End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +1174 -0
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": Infinity,
|
4 |
+
"eval_runtime": 744.5616,
|
5 |
+
"eval_samples": 14765,
|
6 |
+
"eval_samples_per_second": 19.83,
|
7 |
+
"eval_steps_per_second": 2.479,
|
8 |
+
"eval_wer": 0.17098213362202716,
|
9 |
+
"train_loss": 0.4200820018974194,
|
10 |
+
"train_runtime": 124144.1421,
|
11 |
+
"train_samples": 145805,
|
12 |
+
"train_samples_per_second": 5.872,
|
13 |
+
"train_steps_per_second": 0.092
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": Infinity,
|
4 |
+
"eval_runtime": 744.5616,
|
5 |
+
"eval_samples": 14765,
|
6 |
+
"eval_samples_per_second": 19.83,
|
7 |
+
"eval_steps_per_second": 2.479,
|
8 |
+
"eval_wer": 0.17098213362202716
|
9 |
+
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262657585
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:07ed9d47bcc629eedefa940f78ba02bee781864969d815ff5e81d0d73baad90a
|
3 |
size 1262657585
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"train_loss": 0.4200820018974194,
|
4 |
+
"train_runtime": 124144.1421,
|
5 |
+
"train_samples": 145805,
|
6 |
+
"train_samples_per_second": 5.872,
|
7 |
+
"train_steps_per_second": 0.092
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1174 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 5.0,
|
5 |
+
"global_step": 11395,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.04,
|
12 |
+
"eval_loss": Infinity,
|
13 |
+
"eval_runtime": 762.0088,
|
14 |
+
"eval_samples_per_second": 19.376,
|
15 |
+
"eval_steps_per_second": 2.423,
|
16 |
+
"eval_wer": 1.0,
|
17 |
+
"step": 100
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.09,
|
21 |
+
"eval_loss": Infinity,
|
22 |
+
"eval_runtime": 760.7665,
|
23 |
+
"eval_samples_per_second": 19.408,
|
24 |
+
"eval_steps_per_second": 2.427,
|
25 |
+
"eval_wer": 0.9983381717130623,
|
26 |
+
"step": 200
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.13,
|
30 |
+
"eval_loss": Infinity,
|
31 |
+
"eval_runtime": 758.492,
|
32 |
+
"eval_samples_per_second": 19.466,
|
33 |
+
"eval_steps_per_second": 2.434,
|
34 |
+
"eval_wer": 0.7671578598961529,
|
35 |
+
"step": 300
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 0.18,
|
39 |
+
"eval_loss": Infinity,
|
40 |
+
"eval_runtime": 763.9298,
|
41 |
+
"eval_samples_per_second": 19.328,
|
42 |
+
"eval_steps_per_second": 2.416,
|
43 |
+
"eval_wer": 0.6919342715882527,
|
44 |
+
"step": 400
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.22,
|
48 |
+
"learning_rate": 0.000996,
|
49 |
+
"loss": 2.9929,
|
50 |
+
"step": 500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.22,
|
54 |
+
"eval_loss": Infinity,
|
55 |
+
"eval_runtime": 771.5199,
|
56 |
+
"eval_samples_per_second": 19.138,
|
57 |
+
"eval_steps_per_second": 2.393,
|
58 |
+
"eval_wer": 0.6265920108122272,
|
59 |
+
"step": 500
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.26,
|
63 |
+
"eval_loss": Infinity,
|
64 |
+
"eval_runtime": 768.8673,
|
65 |
+
"eval_samples_per_second": 19.204,
|
66 |
+
"eval_steps_per_second": 2.401,
|
67 |
+
"eval_wer": 0.5512649892084595,
|
68 |
+
"step": 600
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.31,
|
72 |
+
"eval_loss": Infinity,
|
73 |
+
"eval_runtime": 774.562,
|
74 |
+
"eval_samples_per_second": 19.062,
|
75 |
+
"eval_steps_per_second": 2.383,
|
76 |
+
"eval_wer": 0.508133304831714,
|
77 |
+
"step": 700
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.35,
|
81 |
+
"eval_loss": Infinity,
|
82 |
+
"eval_runtime": 770.4155,
|
83 |
+
"eval_samples_per_second": 19.165,
|
84 |
+
"eval_steps_per_second": 2.396,
|
85 |
+
"eval_wer": 0.4944870053302625,
|
86 |
+
"step": 800
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.39,
|
90 |
+
"eval_loss": Infinity,
|
91 |
+
"eval_runtime": 769.5888,
|
92 |
+
"eval_samples_per_second": 19.186,
|
93 |
+
"eval_steps_per_second": 2.399,
|
94 |
+
"eval_wer": 0.4720419801270161,
|
95 |
+
"step": 900
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.44,
|
99 |
+
"learning_rate": 0.0009542909591555759,
|
100 |
+
"loss": 0.5311,
|
101 |
+
"step": 1000
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.44,
|
105 |
+
"eval_loss": Infinity,
|
106 |
+
"eval_runtime": 772.8626,
|
107 |
+
"eval_samples_per_second": 19.104,
|
108 |
+
"eval_steps_per_second": 2.389,
|
109 |
+
"eval_wer": 0.4387433544107405,
|
110 |
+
"step": 1000
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.48,
|
114 |
+
"eval_loss": Infinity,
|
115 |
+
"eval_runtime": 769.5289,
|
116 |
+
"eval_samples_per_second": 19.187,
|
117 |
+
"eval_steps_per_second": 2.399,
|
118 |
+
"eval_wer": 0.4410533646851146,
|
