infinitejoy
commited on
Commit
•
21747b1
1
Parent(s):
e7938d6
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 +742 -0
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"eval_loss": 0.11674057692289352,
|
4 |
+
"eval_runtime": 153.2854,
|
5 |
+
"eval_samples": 3172,
|
6 |
+
"eval_samples_per_second": 20.693,
|
7 |
+
"eval_steps_per_second": 20.693,
|
8 |
+
"eval_wer": 0.14212066931713924,
|
9 |
+
"train_loss": 1.0303706246270075,
|
10 |
+
"train_runtime": 15356.7914,
|
11 |
+
"train_samples": 7174,
|
12 |
+
"train_samples_per_second": 23.358,
|
13 |
+
"train_steps_per_second": 0.733
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"eval_loss": 0.11674057692289352,
|
4 |
+
"eval_runtime": 153.2854,
|
5 |
+
"eval_samples": 3172,
|
6 |
+
"eval_samples_per_second": 20.693,
|
7 |
+
"eval_steps_per_second": 20.693,
|
8 |
+
"eval_wer": 0.14212066931713924
|
9 |
+
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262108145
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f443ab6ef2a8ce462c9a5b0263a6ef59a64d1e69d90a60c17ae7e7089b43b22a
|
3 |
size 1262108145
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"train_loss": 1.0303706246270075,
|
4 |
+
"train_runtime": 15356.7914,
|
5 |
+
"train_samples": 7174,
|
6 |
+
"train_samples_per_second": 23.358,
|
7 |
+
"train_steps_per_second": 0.733
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,742 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 50.0,
|
5 |
+
"global_step": 11250,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.44,
|
12 |
+
"learning_rate": 3.4299999999999998e-06,
|
13 |
+
"loss": 12.3928,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.89,
|
18 |
+
"learning_rate": 6.93e-06,
|
19 |
+
"loss": 7.13,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.33,
|
24 |
+
"learning_rate": 1.0429999999999998e-05,
|
25 |
+
"loss": 4.5055,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.78,
|
30 |
+
"learning_rate": 1.3929999999999999e-05,
|
31 |
+
"loss": 3.8494,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 2.22,
|
36 |
+
"learning_rate": 1.7429999999999997e-05,
|
37 |
+
"loss": 3.4447,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 2.67,
|
42 |
+
"learning_rate": 2.0929999999999998e-05,
|
43 |
+
"loss": 3.2021,
|
44 |
+
"step": 600
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 3.11,
|
48 |
+
"learning_rate": 2.4429999999999995e-05,
|
49 |
+
"loss": 3.0761,
|
50 |
+
"step": 700
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 3.56,
|
54 |
+
"learning_rate": 2.793e-05,
|
55 |
+
"loss": 3.0165,
|
56 |
+
"step": 800
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 4.0,
|
60 |
+
"learning_rate": 3.1429999999999996e-05,
|
61 |
+
"loss": 2.9623,
|
62 |
+
"step": 900
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 4.44,
|
66 |
+
"learning_rate": 3.493e-05,
|
67 |
+
"loss": 2.9094,
|
68 |
+
"step": 1000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 4.89,
|
72 |
+
"learning_rate": 3.843e-05,
|
73 |
+
"loss": 2.8831,
|
74 |
+
"step": 1100
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 5.33,
|
78 |
+
"learning_rate": 4.192999999999999e-05,
|
79 |
+
"loss": 2.8667,
|
80 |
+
"step": 1200
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 5.78,
|
84 |
+
"learning_rate": 4.543e-05,
|
85 |
+
"loss": 2.7722,
|
86 |
+
"step": 1300
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 6.22,
|
90 |
+
"learning_rate": 4.8929999999999994e-05,
|
91 |
+
"loss": 2.2496,
|
92 |
+
"step": 1400
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 6.67,
