_per_second": 1.592,
|
987 |
+
"eval_wer": 0.463360473723168,
|
988 |
+
"step": 1400
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 21.69,
|
992 |
+
"learning_rate": 9.660833333333333e-05,
|
993 |
+
"loss": 0.5377,
|
994 |
+
"step": 1410
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 21.85,
|
998 |
+
"learning_rate": 9.652500000000002e-05,
|
999 |
+
"loss": 0.625,
|
1000 |
+
"step": 1420
|
1001 |
+
},
|
1002 |
+
{
|
1003 |
+
"epoch": 22.0,
|
1004 |
+
"learning_rate": 9.644166666666668e-05,
|
1005 |
+
"loss": 0.5206,
|
1006 |
+
"step": 1430
|
1007 |
+
},
|
1008 |
+
{
|
1009 |
+
"epoch": 22.15,
|
1010 |
+
"learning_rate": 9.635833333333334e-05,
|
1011 |
+
"loss": 0.6004,
|
1012 |
+
"step": 1440
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 22.31,
|
1016 |
+
"learning_rate": 9.627500000000001e-05,
|
1017 |
+
"loss": 0.5728,
|
1018 |
+
"step": 1450
|
1019 |
+
},
|
1020 |
+
{
|
1021 |
+
"epoch": 22.46,
|
1022 |
+
"learning_rate": 9.619166666666667e-05,
|
1023 |
+
"loss": 0.5696,
|
1024 |
+
"step": 1460
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 22.62,
|
1028 |
+
"learning_rate": 9.610833333333333e-05,
|
1029 |
+
"loss": 0.5712,
|
1030 |
+
"step": 1470
|
1031 |
+
},
|
1032 |
+
{
|
1033 |
+
"epoch": 22.77,
|
1034 |
+
"learning_rate": 9.6025e-05,
|
1035 |
+
"loss": 0.5303,
|
1036 |
+
"step": 1480
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 22.92,
|
1040 |
+
"learning_rate": 9.594166666666668e-05,
|
1041 |
+
"loss": 0.4954,
|
1042 |
+
"step": 1490
|
1043 |
+
},
|
1044 |
+
{
|
1045 |
+
"epoch": 23.08,
|
1046 |
+
"learning_rate": 9.585833333333334e-05,
|
1047 |
+
"loss": 0.4919,
|
1048 |
+
"step": 1500
|
1049 |
+
},
|
1050 |
+
{
|
1051 |
+
"epoch": 23.08,
|
1052 |
+
"eval_cer": 0.11941299790356394,
|
1053 |
+
"eval_loss": 0.3299271762371063,
|
1054 |
+
"eval_runtime": 27.0051,
|
1055 |
+
"eval_samples_per_second": 12.627,
|
1056 |
+
"eval_steps_per_second": 1.592,
|
1057 |
+
"eval_wer": 0.4052553663952628,
|
1058 |
+
"step": 1500
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 23.23,
|
1062 |
+
"learning_rate": 9.5775e-05,
|
1063 |
+
"loss": 0.391,
|
1064 |
+
"step": 1510
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 23.38,
|
1068 |
+
"learning_rate": 9.569166666666667e-05,
|
1069 |
+
"loss": 0.5205,
|
1070 |
+
"step": 1520
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 23.54,
|
1074 |
+
"learning_rate": 9.560833333333333e-05,
|
1075 |
+
"loss": 0.5302,
|
1076 |
+
"step": 1530
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 23.69,
|
1080 |
+
"learning_rate": 9.5525e-05,
|
1081 |
+
"loss": 0.4806,
|
1082 |
+
"step": 1540
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 23.85,
|
1086 |
+
"learning_rate": 9.544166666666668e-05,
|
1087 |
+
"loss": 0.5882,
|
1088 |
+
"step": 1550
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 24.0,
|
1092 |
+
"learning_rate": 9.535833333333334e-05,
|
1093 |
+
"loss": 0.4541,
|
1094 |
+
"step": 1560
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 24.15,
|
1098 |
+
"learning_rate": 9.5275e-05,
|
1099 |
+
"loss": 0.457,
|
1100 |
+
"step": 1570
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 24.31,
|
1104 |
+
"learning_rate": 9.519166666666667e-05,
|
1105 |
+
"loss": 0.4797,
|
1106 |
+
"step": 1580
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 24.46,
|
1110 |
+
"learning_rate": 9.510833333333333e-05,
|
1111 |
+
"loss": 0.3973,
|
1112 |
+
"step": 1590