119 |
+
"step": 1100
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.53,
|
123 |
+
"eval_loss": Infinity,
|
124 |
+
"eval_runtime": 780.8054,
|
125 |
+
"eval_samples_per_second": 18.91,
|
126 |
+
"eval_steps_per_second": 2.364,
|
127 |
+
"eval_wer": 0.44290826845767167,
|
128 |
+
"step": 1200
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.57,
|
132 |
+
"eval_loss": Infinity,
|
133 |
+
"eval_runtime": 774.2302,
|
134 |
+
"eval_samples_per_second": 19.071,
|
135 |
+
"eval_steps_per_second": 2.384,
|
136 |
+
"eval_wer": 0.43216499679356785,
|
137 |
+
"step": 1300
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.61,
|
141 |
+
"eval_loss": Infinity,
|
142 |
+
"eval_runtime": 768.9729,
|
143 |
+
"eval_samples_per_second": 19.201,
|
144 |
+
"eval_steps_per_second": 2.401,
|
145 |
+
"eval_wer": 0.4531826425138428,
|
146 |
+
"step": 1400
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.66,
|
150 |
+
"learning_rate": 0.0009083983478659936,
|
151 |
+
"loss": 0.4654,
|
152 |
+
"step": 1500
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.66,
|
156 |
+
"eval_loss": Infinity,
|
157 |
+
"eval_runtime": 777.4905,
|
158 |
+
"eval_samples_per_second": 18.991,
|
159 |
+
"eval_steps_per_second": 2.374,
|
160 |
+
"eval_wer": 0.44915563952806836,
|
161 |
+
"step": 1500
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 0.7,
|
165 |
+
"eval_loss": Infinity,
|
166 |
+
"eval_runtime": 773.9985,
|
167 |
+
"eval_samples_per_second": 19.076,
|
168 |
+
"eval_steps_per_second": 2.385,
|
169 |
+
"eval_wer": 0.3878817550561643,
|
170 |
+
"step": 1600
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.75,
|
174 |
+
"eval_loss": Infinity,
|
175 |
+
"eval_runtime": 773.8841,
|
176 |
+
"eval_samples_per_second": 19.079,
|
177 |
+
"eval_steps_per_second": 2.385,
|
178 |
+
"eval_wer": 0.3835927210541922,
|
179 |
+
"step": 1700
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.79,
|
183 |
+
"eval_loss": Infinity,
|
184 |
+
"eval_runtime": 773.515,
|
185 |
+
"eval_samples_per_second": 19.088,
|
186 |
+
"eval_steps_per_second": 2.387,
|
187 |
+
"eval_wer": 0.37425614221388626,
|
188 |
+
"step": 1800
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.83,
|
192 |
+
"eval_loss": Infinity,
|
193 |
+
"eval_runtime": 774.3742,
|
194 |
+
"eval_samples_per_second": 19.067,
|
195 |
+
"eval_steps_per_second": 2.384,
|
196 |
+
"eval_wer": 0.36867074423704155,
|
197 |
+
"step": 1900
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 0.88,
|
201 |
+
"learning_rate": 0.0008625975217989904,
|
202 |
+
"loss": 0.4254,
|
203 |
+
"step": 2000
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 0.88,
|
207 |
+
"eval_loss": Infinity,
|
208 |
+
"eval_runtime": 774.0591,
|
209 |
+
"eval_samples_per_second": 19.075,
|
210 |
+
"eval_steps_per_second": 2.385,
|
211 |
+
"eval_wer": 0.3792830003930465,
|
212 |
+
"step": 2000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.92,
|
216 |
+
"eval_loss": Infinity,
|
217 |
+
"eval_runtime": 768.0215,
|
218 |
+
"eval_samples_per_second": 19.225,
|
219 |
+
"eval_steps_per_second": 2.404,
|
220 |
+
"eval_wer": 0.3766006302535495,
|
221 |
+
"step": 2100
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.97,
|
225 |
+
"eval_loss": Infinity,
|
226 |
+
"eval_runtime": 772.2372,
|
227 |
+
"eval_samples_per_second": 19.12,
|
228 |
+
"eval_steps_per_second": 2.39,
|
229 |
+
"eval_wer": 0.3705256480095986,
|
230 |
+
"step": 2200
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 1.01,
|
234 |
+
"eval_loss": Infinity,
|
235 |
+
"eval_runtime": 767.2028,
|
236 |
+
"eval_samples_per_second": 19.245,
|
237 |
+
"eval_steps_per_second": 2.406,
|
238 |
+
"eval_wer": 0.3271802014880604,
|
239 |
+
"step": 2300
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 1.05,
|
243 |
+
"eval_loss": Infinity,
|
244 |
+
"eval_runtime": 768.4281,
|
245 |
+
"eval_samples_per_second": 19.215,
|
246 |
+
"eval_steps_per_second": 2.402,
|
247 |
+
"eval_wer": 0.31850559574130644,
|
248 |
+
"step": 2400
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.1,
|
252 |
+
"learning_rate": 0.000816704910509408,
|
253 |
+
"loss": 0.3997,
|
254 |
+
"step": 2500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 1.1,
|
258 |
+
"eval_loss": Infinity,
|
259 |
+
"eval_runtime": 767.2666,
|
260 |
+
"eval_samples_per_second": 19.244,
|
261 |
+
"eval_steps_per_second": 2.406,
|
262 |
+
"eval_wer": 0.3244426669241006,
|
263 |
+
"step": 2500
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.14,
|
267 |
+
"eval_loss": Infinity,
|
268 |
+
"eval_runtime": 767.2447,
|
269 |
+
"eval_samples_per_second": 19.244,
|
270 |
+
"eval_steps_per_second": 2.406,
|
271 |
+