|
96 |
+
"learning_rate": 5.243e-05,
|
97 |
+
"loss": 1.7213,
|
98 |
+
"step": 1500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 7.11,
|
102 |
+
"learning_rate": 5.593e-05,
|
103 |
+
"loss": 1.5195,
|
104 |
+
"step": 1600
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 7.56,
|
108 |
+
"learning_rate": 5.942999999999999e-05,
|
109 |
+
"loss": 1.3847,
|
110 |
+
"step": 1700
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 8.0,
|
114 |
+
"learning_rate": 6.293e-05,
|
115 |
+
"loss": 1.3296,
|
116 |
+
"step": 1800
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 8.44,
|
120 |
+
"learning_rate": 6.642999999999999e-05,
|
121 |
+
"loss": 1.2533,
|
122 |
+
"step": 1900
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 8.89,
|
126 |
+
"learning_rate": 6.992999999999999e-05,
|
127 |
+
"loss": 1.1973,
|
128 |
+
"step": 2000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 8.89,
|
132 |
+
"eval_loss": 0.44813069701194763,
|
133 |
+
"eval_runtime": 151.5294,
|
134 |
+
"eval_samples_per_second": 20.933,
|
135 |
+
"eval_steps_per_second": 20.933,
|
136 |
+
"eval_wer": 0.4849491021888243,
|
137 |
+
"step": 2000
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 9.33,
|
141 |
+
"learning_rate": 6.925837837837837e-05,
|
142 |
+
"loss": 1.08,
|
143 |
+
"step": 2100
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 9.78,
|
147 |
+
"learning_rate": 6.850162162162162e-05,
|
148 |
+
"loss": 0.9484,
|
149 |
+
"step": 2200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 10.22,
|
153 |
+
"learning_rate": 6.774486486486486e-05,
|
154 |
+
"loss": 0.8818,
|
155 |
+
"step": 2300
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 10.67,
|
159 |
+
"learning_rate": 6.698810810810811e-05,
|
160 |
+
"loss": 0.8043,
|
161 |
+
"step": 2400
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 11.11,
|
165 |
+
"learning_rate": 6.623135135135134e-05,
|
166 |
+
"loss": 0.7752,
|
167 |
+
"step": 2500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 11.56,
|
171 |
+
"learning_rate": 6.547459459459459e-05,
|
172 |
+
"loss": 0.7497,
|
173 |
+
"step": 2600
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 12.0,
|
177 |
+
"learning_rate": 6.471783783783783e-05,
|
178 |
+
"loss": 0.7486,
|
179 |
+
"step": 2700
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 12.44,
|
183 |
+
"learning_rate": 6.396108108108108e-05,
|
184 |
+
"loss": 0.7016,
|
185 |
+
"step": 2800
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 12.89,
|
189 |
+
"learning_rate": 6.320432432432433e-05,
|
190 |
+
"loss": 0.6927,
|
191 |
+
"step": 2900
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 13.33,
|
195 |
+
"learning_rate": 6.244756756756756e-05,
|
196 |
+
"loss": 0.6609,
|
197 |
+
"step": 3000
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 13.78,
|
201 |
+
"learning_rate": 6.169081081081081e-05,
|
202 |
+
"loss": 0.6633,
|
203 |
+
"step": 3100
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 14.22,
|
207 |
+
"learning_rate": 6.093405405405405e-05,
|
208 |
+
"loss": 0.6781,
|
209 |
+
"step": 3200
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 14.67,
|
213 |
+
"learning_rate": 6.017729729729729e-05,
|
214 |
+
"loss": 0.6486,
|
215 |
+
"step": 3300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 15.11,
|
219 |
+
"learning_rate": 5.942054054054054e-05,
|
220 |
+
"loss": 0.6217,
|
221 |
+
"step": 3400
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 15.56,
|
225 |
+
"learning_rate": 5.866378378378378e-05,
|
226 |
+