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 24.62,
|
1116 |
+
"learning_rate": 9.5025e-05,
|
1117 |
+
"loss": 0.4999,
|
1118 |
+
"step": 1600
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 24.62,
|
1122 |
+
"eval_cer": 0.1020545073375262,
|
1123 |
+
"eval_loss": 0.2569698691368103,
|
1124 |
+
"eval_runtime": 30.0585,
|
1125 |
+
"eval_samples_per_second": 11.345,
|
1126 |
+
"eval_steps_per_second": 1.431,
|
1127 |
+
"eval_wer": 0.3604737231680237,
|
1128 |
+
"step": 1600
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 24.77,
|
1132 |
+
"learning_rate": 9.494166666666668e-05,
|
1133 |
+
"loss": 0.452,
|
1134 |
+
"step": 1610
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 24.92,
|
1138 |
+
"learning_rate": 9.485833333333334e-05,
|
1139 |
+
"loss": 0.4148,
|
1140 |
+
"step": 1620
|
1141 |
+
},
|
1142 |
+
{
|
1143 |
+
"epoch": 25.08,
|
1144 |
+
"learning_rate": 9.4775e-05,
|
1145 |
+
"loss": 0.4316,
|
1146 |
+
"step": 1630
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 25.23,
|
1150 |
+
"learning_rate": 9.469166666666667e-05,
|
1151 |
+
"loss": 0.3673,
|
1152 |
+
"step": 1640
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 25.38,
|
1156 |
+
"learning_rate": 9.460833333333335e-05,
|
1157 |
+
"loss": 0.4365,
|
1158 |
+
"step": 1650
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 25.54,
|
1162 |
+
"learning_rate": 9.452500000000001e-05,
|
1163 |
+
"loss": 0.4191,
|
1164 |
+
"step": 1660
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 25.69,
|
1168 |
+
"learning_rate": 9.444166666666667e-05,
|
1169 |
+
"loss": 0.3857,
|
1170 |
+
"step": 1670
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 25.85,
|
1174 |
+
"learning_rate": 9.435833333333334e-05,
|
1175 |
+
"loss": 0.4627,
|
1176 |
+
"step": 1680
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 26.0,
|
1180 |
+
"learning_rate": 9.4275e-05,
|
1181 |
+
"loss": 0.3845,
|
1182 |
+
"step": 1690
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 26.15,
|
1186 |
+
"learning_rate": 9.419166666666666e-05,
|
1187 |
+
"loss": 0.4349,
|
1188 |
+
"step": 1700
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 26.15,
|
1192 |
+
"eval_cer": 0.09207547169811321,
|
1193 |
+
"eval_loss": 0.21595901250839233,
|
1194 |
+
"eval_runtime": 27.0141,
|
1195 |
+
"eval_samples_per_second": 12.623,
|
1196 |
+
"eval_steps_per_second": 1.592,
|
1197 |
+
"eval_wer": 0.33826794966691337,
|
1198 |
+
"step": 1700
|
1199 |
+
},
|
1200 |
+
{
|
1201 |
+
"epoch": 26.31,
|
1202 |
+
"learning_rate": 9.410833333333335e-05,
|
1203 |
+
"loss": 0.359,
|
1204 |
+
"step": 1710
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 26.46,
|
1208 |
+
"learning_rate": 9.402500000000001e-05,
|
1209 |
+
"loss": 0.3897,
|
1210 |
+
"step": 1720
|
1211 |
+
},
|
1212 |
+
{
|
1213 |
+
"epoch": 26.62,
|
1214 |
+
"learning_rate": 9.394166666666667e-05,
|
1215 |
+
"loss": 0.3857,
|
1216 |
+
"step": 1730
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 26.77,
|
1220 |
+
"learning_rate": 9.385833333333334e-05,
|
1221 |
+
"loss": 0.361,
|
1222 |
+
"step": 1740
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 26.92,
|
1226 |
+
"learning_rate": 9.3775e-05,
|
1227 |
+
"loss": 0.4116,
|
1228 |
+
"step": 1750
|
1229 |
+
},
|
1230 |
+
{
|
1231 |
+
"epoch": 27.08,
|
1232 |
+
"learning_rate": 9.369166666666666e-05,
|
1233 |
+
"loss": 0.427,
|
1234 |
+
"step": 1760
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 27.23,
|
1238 |