"eval_wer": 0.30816916170761477,
|
272 |
+
"step": 2600
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 1.18,
|
276 |
+
"eval_loss": Infinity,
|
277 |
+
"eval_runtime": 771.1785,
|
278 |
+
"eval_samples_per_second": 19.146,
|
279 |
+
"eval_steps_per_second": 2.394,
|
280 |
+
"eval_wer": 0.30403872542597277,
|
281 |
+
"step": 2700
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.23,
|
285 |
+
"eval_loss": Infinity,
|
286 |
+
"eval_runtime": 761.2544,
|
287 |
+
"eval_samples_per_second": 19.396,
|
288 |
+
"eval_steps_per_second": 2.425,
|
289 |
+
"eval_wer": 0.30282510808779417,
|
290 |
+
"step": 2800
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.27,
|
294 |
+
"eval_loss": Infinity,
|
295 |
+
"eval_runtime": 770.9042,
|
296 |
+
"eval_samples_per_second": 19.153,
|
297 |
+
"eval_steps_per_second": 2.395,
|
298 |
+
"eval_wer": 0.3112101006061191,
|
299 |
+
"step": 2900
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.32,
|
303 |
+
"learning_rate": 0.0007709040844424048,
|
304 |
+
"loss": 0.3668,
|
305 |
+
"step": 3000
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 1.32,
|
309 |
+
"eval_loss": Infinity,
|
310 |
+
"eval_runtime": 763.6174,
|
311 |
+
"eval_samples_per_second": 19.336,
|
312 |
+
"eval_steps_per_second": 2.417,
|
313 |
+
"eval_wer": 0.31098944290826847,
|
314 |
+
"step": 3000
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 1.36,
|
318 |
+
"eval_loss": Infinity,
|
319 |
+
"eval_runtime": 765.7415,
|
320 |
+
"eval_samples_per_second": 19.282,
|
321 |
+
"eval_steps_per_second": 2.411,
|
322 |
+
"eval_wer": 0.3066521400348915,
|
323 |
+
"step": 3100
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 1.4,
|
327 |
+
"eval_loss": Infinity,
|
328 |
+
"eval_runtime": 762.1465,
|
329 |
+
"eval_samples_per_second": 19.373,
|
330 |
+
"eval_steps_per_second": 2.422,
|
331 |
+
"eval_wer": 0.29612952606863835,
|
332 |
+
"step": 3200
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.45,
|
336 |
+
"eval_loss": Infinity,
|
337 |
+
"eval_runtime": 764.1293,
|
338 |
+
"eval_samples_per_second": 19.323,
|
339 |
+
"eval_steps_per_second": 2.416,
|
340 |
+
"eval_wer": 0.3080933106239786,
|
341 |
+
"step": 3300
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 1.49,
|
345 |
+
"eval_loss": Infinity,
|
346 |
+
"eval_runtime": 764.6533,
|
347 |
+
"eval_samples_per_second": 19.309,
|
348 |
+
"eval_steps_per_second": 2.414,
|
349 |
+
"eval_wer": 0.2936195447555871,
|
350 |
+
"step": 3400
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 1.54,
|
354 |
+
"learning_rate": 0.0007250114731528225,
|
355 |
+
"loss": 0.3645,
|
356 |
+
"step": 3500
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.54,
|
360 |
+
"eval_loss": Infinity,
|
361 |
+
"eval_runtime": 762.2683,
|
362 |
+
"eval_samples_per_second": 19.37,
|
363 |
+
"eval_steps_per_second": 2.422,
|
364 |
+
"eval_wer": 0.30368015666696546,
|
365 |
+
"step": 3500
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 1.58,
|
369 |
+
"eval_loss": Infinity,
|
370 |
+
"eval_runtime": 760.0026,
|
371 |
+
"eval_samples_per_second": 19.428,
|
372 |
+
"eval_steps_per_second": 2.429,
|
373 |
+
"eval_wer": 0.2973914122782218,
|
374 |
+
"step": 3600
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.62,
|
378 |
+
"eval_loss": Infinity,
|
379 |
+
"eval_runtime": 760.5942,
|
380 |
+
"eval_samples_per_second": 19.412,
|
381 |
+
"eval_steps_per_second": 2.427,
|
382 |
+
"eval_wer": 0.30096330876217925,
|
383 |
+
"step": 3700
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 1.67,
|
387 |
+
"eval_loss": Infinity,
|
388 |
+
"eval_runtime": 761.2356,
|
389 |
+
"eval_samples_per_second": 19.396,
|
390 |
+
"eval_steps_per_second": 2.425,
|
391 |
+
"eval_wer": 0.2985498651919377,
|
392 |
+
"step": 3800
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 1.71,
|
396 |
+
"eval_loss": Infinity,
|
397 |
+
"eval_runtime": 767.0946,
|
398 |
+
"eval_samples_per_second": 19.248,
|
399 |
+
"eval_steps_per_second": 2.406,
|
400 |
+
"eval_wer": 0.2975775922107833,
|
401 |
+
"step": 3900
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 1.76,
|
405 |
+
"learning_rate": 0.00067911886186324,
|
406 |
+
"loss": 0.3624,
|
407 |
+
"step": 4000
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 1.76,
|
411 |
+
"eval_loss": Infinity,
|
412 |
+
"eval_runtime": 764.3373,
|
413 |
+
"eval_samples_per_second": 19.317,
|
414 |
+
"eval_steps_per_second": 2.415,
|
415 |
+
"eval_wer": 0.292757600623358,
|
416 |
+
"step": 4000
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.8,
|
420 |
+
"eval_loss": Infinity,
|
421 |
+
"eval_runtime": 778.1044,
|
422 |
+
"eval_samples_per_second": 18.976,
|
423 |
+