"loss": 0.6348,
|
227 |
+
"step": 3500
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 16.0,
|
231 |
+
"learning_rate": 5.7907027027027026e-05,
|
232 |
+
"loss": 0.6555,
|
233 |
+
"step": 3600
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 16.44,
|
237 |
+
"learning_rate": 5.715027027027027e-05,
|
238 |
+
"loss": 0.6179,
|
239 |
+
"step": 3700
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 16.89,
|
243 |
+
"learning_rate": 5.639351351351351e-05,
|
244 |
+
"loss": 0.6116,
|
245 |
+
"step": 3800
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 17.33,
|
249 |
+
"learning_rate": 5.5636756756756754e-05,
|
250 |
+
"loss": 0.586,
|
251 |
+
"step": 3900
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 17.78,
|
255 |
+
"learning_rate": 5.4879999999999996e-05,
|
256 |
+
"loss": 0.6005,
|
257 |
+
"step": 4000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 17.78,
|
261 |
+
"eval_loss": 0.1420038342475891,
|
262 |
+
"eval_runtime": 156.5701,
|
263 |
+
"eval_samples_per_second": 20.259,
|
264 |
+
"eval_steps_per_second": 20.259,
|
265 |
+
"eval_wer": 0.17772729258595832,
|
266 |
+
"step": 4000
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 18.22,
|
270 |
+
"learning_rate": 5.412324324324324e-05,
|
271 |
+
"loss": 0.6199,
|
272 |
+
"step": 4100
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 18.67,
|
276 |
+
"learning_rate": 5.336648648648648e-05,
|
277 |
+
"loss": 0.6017,
|
278 |
+
"step": 4200
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 19.11,
|
282 |
+
"learning_rate": 5.2609729729729724e-05,
|
283 |
+
"loss": 0.5722,
|
284 |
+
"step": 4300
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 19.56,
|
288 |
+
"learning_rate": 5.1852972972972974e-05,
|
289 |
+
"loss": 0.5755,
|
290 |
+
"step": 4400
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 20.0,
|
294 |
+
"learning_rate": 5.1096216216216216e-05,
|
295 |
+
"loss": 0.6083,
|
296 |
+
"step": 4500
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 20.44,
|
300 |
+
"learning_rate": 5.033945945945946e-05,
|
301 |
+
"loss": 0.5677,
|
302 |
+
"step": 4600
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 20.89,
|
306 |
+
"learning_rate": 4.9590270270270266e-05,
|
307 |
+
"loss": 0.5652,
|
308 |
+
"step": 4700
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 21.33,
|
312 |
+
"learning_rate": 4.8833513513513516e-05,
|
313 |
+
"loss": 0.5509,
|
314 |
+
"step": 4800
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 21.78,
|
318 |
+
"learning_rate": 4.807675675675676e-05,
|
319 |
+
"loss": 0.5526,
|
320 |
+
"step": 4900
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 22.22,
|
324 |
+
"learning_rate": 4.732e-05,
|
325 |
+
"loss": 0.5589,
|
326 |
+
"step": 5000
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 22.67,
|
330 |
+
"learning_rate": 4.6563243243243244e-05,
|
331 |
+
"loss": 0.5394,
|
332 |
+
"step": 5100
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 23.11,
|
336 |
+
"learning_rate": 4.5806486486486486e-05,
|
337 |
+
"loss": 0.5329,
|
338 |
+
"step": 5200
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 23.56,
|
342 |
+
"learning_rate": 4.504972972972973e-05,
|
343 |
+
"loss": 0.5353,
|
344 |
+
"step": 5300
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 24.0,
|
348 |
+
"learning_rate": 4.429297297297297e-05,
|
349 |
+
"loss": 0.5643,
|
350 |
+
"step": 5400
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 24.44,
|
354 |
+
"learning_rate": 4.3536216216216214e-05,
|
355 |
+
"loss": 0.537,
|
356 |
+