+
"learning_rate": 9.360833333333334e-05,
|
1239 |
+
"loss": 0.2724,
|
1240 |
+
"step": 1770
|
1241 |
+
},
|
1242 |
+
{
|
1243 |
+
"epoch": 27.38,
|
1244 |
+
"learning_rate": 9.352500000000001e-05,
|
1245 |
+
"loss": 0.416,
|
1246 |
+
"step": 1780
|
1247 |
+
},
|
1248 |
+
{
|
1249 |
+
"epoch": 27.54,
|
1250 |
+
"learning_rate": 9.344166666666667e-05,
|
1251 |
+
"loss": 0.3837,
|
1252 |
+
"step": 1790
|
1253 |
+
},
|
1254 |
+
{
|
1255 |
+
"epoch": 27.69,
|
1256 |
+
"learning_rate": 9.335833333333333e-05,
|
1257 |
+
"loss": 0.3205,
|
1258 |
+
"step": 1800
|
1259 |
+
},
|
1260 |
+
{
|
1261 |
+
"epoch": 27.69,
|
1262 |
+
"eval_cer": 0.07907756813417191,
|
1263 |
+
"eval_loss": 0.17836497724056244,
|
1264 |
+
"eval_runtime": 26.9274,
|
1265 |
+
"eval_samples_per_second": 12.664,
|
1266 |
+
"eval_steps_per_second": 1.597,
|
1267 |
+
"eval_wer": 0.29533678756476683,
|
1268 |
+
"step": 1800
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 27.85,
|
1272 |
+
"learning_rate": 9.3275e-05,
|
1273 |
+
"loss": 0.3578,
|
1274 |
+
"step": 1810
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 28.0,
|
1278 |
+
"learning_rate": 9.319166666666666e-05,
|
1279 |
+
"loss": 0.3527,
|
1280 |
+
"step": 1820
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 28.15,
|
1284 |
+
"learning_rate": 9.310833333333334e-05,
|
1285 |
+
"loss": 0.3458,
|
1286 |
+
"step": 1830
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 28.31,
|
1290 |
+
"learning_rate": 9.302500000000001e-05,
|
1291 |
+
"loss": 0.3527,
|
1292 |
+
"step": 1840
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 28.46,
|
1296 |
+
"learning_rate": 9.294166666666667e-05,
|
1297 |
+
"loss": 0.3244,
|
1298 |
+
"step": 1850
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 28.62,
|
1302 |
+
"learning_rate": 9.285833333333333e-05,
|
1303 |
+
"loss": 0.393,
|
1304 |
+
"step": 1860
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 28.77,
|
1308 |
+
"learning_rate": 9.2775e-05,
|
1309 |
+
"loss": 0.3255,
|
1310 |
+
"step": 1870
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 28.92,
|
1314 |
+
"learning_rate": 9.269166666666668e-05,
|
1315 |
+
"loss": 0.2884,
|
1316 |
+
"step": 1880
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 29.08,
|
1320 |
+
"learning_rate": 9.260833333333334e-05,
|
1321 |
+
"loss": 0.3382,
|
1322 |
+
"step": 1890
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 29.23,
|
1326 |
+
"learning_rate": 9.252500000000001e-05,
|
1327 |
+
"loss": 0.2717,
|
1328 |
+
"step": 1900
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 29.23,
|
1332 |
+
"eval_cer": 0.07127882599580712,
|
1333 |
+
"eval_loss": 0.14557726681232452,
|
1334 |
+
"eval_runtime": 26.923,
|
1335 |
+
"eval_samples_per_second": 12.666,
|
1336 |
+
"eval_steps_per_second": 1.597,
|
1337 |
+
"eval_wer": 0.2701702442635085,
|
1338 |
+
"step": 1900
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 29.38,
|
1342 |
+
"learning_rate": 9.244166666666667e-05,
|
1343 |
+
"loss": 0.3271,
|
1344 |
+
"step": 1910
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 29.54,
|
1348 |
+
"learning_rate": 9.235833333333333e-05,
|
1349 |
+
"loss": 0.3439,
|
1350 |
+
"step": 1920
|
1351 |
+
},
|
1352 |
+
{
|
1353 |
+
"epoch": 29.69,
|
1354 |
+
"learning_rate": 9.2275e-05,
|
1355 |
+
"loss": 0.3184,
|
1356 |
+
"step": 1930
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 29.85,
|
1360 |
+
"learning_rate": 9.219166666666668e-05,
|
1361 |
+