"eval_steps_per_second": 2.372,
|
424 |
+
"eval_wer": 0.2859516897552768,
|
425 |
+
"step": 4100
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 1.84,
|
429 |
+
"eval_loss": Infinity,
|
430 |
+
"eval_runtime": 780.7174,
|
431 |
+
"eval_samples_per_second": 18.912,
|
432 |
+
"eval_steps_per_second": 2.364,
|
433 |
+
"eval_wer": 0.29222664303790485,
|
434 |
+
"step": 4200
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 1.89,
|
438 |
+
"eval_loss": Infinity,
|
439 |
+
"eval_runtime": 778.463,
|
440 |
+
"eval_samples_per_second": 18.967,
|
441 |
+
"eval_steps_per_second": 2.371,
|
442 |
+
"eval_wer": 0.2865929761896553,
|
443 |
+
"step": 4300
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 1.93,
|
447 |
+
"eval_loss": Infinity,
|
448 |
+
"eval_runtime": 777.1309,
|
449 |
+
"eval_samples_per_second": 18.999,
|
450 |
+
"eval_steps_per_second": 2.375,
|
451 |
+
"eval_wer": 0.2775529061308362,
|
452 |
+
"step": 4400
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 1.97,
|
456 |
+
"learning_rate": 0.0006332262505736577,
|
457 |
+
"loss": 0.3527,
|
458 |
+
"step": 4500
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.97,
|
462 |
+
"eval_loss": Infinity,
|
463 |
+
"eval_runtime": 776.0915,
|
464 |
+
"eval_samples_per_second": 19.025,
|
465 |
+
"eval_steps_per_second": 2.379,
|
466 |
+
"eval_wer": 0.27924921218306314,
|
467 |
+
"step": 4500
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 2.02,
|
471 |
+
"eval_loss": Infinity,
|
472 |
+
"eval_runtime": 777.8615,
|
473 |
+
"eval_samples_per_second": 18.982,
|
474 |
+
"eval_steps_per_second": 2.373,
|
475 |
+
"eval_wer": 0.2857655098227153,
|
476 |
+
"step": 4600
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 2.06,
|
480 |
+
"eval_loss": Infinity,
|
481 |
+
"eval_runtime": 779.7088,
|
482 |
+
"eval_samples_per_second": 18.937,
|
483 |
+
"eval_steps_per_second": 2.368,
|
484 |
+
"eval_wer": 0.27665648423331796,
|
485 |
+
"step": 4700
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 2.11,
|
489 |
+
"eval_loss": Infinity,
|
490 |
+
"eval_runtime": 780.5896,
|
491 |
+
"eval_samples_per_second": 18.915,
|
492 |
+
"eval_steps_per_second": 2.365,
|
493 |
+
"eval_wer": 0.28240047993049283,
|
494 |
+
"step": 4800
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.15,
|
498 |
+
"eval_loss": Infinity,
|
499 |
+
"eval_runtime": 781.7435,
|
500 |
+
"eval_samples_per_second": 18.887,
|
501 |
+
"eval_steps_per_second": 2.361,
|
502 |
+
"eval_wer": 0.27986291640521027,
|
503 |
+
"step": 4900
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 2.19,
|
507 |
+
"learning_rate": 0.0005873336392840752,
|
508 |
+
"loss": 0.3162,
|
509 |
+
"step": 5000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 2.19,
|
513 |
+
"eval_loss": Infinity,
|
514 |
+
"eval_runtime": 785.7508,
|
515 |
+
"eval_samples_per_second": 18.791,
|
516 |
+
"eval_steps_per_second": 2.349,
|
517 |
+
"eval_wer": 0.26727853207466507,
|
518 |
+
"step": 5000
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 2.24,
|
522 |
+
"eval_loss": Infinity,
|
523 |
+
"eval_runtime": 789.097,
|
524 |
+
"eval_samples_per_second": 18.711,
|
525 |
+
"eval_steps_per_second": 2.339,
|
526 |
+
"eval_wer": 0.2961915860461588,
|
527 |
+
"step": 5100
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 2.28,
|
531 |
+
"eval_loss": Infinity,
|
532 |
+
"eval_runtime": 782.4986,
|
533 |
+
"eval_samples_per_second": 18.869,
|
534 |
+
"eval_steps_per_second": 2.359,
|
535 |
+
"eval_wer": 0.2735948586756401,
|
536 |
+
"step": 5200
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.33,
|
540 |
+
"eval_loss": Infinity,
|
541 |
+
"eval_runtime": 780.5596,
|
542 |
+
"eval_samples_per_second": 18.916,
|
543 |
+
"eval_steps_per_second": 2.365,
|
544 |
+
"eval_wer": 0.2651547017328525,
|
545 |
+
"step": 5300
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 2.37,
|
549 |
+
"eval_loss": Infinity,
|
550 |
+
"eval_runtime": 782.8773,
|
551 |
+
"eval_samples_per_second": 18.86,
|
552 |
+
"eval_steps_per_second": 2.358,
|
553 |
+
"eval_wer": 0.2550527165031271,
|
554 |
+
"step": 5400
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 2.41,
|
558 |
+
"learning_rate": 0.0005414410279944929,
|
559 |
+
"loss": 0.3063,
|
560 |
+
"step": 5500
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 2.41,
|
564 |
+
"eval_loss": Infinity,
|
565 |
+
"eval_runtime": 779.9742,
|
566 |
+
"eval_samples_per_second": 18.93,
|
567 |
+
"eval_steps_per_second": 2.367,
|
568 |
+
"eval_wer": 0.26804393846408453,
|
569 |
+
"step": 5500
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 2.46,
|
573 |
+
"eval_loss": Infinity,
|
574 |
+
"eval_runtime": 783.4711,
|
575 |