"step": 5500
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 24.89,
|
360 |
+
"learning_rate": 4.277945945945946e-05,
|
361 |
+
"loss": 0.5502,
|
362 |
+
"step": 5600
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 25.33,
|
366 |
+
"learning_rate": 4.20227027027027e-05,
|
367 |
+
"loss": 0.5126,
|
368 |
+
"step": 5700
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 25.78,
|
372 |
+
"learning_rate": 4.126594594594594e-05,
|
373 |
+
"loss": 0.5315,
|
374 |
+
"step": 5800
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 26.22,
|
378 |
+
"learning_rate": 4.050918918918919e-05,
|
379 |
+
"loss": 0.5424,
|
380 |
+
"step": 5900
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 26.67,
|
384 |
+
"learning_rate": 3.9752432432432434e-05,
|
385 |
+
"loss": 0.5248,
|
386 |
+
"step": 6000
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 26.67,
|
390 |
+
"eval_loss": 0.13026614487171173,
|
391 |
+
"eval_runtime": 153.4664,
|
392 |
+
"eval_samples_per_second": 20.669,
|
393 |
+
"eval_steps_per_second": 20.669,
|
394 |
+
"eval_wer": 0.16505745117742146,
|
395 |
+
"step": 6000
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 27.11,
|
399 |
+
"learning_rate": 3.8995675675675676e-05,
|
400 |
+
"loss": 0.5111,
|
401 |
+
"step": 6100
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 27.56,
|
405 |
+
"learning_rate": 3.823891891891892e-05,
|
406 |
+
"loss": 0.5226,
|
407 |
+
"step": 6200
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 28.0,
|
411 |
+
"learning_rate": 3.748216216216216e-05,
|
412 |
+
"loss": 0.5335,
|
413 |
+
"step": 6300
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 28.44,
|
417 |
+
"learning_rate": 3.6725405405405404e-05,
|
418 |
+
"loss": 0.5031,
|
419 |
+
"step": 6400
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 28.89,
|
423 |
+
"learning_rate": 3.596864864864865e-05,
|
424 |
+
"loss": 0.5219,
|
425 |
+
"step": 6500
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 29.33,
|
429 |
+
"learning_rate": 3.521189189189189e-05,
|
430 |
+
"loss": 0.4853,
|
431 |
+
"step": 6600
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 29.78,
|
435 |
+
"learning_rate": 3.445513513513513e-05,
|
436 |
+
"loss": 0.5062,
|
437 |
+
"step": 6700
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 30.22,
|
441 |
+
"learning_rate": 3.370594594594594e-05,
|
442 |
+
"loss": 0.5395,
|
443 |
+
"step": 6800
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 30.67,
|
447 |
+
"learning_rate": 3.294918918918919e-05,
|
448 |
+
"loss": 0.4876,
|
449 |
+
"step": 6900
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 31.11,
|
453 |
+
"learning_rate": 3.219243243243243e-05,
|
454 |
+
"loss": 0.4981,
|
455 |
+
"step": 7000
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 31.56,
|
459 |
+
"learning_rate": 3.1435675675675674e-05,
|
460 |
+
"loss": 0.5011,
|
461 |
+
"step": 7100
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 32.0,
|
465 |
+
"learning_rate": 3.067891891891892e-05,
|
466 |
+
"loss": 0.511,
|
467 |
+
"step": 7200
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 32.44,
|
471 |
+
"learning_rate": 2.992216216216216e-05,
|
472 |
+
"loss": 0.4935,
|
473 |
+
"step": 7300
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 32.89,
|
477 |
+
"learning_rate": 2.9165405405405402e-05,
|
478 |
+
"loss": 0.4951,
|
479 |
+
"step": 7400
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 33.33,
|
483 |
+
"learning_rate": 2.8408648648648645e-05,
|
484 |
+
"loss": 0.4655,
|
485 |
+
"step": 7500