"loss": 0.3458,
|
1362 |
+
"step": 1940
|
1363 |
+
},
|
1364 |
+
{
|
1365 |
+
"epoch": 30.0,
|
1366 |
+
"learning_rate": 9.210833333333334e-05,
|
1367 |
+
"loss": 0.2775,
|
1368 |
+
"step": 1950
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 30.15,
|
1372 |
+
"learning_rate": 9.2025e-05,
|
1373 |
+
"loss": 0.2959,
|
1374 |
+
"step": 1960
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 30.31,
|
1378 |
+
"learning_rate": 9.194166666666667e-05,
|
1379 |
+
"loss": 0.2861,
|
1380 |
+
"step": 1970
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 30.46,
|
1384 |
+
"learning_rate": 9.185833333333333e-05,
|
1385 |
+
"loss": 0.3047,
|
1386 |
+
"step": 1980
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 30.62,
|
1390 |
+
"learning_rate": 9.1775e-05,
|
1391 |
+
"loss": 0.3244,
|
1392 |
+
"step": 1990
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 30.77,
|
1396 |
+
"learning_rate": 9.169166666666668e-05,
|
1397 |
+
"loss": 0.2903,
|
1398 |
+
"step": 2000
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 30.77,
|
1402 |
+
"eval_cer": 0.06532494758909853,
|
1403 |
+
"eval_loss": 0.1265391856431961,
|
1404 |
+
"eval_runtime": 29.4587,
|
1405 |
+
"eval_samples_per_second": 11.576,
|
1406 |
+
"eval_steps_per_second": 1.46,
|
1407 |
+
"eval_wer": 0.2527757216876388,
|
1408 |
+
"step": 2000
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 30.92,
|
1412 |
+
"learning_rate": 9.160833333333334e-05,
|
1413 |
+
"loss": 0.2849,
|
1414 |
+
"step": 2010
|
1415 |
+
},
|
1416 |
+
{
|
1417 |
+
"epoch": 31.08,
|
1418 |
+
"learning_rate": 9.1525e-05,
|
1419 |
+
"loss": 0.3266,
|
1420 |
+
"step": 2020
|
1421 |
+
},
|
1422 |
+
{
|
1423 |
+
"epoch": 31.23,
|
1424 |
+
"learning_rate": 9.144166666666668e-05,
|
1425 |
+
"loss": 0.2915,
|
1426 |
+
"step": 2030
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 31.38,
|
1430 |
+
"learning_rate": 9.135833333333334e-05,
|
1431 |
+
"loss": 0.3066,
|
1432 |
+
"step": 2040
|
1433 |
+
},
|
1434 |
+
{
|
1435 |
+
"epoch": 31.54,
|
1436 |
+
"learning_rate": 9.1275e-05,
|
1437 |
+
"loss": 0.3167,
|
1438 |
+
"step": 2050
|
1439 |
+
},
|
1440 |
+
{
|
1441 |
+
"epoch": 31.69,
|
1442 |
+
"learning_rate": 9.119166666666667e-05,
|
1443 |
+
"loss": 0.2572,
|
1444 |
+
"step": 2060
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 31.85,
|
1448 |
+
"learning_rate": 9.110833333333334e-05,
|
1449 |
+
"loss": 0.3269,
|
1450 |
+
"step": 2070
|
1451 |
+
},
|
1452 |
+
{
|
1453 |
+
"epoch": 32.0,
|
1454 |
+
"learning_rate": 9.1025e-05,
|
1455 |
+
"loss": 0.2422,
|
1456 |
+
"step": 2080
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 32.15,
|
1460 |
+
"learning_rate": 9.094166666666666e-05,
|
1461 |
+
"loss": 0.3152,
|
1462 |
+
"step": 2090
|
1463 |
+
},
|
1464 |
+
{
|
1465 |
+
"epoch": 32.31,
|
1466 |
+
"learning_rate": 9.085833333333334e-05,
|
1467 |
+
"loss": 0.2703,
|
1468 |
+
"step": 2100
|
1469 |
+
},
|
1470 |
+
{
|
1471 |
+
"epoch": 32.31,
|
1472 |
+
"eval_cer": 0.060964360587002095,
|
1473 |
+
"eval_loss": 0.11160185188055038,
|
1474 |
+
"eval_runtime": 26.8072,
|
1475 |
+
"eval_samples_per_second": 12.72,
|
1476 |
+
"eval_steps_per_second": 1.604,
|
1477 |
+
"eval_wer": 0.24426350851221318,
|
1478 |
+
"step": 2100
|
1479 |
+
}
|
1480 |
+
],
|
1481 |
+
"max_steps": 13000,
|
1482 |
+
"num_train_epochs": 200,
|
1483 |
+
"total_flos": 8.493815951541043e+18,
|
1484 |
+
"trial_name": null,
|
1485 |
+
"trial_params": null
|
1486 |
+
}
|