+
"eval_samples_per_second": 18.846,
|
576 |
+
"eval_steps_per_second": 2.356,
|
577 |
+
"eval_wer": 0.2557767495741996,
|
578 |
+
"step": 5600
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.5,
|
582 |
+
"eval_loss": Infinity,
|
583 |
+
"eval_runtime": 792.9653,
|
584 |
+
"eval_samples_per_second": 18.62,
|
585 |
+
"eval_steps_per_second": 2.328,
|
586 |
+
"eval_wer": 0.2597830659008006,
|
587 |
+
"step": 5700
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 2.54,
|
591 |
+
"eval_loss": Infinity,
|
592 |
+
"eval_runtime": 788.5494,
|
593 |
+
"eval_samples_per_second": 18.724,
|
594 |
+
"eval_steps_per_second": 2.341,
|
595 |
+
"eval_wer": 0.25175664214148297,
|
596 |
+
"step": 5800
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 2.59,
|
600 |
+
"eval_loss": Infinity,
|
601 |
+
"eval_runtime": 777.5477,
|
602 |
+
"eval_samples_per_second": 18.989,
|
603 |
+
"eval_steps_per_second": 2.374,
|
604 |
+
"eval_wer": 0.254149399052551,
|
605 |
+
"step": 5900
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 2.63,
|
609 |
+
"learning_rate": 0.0004955484167049105,
|
610 |
+
"loss": 0.2913,
|
611 |
+
"step": 6000
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 2.63,
|
615 |
+
"eval_loss": Infinity,
|
616 |
+
"eval_runtime": 785.1255,
|
617 |
+
"eval_samples_per_second": 18.806,
|
618 |
+
"eval_steps_per_second": 2.351,
|
619 |
+
"eval_wer": 0.2507361002889237,
|
620 |
+
"step": 6000
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 2.68,
|
624 |
+
"eval_loss": Infinity,
|
625 |
+
"eval_runtime": 785.3479,
|
626 |
+
"eval_samples_per_second": 18.801,
|
627 |
+
"eval_steps_per_second": 2.351,
|
628 |
+
"eval_wer": 0.24998448500561987,
|
629 |
+
"step": 6100
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 2.72,
|
633 |
+
"eval_loss": Infinity,
|
634 |
+
"eval_runtime": 781.3987,
|
635 |
+
"eval_samples_per_second": 18.896,
|
636 |
+
"eval_steps_per_second": 2.362,
|
637 |
+
"eval_wer": 0.24352335179043036,
|
638 |
+
"step": 6200
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 2.76,
|
642 |
+
"eval_loss": Infinity,
|
643 |
+
"eval_runtime": 781.4686,
|
644 |
+
"eval_samples_per_second": 18.894,
|
645 |
+
"eval_steps_per_second": 2.362,
|
646 |
+
"eval_wer": 0.23755180284234698,
|
647 |
+
"step": 6300
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 2.81,
|
651 |
+
"eval_loss": Infinity,
|
652 |
+
"eval_runtime": 775.7634,
|
653 |
+
"eval_samples_per_second": 19.033,
|
654 |
+
"eval_steps_per_second": 2.38,
|
655 |
+
"eval_wer": 0.23476599940698245,
|
656 |
+
"step": 6400
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 2.85,
|
660 |
+
"learning_rate": 0.0004497475906379073,
|
661 |
+
"loss": 0.2797,
|
662 |
+
"step": 6500
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 2.85,
|
666 |
+
"eval_loss": Infinity,
|
667 |
+
"eval_runtime": 772.9219,
|
668 |
+
"eval_samples_per_second": 19.103,
|
669 |
+
"eval_steps_per_second": 2.388,
|
670 |
+
"eval_wer": 0.2512049978968563,
|
671 |
+
"step": 6500
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 2.9,
|
675 |
+
"eval_loss": Infinity,
|
676 |
+
"eval_runtime": 771.2336,
|
677 |
+
"eval_samples_per_second": 19.145,
|
678 |
+
"eval_steps_per_second": 2.394,
|
679 |
+
"eval_wer": 0.23818619372366762,
|
680 |
+
"step": 6600
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 2.94,
|
684 |
+
"eval_loss": Infinity,
|
685 |
+
"eval_runtime": 777.0185,
|
686 |
+
"eval_samples_per_second": 19.002,
|
687 |
+
"eval_steps_per_second": 2.376,
|
688 |
+
"eval_wer": 0.2523082863861096,
|
689 |
+
"step": 6700
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 2.98,
|
693 |
+
"eval_loss": Infinity,
|
694 |
+
"eval_runtime": 773.6017,
|
695 |
+
"eval_samples_per_second": 19.086,
|
696 |
+
"eval_steps_per_second": 2.386,
|
697 |
+
"eval_wer": 0.25221864419635776,
|
698 |
+
"step": 6800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 3.03,
|
702 |
+
"eval_loss": Infinity,
|
703 |
+
"eval_runtime": 780.1394,
|
704 |
+
"eval_samples_per_second": 18.926,
|
705 |
+
"eval_steps_per_second": 2.366,
|
706 |
+
"eval_wer": 0.24091683273456949,
|
707 |
+
"step": 6900
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 3.07,
|
711 |
+
"learning_rate": 0.0004038549793483249,
|
712 |
+
"loss": 0.2766,
|
713 |
+
"step": 7000
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 3.07,
|
717 |
+
"eval_loss": Infinity,
|
718 |
+
"eval_runtime": 783.1904,
|
719 |
+
"eval_samples_per_second": 18.852,
|
720 |
+
"eval_steps_per_second": 2.357,
|
721 |
+
"eval_wer": 0.24534377779769825,
|
722 |
+
"step": 7000
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 3.12,
|
726 |
+