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 33.78,
|
489 |
+
"learning_rate": 2.765189189189189e-05,
|
490 |
+
"loss": 0.4926,
|
491 |
+
"step": 7600
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 34.22,
|
495 |
+
"learning_rate": 2.6895135135135133e-05,
|
496 |
+
"loss": 0.5083,
|
497 |
+
"step": 7700
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 34.67,
|
501 |
+
"learning_rate": 2.6138378378378376e-05,
|
502 |
+
"loss": 0.4849,
|
503 |
+
"step": 7800
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 35.11,
|
507 |
+
"learning_rate": 2.538162162162162e-05,
|
508 |
+
"loss": 0.4673,
|
509 |
+
"step": 7900
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 35.56,
|
513 |
+
"learning_rate": 2.462486486486486e-05,
|
514 |
+
"loss": 0.4871,
|
515 |
+
"step": 8000
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 35.56,
|
519 |
+
"eval_loss": 0.12074683606624603,
|
520 |
+
"eval_runtime": 154.3437,
|
521 |
+
"eval_samples_per_second": 20.552,
|
522 |
+
"eval_steps_per_second": 20.552,
|
523 |
+
"eval_wer": 0.1523439206605793,
|
524 |
+
"step": 8000
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 36.0,
|
528 |
+
"learning_rate": 2.3875675675675676e-05,
|
529 |
+
"loss": 0.4911,
|
530 |
+
"step": 8100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 36.44,
|
534 |
+
"learning_rate": 2.3118918918918918e-05,
|
535 |
+
"loss": 0.4724,
|
536 |
+
"step": 8200
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 36.89,
|
540 |
+
"learning_rate": 2.236216216216216e-05,
|
541 |
+
"loss": 0.4784,
|
542 |
+
"step": 8300
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 37.33,
|
546 |
+
"learning_rate": 2.1605405405405403e-05,
|
547 |
+
"loss": 0.466,
|
548 |
+
"step": 8400
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 37.78,
|
552 |
+
"learning_rate": 2.0848648648648646e-05,
|
553 |
+
"loss": 0.4761,
|
554 |
+
"step": 8500
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 38.22,
|
558 |
+
"learning_rate": 2.0091891891891892e-05,
|
559 |
+
"loss": 0.4772,
|
560 |
+
"step": 8600
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 38.67,
|
564 |
+
"learning_rate": 1.9335135135135135e-05,
|
565 |
+
"loss": 0.4524,
|
566 |
+
"step": 8700
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 39.11,
|
570 |
+
"learning_rate": 1.8578378378378377e-05,
|
571 |
+
"loss": 0.4436,
|
572 |
+
"step": 8800
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 39.56,
|
576 |
+
"learning_rate": 1.782162162162162e-05,
|
577 |
+
"loss": 0.4673,
|
578 |
+
"step": 8900
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 40.0,
|
582 |
+
"learning_rate": 1.7064864864864862e-05,
|
583 |
+
"loss": 0.4848,
|
584 |
+
"step": 9000
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 40.44,
|
588 |
+
"learning_rate": 1.630810810810811e-05,
|
589 |
+
"loss": 0.461,
|
590 |
+
"step": 9100
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 40.89,
|
594 |
+
"learning_rate": 1.555135135135135e-05,
|
595 |
+
"loss": 0.465,
|
596 |
+
"step": 9200
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 41.33,
|
600 |
+
"learning_rate": 1.4794594594594594e-05,
|
601 |
+
"loss": 0.4398,
|
602 |
+
"step": 9300
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 41.78,
|
606 |
+
"learning_rate": 1.4045405405405405e-05,
|
607 |
+
"loss": 0.4552,
|
608 |
+
"step": 9400
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 42.22,
|
612 |
+
"learning_rate": 1.3288648648648647e-05,
|
613 |
+
"loss": 0.47,
|
614 |
+
"step": 9500
