patrickvonplaten
commited on
Commit
•
0307e54
1
Parent(s):
26bb62f
End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.99,
|
3 |
+
"eval_loss": 0.6482358574867249,
|
4 |
+
"eval_runtime": 36.4371,
|
5 |
+
"eval_samples": 4620,
|
6 |
+
"eval_samples_per_second": 126.794,
|
7 |
+
"eval_steps_per_second": 2.003,
|
8 |
+
"eval_wer": 0.6389279813987639,
|
9 |
+
"train_loss": 3.356673427847929,
|
10 |
+
"train_runtime": 4577.6572,
|
11 |
+
"train_samples": 11030,
|
12 |
+
"train_samples_per_second": 120.476,
|
13 |
+
"train_steps_per_second": 0.47
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.99,
|
3 |
+
"eval_loss": 0.6482358574867249,
|
4 |
+
"eval_runtime": 36.4371,
|
5 |
+
"eval_samples": 4620,
|
6 |
+
"eval_samples_per_second": 126.794,
|
7 |
+
"eval_steps_per_second": 2.003,
|
8 |
+
"eval_wer": 0.6389279813987639
|
9 |
+
}
|
runs/Feb02_17-57-08_ip-172-31-1-149/events.out.tfevents.1643829611.ip-172-31-1-149.13347.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:198beece3d1d9f9677ed816111bf850be48a2b0725d690a4f0d3887da8a25fab
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.99,
|
3 |
+
"train_loss": 3.356673427847929,
|
4 |
+
"train_runtime": 4577.6572,
|
5 |
+
"train_samples": 11030,
|
6 |
+
"train_samples_per_second": 120.476,
|
7 |
+
"train_steps_per_second": 0.47
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,187 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 49.994219653179194,
|
5 |
+
"global_step": 2150,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 2.32,
|
12 |
+
"learning_rate": 3.7125e-06,
|
13 |
+
"loss": 14.869,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 4.65,
|
18 |
+
"learning_rate": 7.4625e-06,
|
19 |
+
"loss": 6.701,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 6.97,
|
24 |
+
"learning_rate": 1.1212499999999998e-05,
|
25 |
+
"loss": 4.3877,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 9.3,
|
30 |
+
"learning_rate": 1.49625e-05,
|
31 |
+
"loss": 3.788,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 11.62,
|
36 |
+
"learning_rate": 1.8712499999999997e-05,
|
37 |
+
"loss": 3.3873,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 11.62,
|
42 |
+
"eval_loss": 3.3324315547943115,
|
43 |
+
"eval_runtime": 36.9219,
|
44 |
+
"eval_samples_per_second": 125.129,
|
45 |
+
"eval_steps_per_second": 1.977,
|
46 |
+
"eval_wer": 1.0,
|
47 |
+
"step": 500
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 13.95,
|
51 |
+
"learning_rate": 2.2462499999999997e-05,
|
52 |
+
"loss": 3.1689,
|
53 |
+
"step": 600
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 16.28,
|
57 |
+
"learning_rate": 2.6212499999999997e-05,
|
58 |
+
"loss": 3.0785,
|
59 |
+
"step": 700
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 18.6,
|
63 |
+
"learning_rate": 2.99625e-05,
|
64 |
+
"loss": 2.9982,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 20.92,
|
69 |
+
"learning_rate": 3.37125e-05,
|
70 |
+
"loss": 2.9572,
|
71 |
+
"step": 900
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 23.25,
|
75 |
+
"learning_rate": 3.7462499999999996e-05,
|
76 |
+
"loss": 2.9207,
|
77 |
+
"step": 1000
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 23.25,
|
81 |
+
"eval_loss": 2.8923747539520264,
|
82 |
+
"eval_runtime": 36.3858,
|
83 |
+
"eval_samples_per_second": 126.972,
|
84 |
+
"eval_steps_per_second": 2.006,
|
85 |
+
"eval_wer": 0.9987150461971487,
|
86 |
+
"step": 1000
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 25.58,
|
90 |
+
"learning_rate": 4.12125e-05,
|
91 |
+
"loss": 2.8705,
|
92 |
+
"step": 1100
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 27.9,
|
96 |
+
"learning_rate": 4.4962499999999995e-05,
|
97 |
+
"loss": 2.8024,
|
98 |
+
"step": 1200
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 30.23,
|
102 |
+
"learning_rate": 4.8675e-05,
|
103 |
+
"loss": 2.6545,
|
104 |
+
"step": 1300
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 32.55,
|
108 |
+
"learning_rate": 5.2424999999999994e-05,
|
109 |
+
"loss": 2.4772,
|
110 |
+
"step": 1400
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 34.88,
|
114 |
+
"learning_rate": 5.61375e-05,
|
115 |
+
"loss": 2.1463,
|
116 |
+
"step": 1500
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 34.88,
|
120 |
+
"eval_loss": 1.3062961101531982,
|
121 |
+
"eval_runtime": 36.8007,
|
122 |
+
"eval_samples_per_second": 125.541,
|
123 |
+
"eval_steps_per_second": 1.984,
|
124 |
+
"eval_wer": 0.9138775010707948,
|
125 |
+
"step": 1500
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 37.21,
|
129 |
+
"learning_rate": 5.988749999999999e-05,
|
130 |
+
"loss": 1.911,
|
131 |
+
"step": 1600
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 39.53,
|
135 |
+
"learning_rate": 6.36375e-05,
|
136 |
+
"loss": 1.7907,
|
137 |
+
"step": 1700
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 41.86,
|
141 |
+
"learning_rate": 6.738749999999999e-05,
|
142 |
+
"loss": 1.7065,
|
143 |
+
"step": 1800
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 44.18,
|
147 |
+
"learning_rate": 7.11375e-05,
|
148 |
+
"loss": 1.6557,
|
149 |
+
"step": 1900
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 46.51,
|
153 |
+
"learning_rate": 7.48875e-05,
|
154 |
+
"loss": 1.607,
|
155 |
+
"step": 2000
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 46.51,
|
159 |
+
"eval_loss": 0.7204738855361938,
|
160 |
+
"eval_runtime": 36.377,
|
161 |
+
"eval_samples_per_second": 127.003,
|
162 |
+
"eval_steps_per_second": 2.007,
|
163 |
+
"eval_wer": 0.6856146362356973,
|
164 |
+
"step": 2000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 48.83,
|
168 |
+
"learning_rate": 2.6499999999999997e-05,
|
169 |
+
"loss": 1.5509,
|
170 |
+
"step": 2100
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 49.99,
|
174 |
+
"step": 2150,
|
175 |
+
"total_flos": 6.648469351402203e+19,
|
176 |
+
"train_loss": 3.356673427847929,
|
177 |
+
"train_runtime": 4577.6572,
|
178 |
+
"train_samples_per_second": 120.476,
|
179 |
+
"train_steps_per_second": 0.47
|
180 |
+
}
|
181 |
+
],
|
182 |
+
"max_steps": 2150,
|
183 |
+
"num_train_epochs": 50,
|
184 |
+
"total_flos": 6.648469351402203e+19,
|
185 |
+
"trial_name": null,
|
186 |
+
"trial_params": null
|
187 |
+
}
|