"eval_loss": Infinity,
|
727 |
+
"eval_runtime": 777.1186,
|
728 |
+
"eval_samples_per_second": 19.0,
|
729 |
+
"eval_steps_per_second": 2.375,
|
730 |
+
"eval_wer": 0.2326076912998807,
|
731 |
+
"step": 7100
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 3.16,
|
735 |
+
"eval_loss": Infinity,
|
736 |
+
"eval_runtime": 781.1613,
|
737 |
+
"eval_samples_per_second": 18.901,
|
738 |
+
"eval_steps_per_second": 2.363,
|
739 |
+
"eval_wer": 0.22860137497327973,
|
740 |
+
"step": 7200
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 3.2,
|
744 |
+
"eval_loss": Infinity,
|
745 |
+
"eval_runtime": 786.2675,
|
746 |
+
"eval_samples_per_second": 18.779,
|
747 |
+
"eval_steps_per_second": 2.348,
|
748 |
+
"eval_wer": 0.23420745960929795,
|
749 |
+
"step": 7300
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 3.25,
|
753 |
+
"eval_loss": Infinity,
|
754 |
+
"eval_runtime": 781.7233,
|
755 |
+
"eval_samples_per_second": 18.888,
|
756 |
+
"eval_steps_per_second": 2.361,
|
757 |
+
"eval_wer": 0.23047006985195248,
|
758 |
+
"step": 7400
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 3.29,
|
762 |
+
"learning_rate": 0.0003579623680587425,
|
763 |
+
"loss": 0.2468,
|
764 |
+
"step": 7500
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 3.29,
|
768 |
+
"eval_loss": Infinity,
|
769 |
+
"eval_runtime": 777.7207,
|
770 |
+
"eval_samples_per_second": 18.985,
|
771 |
+
"eval_steps_per_second": 2.374,
|
772 |
+
"eval_wer": 0.2238089655980858,
|
773 |
+
"step": 7500
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 3.33,
|
777 |
+
"eval_loss": Infinity,
|
778 |
+
"eval_runtime": 786.7846,
|
779 |
+
"eval_samples_per_second": 18.766,
|
780 |
+
"eval_steps_per_second": 2.346,
|
781 |
+
"eval_wer": 0.23211121147971672,
|
782 |
+
"step": 7600
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 3.38,
|
786 |
+
"eval_loss": Infinity,
|
787 |
+
"eval_runtime": 784.5553,
|
788 |
+
"eval_samples_per_second": 18.82,
|
789 |
+
"eval_steps_per_second": 2.353,
|
790 |
+
"eval_wer": 0.23048386095806814,
|
791 |
+
"step": 7700
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 3.42,
|
795 |
+
"eval_loss": Infinity,
|
796 |
+
"eval_runtime": 776.8796,
|
797 |
+
"eval_samples_per_second": 19.006,
|
798 |
+
"eval_steps_per_second": 2.376,
|
799 |
+
"eval_wer": 0.21740989236041677,
|
800 |
+
"step": 7800
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 3.47,
|
804 |
+
"eval_loss": Infinity,
|
805 |
+
"eval_runtime": 778.8968,
|
806 |
+
"eval_samples_per_second": 18.956,
|
807 |
+
"eval_steps_per_second": 2.37,
|
808 |
+
"eval_wer": 0.22008536694685596,
|
809 |
+
"step": 7900
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 3.51,
|
813 |
+
"learning_rate": 0.00031206975676916014,
|
814 |
+
"loss": 0.2439,
|
815 |
+
"step": 8000
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 3.51,
|
819 |
+
"eval_loss": Infinity,
|
820 |
+
"eval_runtime": 785.8625,
|
821 |
+
"eval_samples_per_second": 18.788,
|
822 |
+
"eval_steps_per_second": 2.349,
|
823 |
+
"eval_wer": 0.21332772495017963,
|
824 |
+
"step": 8000
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 3.55,
|
828 |
+
"eval_loss": Infinity,
|
829 |
+
"eval_runtime": 778.9388,
|
830 |
+
"eval_samples_per_second": 18.955,
|
831 |
+
"eval_steps_per_second": 2.37,
|
832 |
+
"eval_wer": 0.22168513525627323,
|
833 |
+
"step": 8100
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 3.6,
|
837 |
+
"eval_loss": Infinity,
|
838 |
+
"eval_runtime": 779.3965,
|
839 |
+
"eval_samples_per_second": 18.944,
|
840 |
+
"eval_steps_per_second": 2.368,
|
841 |
+
"eval_wer": 0.21885106294950388,
|
842 |
+
"step": 8200
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 3.64,
|
846 |
+
"eval_loss": Infinity,
|
847 |
+
"eval_runtime": 786.4104,
|
848 |
+
"eval_samples_per_second": 18.775,
|
849 |
+
"eval_steps_per_second": 2.347,
|
850 |
+
"eval_wer": 0.2105212348556416,
|
851 |
+
"step": 8300
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 3.69,
|
855 |
+
"eval_loss": Infinity,
|
856 |
+
"eval_runtime": 784.1132,
|
857 |
+
"eval_samples_per_second": 18.83,
|
858 |
+
"eval_steps_per_second": 2.354,
|
859 |
+
"eval_wer": 0.21176932995910938,
|
860 |
+
"step": 8400
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 3.73,
|
864 |
+
"learning_rate": 0.0002661771454795778,
|
865 |
+
"loss": 0.2357,
|
866 |
+
"step": 8500
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 3.73,
|
870 |
+
"eval_loss": Infinity,
|
871 |
+
"eval_runtime": 776.0702,
|
872 |
+
"eval_samples_per_second": 19.025,
|
873 |
+
"eval_steps_per_second": 2.379,
|
874 |
+
"eval_wer": 0.20932830417663648,
|
875 |
+
"step": 8500