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 42.67,
|
618 |
+
"learning_rate": 1.253189189189189e-05,
|
619 |
+
"loss": 0.4599,
|
620 |
+
"step": 9600
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 43.11,
|
624 |
+
"learning_rate": 1.1775135135135134e-05,
|
625 |
+
"loss": 0.4273,
|
626 |
+
"step": 9700
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 43.56,
|
630 |
+
"learning_rate": 1.1018378378378377e-05,
|
631 |
+
"loss": 0.4533,
|
632 |
+
"step": 9800
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 44.0,
|
636 |
+
"learning_rate": 1.0261621621621621e-05,
|
637 |
+
"loss": 0.4573,
|
638 |
+
"step": 9900
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 44.44,
|
642 |
+
"learning_rate": 9.504864864864864e-06,
|
643 |
+
"loss": 0.4428,
|
644 |
+
"step": 10000
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 44.44,
|
648 |
+
"eval_loss": 0.11431078612804413,
|
649 |
+
"eval_runtime": 152.8495,
|
650 |
+
"eval_samples_per_second": 20.752,
|
651 |
+
"eval_steps_per_second": 20.752,
|
652 |
+
"eval_wer": 0.14247018218358162,
|
653 |
+
"step": 10000
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 44.89,
|
657 |
+
"learning_rate": 8.748108108108106e-06,
|
658 |
+
"loss": 0.4431,
|
659 |
+
"step": 10100
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 45.33,
|
663 |
+
"learning_rate": 7.99135135135135e-06,
|
664 |
+
"loss": 0.4124,
|
665 |
+
"step": 10200
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 45.78,
|
669 |
+
"learning_rate": 7.234594594594593e-06,
|
670 |
+
"loss": 0.4437,
|
671 |
+
"step": 10300
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 46.22,
|
675 |
+
"learning_rate": 6.4778378378378375e-06,
|
676 |
+
"loss": 0.4694,
|
677 |
+
"step": 10400
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 46.67,
|
681 |
+
"learning_rate": 5.721081081081081e-06,
|
682 |
+
"loss": 0.4408,
|
683 |
+
"step": 10500
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 47.11,
|
687 |
+
"learning_rate": 4.9643243243243245e-06,
|
688 |
+
"loss": 0.428,
|
689 |
+
"step": 10600
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 47.56,
|
693 |
+
"learning_rate": 4.207567567567567e-06,
|
694 |
+
"loss": 0.4418,
|
695 |
+
"step": 10700
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 48.0,
|
699 |
+
"learning_rate": 3.4508108108108105e-06,
|
700 |
+
"loss": 0.4527,
|
701 |
+
"step": 10800
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 48.44,
|
705 |
+
"learning_rate": 2.6940540540540536e-06,
|
706 |
+
"loss": 0.448,
|
707 |
+
"step": 10900
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 48.89,
|
711 |
+
"learning_rate": 1.937297297297297e-06,
|
712 |
+
"loss": 0.4399,
|
713 |
+
"step": 11000
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 49.33,
|
717 |
+
"learning_rate": 1.1805405405405403e-06,
|
718 |
+
"loss": 0.4111,
|
719 |
+
"step": 11100
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 49.78,
|
723 |
+
"learning_rate": 4.237837837837838e-07,
|
724 |
+
"loss": 0.4214,
|
725 |
+
"step": 11200
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 50.0,
|
729 |
+
"step": 11250,
|
730 |
+
"total_flos": 4.148416605366081e+19,
|
731 |
+
"train_loss": 1.0303706246270075,
|
732 |
+
"train_runtime": 15356.7914,
|
733 |
+
"train_samples_per_second": 23.358,
|
734 |
+
"train_steps_per_second": 0.733
|
735 |
+
}
|
736 |
+
],
|
737 |
+
"max_steps": 11250,
|
738 |
+
"num_train_epochs": 50,
|
739 |
+
"total_flos": 4.148416605366081e+19,
|
740 |
+
"trial_name": null,
|
741 |
+
"trial_params": null
|
742 |
+
}
|