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 3.77,
|
879 |
+
"eval_loss": Infinity,
|
880 |
+
"eval_runtime": 777.6609,
|
881 |
+
"eval_samples_per_second": 18.986,
|
882 |
+
"eval_steps_per_second": 2.374,
|
883 |
+
"eval_wer": 0.21030747271084876,
|
884 |
+
"step": 8600
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 3.82,
|
888 |
+
"eval_loss": Infinity,
|
889 |
+
"eval_runtime": 773.9787,
|
890 |
+
"eval_samples_per_second": 19.077,
|
891 |
+
"eval_steps_per_second": 2.385,
|
892 |
+
"eval_wer": 0.20353603960805677,
|
893 |
+
"step": 8700
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 3.86,
|
897 |
+
"eval_loss": Infinity,
|
898 |
+
"eval_runtime": 780.5659,
|
899 |
+
"eval_samples_per_second": 18.916,
|
900 |
+
"eval_steps_per_second": 2.365,
|
901 |
+
"eval_wer": 0.2019362712986395,
|
902 |
+
"step": 8800
|
903 |
+
},
|
904 |
+
{
|
905 |
+
"epoch": 3.91,
|
906 |
+
"eval_loss": Infinity,
|
907 |
+
"eval_runtime": 779.996,
|
908 |
+
"eval_samples_per_second": 18.93,
|
909 |
+
"eval_steps_per_second": 2.367,
|
910 |
+
"eval_wer": 0.20323953082656995,
|
911 |
+
"step": 8900
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 3.95,
|
915 |
+
"learning_rate": 0.0002203763194125746,
|
916 |
+
"loss": 0.2217,
|
917 |
+
"step": 9000
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 3.95,
|
921 |
+
"eval_loss": Infinity,
|
922 |
+
"eval_runtime": 778.6538,
|
923 |
+
"eval_samples_per_second": 18.962,
|
924 |
+
"eval_steps_per_second": 2.371,
|
925 |
+
"eval_wer": 0.20561849663152232,
|
926 |
+
"step": 9000
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 3.99,
|
930 |
+
"eval_loss": Infinity,
|
931 |
+
"eval_runtime": 778.3405,
|
932 |
+
"eval_samples_per_second": 18.97,
|
933 |
+
"eval_steps_per_second": 2.372,
|
934 |
+
"eval_wer": 0.20215003344343233,
|
935 |
+
"step": 9100
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 4.04,
|
939 |
+
"eval_loss": Infinity,
|
940 |
+
"eval_runtime": 787.2038,
|
941 |
+
"eval_samples_per_second": 18.756,
|
942 |
+
"eval_steps_per_second": 2.345,
|
943 |
+
"eval_wer": 0.19324787444576993,
|
944 |
+
"step": 9200
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 4.08,
|
948 |
+
"eval_loss": Infinity,
|
949 |
+
"eval_runtime": 775.7143,
|
950 |
+
"eval_samples_per_second": 19.034,
|
951 |
+
"eval_steps_per_second": 2.38,
|
952 |
+
"eval_wer": 0.19350300990890976,
|
953 |
+
"step": 9300
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 4.12,
|
957 |
+
"eval_loss": Infinity,
|
958 |
+
"eval_runtime": 773.9153,
|
959 |
+
"eval_samples_per_second": 19.078,
|
960 |
+
"eval_steps_per_second": 2.385,
|
961 |
+
"eval_wer": 0.1905655043062729,
|
962 |
+
"step": 9400
|
963 |
+
},
|
964 |
+
{
|
965 |
+
"epoch": 4.17,
|
966 |
+
"learning_rate": 0.0001744837081229922,
|
967 |
+
"loss": 0.2025,
|
968 |
+
"step": 9500
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 4.17,
|
972 |
+
"eval_loss": Infinity,
|
973 |
+
"eval_runtime": 779.0392,
|
974 |
+
"eval_samples_per_second": 18.953,
|
975 |
+
"eval_steps_per_second": 2.37,
|
976 |
+
"eval_wer": 0.18794519414429633,
|
977 |
+
"step": 9500
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 4.21,
|
981 |
+
"eval_loss": Infinity,
|
982 |
+
"eval_runtime": 779.0747,
|
983 |
+
"eval_samples_per_second": 18.952,
|
984 |
+
"eval_steps_per_second": 2.369,
|
985 |
+
"eval_wer": 0.18824859847884098,
|
986 |
+
"step": 9600
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 4.26,
|
990 |
+
"eval_loss": Infinity,
|
991 |
+
"eval_runtime": 777.6573,
|
992 |
+
"eval_samples_per_second": 18.987,
|
993 |
+
"eval_steps_per_second": 2.374,
|
994 |
+
"eval_wer": 0.18538004840678246,
|
995 |
+
"step": 9700
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 4.3,
|
999 |
+
"eval_loss": Infinity,
|
1000 |
+
"eval_runtime": 772.7615,
|
1001 |
+
"eval_samples_per_second": 19.107,
|
1002 |
+
"eval_steps_per_second": 2.389,
|
1003 |
+
"eval_wer": 0.18654539687355626,
|
1004 |
+
"step": 9800
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 4.34,
|
1008 |
+
"eval_loss": Infinity,
|
1009 |
+
"eval_runtime": 778.0179,
|
1010 |
+
"eval_samples_per_second": 18.978,
|
1011 |
+
"eval_steps_per_second": 2.373,
|
1012 |
+
"eval_wer": 0.18438708876645452,
|
1013 |
+
"step": 9900
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 4.39,
|
1017 |
+
"learning_rate": 0.00012859109683340984,
|
1018 |
+
"loss": 0.1869,
|
1019 |
+
"step": 10000
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 4.39,
|
1023 |
+
"eval_loss": Infinity,
|
1024 |
+
"eval_runtime": 778.8028,
|
1025 |
+
"eval_samples_per_second": 18.959,
|
1026 |
+
"eval_steps_per_second": 2.37,
|
1027 |
+
"eval_wer": 0.1821667206818323,
|
1028 |
+
"step": 10000
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 4.43,
|
1032 |
+
"eval_loss": Infinity,
|
1033 |
+
"eval_runtime": 777.366,
|
1034 |
+
"eval_samples_per_second": 18.994,
|
1035 |
+
"eval_steps_per_second": 2.375,
|
1036 |
+
"eval_wer": 0.1815461209066273,
|
1037 |
+
"step": 10100
|
1038 |
+
},
|
1039 |
+
{
|
1040 |
+
"epoch": 4.48,
|
1041 |
+
"eval_loss": Infinity,
|
1042 |
+
"eval_runtime": 781.9611,
|
1043 |
+
"eval_samples_per_second": 18.882,
|
1044 |
+
"eval_steps_per_second": 2.361,
|
1045 |
+
"eval_wer": 0.18118755214762,
|
1046 |
+
"step": 10200
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 4.52,
|
1050 |
+
"eval_loss": Infinity,
|
1051 |
+
"eval_runtime": 760.9314,
|
1052 |
+
"eval_samples_per_second": 19.404,
|
1053 |
+
"eval_steps_per_second": 2.426,
|
1054 |
+
"eval_wer": 0.17923611063225325,
|
1055 |
+
"step": 10300
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 4.56,
|
1059 |
+
"eval_loss": Infinity,
|
1060 |
+
"eval_runtime": 761.9088,
|
1061 |
+
"eval_samples_per_second": 19.379,
|
1062 |
+
"eval_steps_per_second": 2.423,
|
1063 |
+
"eval_wer": 0.17967742602795458,
|
1064 |
+
"step": 10400
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 4.61,
|
1068 |
+
"learning_rate": 8.269848554382745e-05,
|
1069 |
+
"loss": 0.1863,
|
1070 |
+
"step": 10500
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 4.61,
|
1074 |
+
"eval_loss": Infinity,
|
1075 |
+
"eval_runtime": 760.7934,
|
1076 |
+
"eval_samples_per_second": 19.407,
|
1077 |
+
"eval_steps_per_second": 2.426,
|
1078 |
+
"eval_wer": 0.1773812068596962,
|
1079 |
+
"step": 10500
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 4.65,
|
1083 |
+
"eval_loss": Infinity,
|
1084 |
+
"eval_runtime": 755.7192,
|
1085 |
+
"eval_samples_per_second": 19.538,
|
1086 |
+
"eval_steps_per_second": 2.443,
|
1087 |
+
"eval_wer": 0.1767330248722599,
|
1088 |
+
"step": 10600
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 4.7,
|
1092 |
+
"eval_loss": Infinity,
|
1093 |
+
"eval_runtime": 756.3189,
|
1094 |
+
"eval_samples_per_second": 19.522,
|
1095 |
+
"eval_steps_per_second": 2.441,
|
1096 |
+
"eval_wer": 0.17650547162135138,
|
1097 |
+
"step": 10700
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 4.74,
|
1101 |
+
"eval_loss": Infinity,
|
1102 |
+
"eval_runtime": 761.0501,
|
1103 |
+
"eval_samples_per_second": 19.401,
|
1104 |
+
"eval_steps_per_second": 2.426,
|
1105 |
+
"eval_wer": 0.1752918542831728,
|
1106 |
+
"step": 10800
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 4.78,
|
1110 |
+
"eval_loss": Infinity,
|
1111 |
+
"eval_runtime": 756.3701,
|
1112 |
+
"eval_samples_per_second": 19.521,
|
1113 |
+
"eval_steps_per_second": 2.441,
|
1114 |
+
"eval_wer": 0.17308527730466622,
|
1115 |
+
"step": 10900
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 4.83,
|
1119 |
+
"learning_rate": 3.6805874254245066e-05,
|
1120 |
+
"loss": 0.178,
|
1121 |
+
"step": 11000
|
1122 |
+
},
|
1123 |
+
{
|
1124 |
+
"epoch": 4.83,
|
1125 |
+
"eval_loss": Infinity,
|
1126 |
+
"eval_runtime": 758.6147,
|
1127 |
+
"eval_samples_per_second": 19.463,
|
1128 |
+
"eval_steps_per_second": 2.433,
|
1129 |
+
"eval_wer": 0.17265775301508057,
|
1130 |
+
"step": 11000
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 4.87,
|
1134 |
+
"eval_loss": Infinity,
|
1135 |
+
"eval_runtime": 759.5074,
|
1136 |
+
"eval_samples_per_second": 19.44,
|
1137 |
+
"eval_steps_per_second": 2.431,
|
1138 |
+
"eval_wer": 0.17243709531722992,
|
1139 |
+
"step": 11100
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 4.91,
|
1143 |
+
"eval_loss": Infinity,
|
1144 |
+
"eval_runtime": 757.3653,
|
1145 |
+
"eval_samples_per_second": 19.495,
|
1146 |
+
"eval_steps_per_second": 2.437,
|
1147 |
+
"eval_wer": 0.17223022872549493,
|
1148 |
+
"step": 11200
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 4.96,
|
1152 |
+
"eval_loss": Infinity,
|
1153 |
+
"eval_runtime": 756.4761,
|
1154 |
+
"eval_samples_per_second": 19.518,
|
1155 |
+
"eval_steps_per_second": 2.44,
|
1156 |
+
"eval_wer": 0.17115452244847298,
|
1157 |
+
"step": 11300
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 5.0,
|
1161 |
+
"step": 11395,
|
1162 |
+
"total_flos": 1.0194681891264653e+20,
|
1163 |
+
"train_loss": 0.4200820018974194,
|
1164 |
+
"train_runtime": 124144.1421,
|
1165 |
+
"train_samples_per_second": 5.872,
|
1166 |
+
"train_steps_per_second": 0.092
|
1167 |
+
}
|
1168 |
+
],
|
1169 |
+
"max_steps": 11395,
|
1170 |
+
"num_train_epochs": 5,
|
1171 |
+
"total_flos": 1.0194681891264653e+20,
|
1172 |
+
"trial_name": null,
|
1173 |
+
"trial_params": null
|
1174 |
+
}
|