Commit
•
df2a1e4
1
Parent(s):
d19eb4f
End of training
Browse files- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Jan24_21-45-08_nid007419/events.out.tfevents.1674630701.nid007419.50761.2 +3 -0
- train_results.json +7 -0
- trainer_state.json +541 -0
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 16.13,
|
3 |
+
"eval_loss": 0.3861992359161377,
|
4 |
+
"eval_runtime": 2316.9306,
|
5 |
+
"eval_samples_per_second": 0.555,
|
6 |
+
"eval_steps_per_second": 0.555,
|
7 |
+
"eval_wer": 15.541324622856406,
|
8 |
+
"train_loss": 0.032722438401542604,
|
9 |
+
"train_runtime": 36967.2072,
|
10 |
+
"train_samples_per_second": 1.731,
|
11 |
+
"train_steps_per_second": 0.054
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 16.13,
|
3 |
+
"eval_loss": 0.3861992359161377,
|
4 |
+
"eval_runtime": 2316.9306,
|
5 |
+
"eval_samples_per_second": 0.555,
|
6 |
+
"eval_steps_per_second": 0.555,
|
7 |
+
"eval_wer": 15.541324622856406
|
8 |
+
}
|
runs/Jan24_21-45-08_nid007419/events.out.tfevents.1674630701.nid007419.50761.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b2235ca7f02016cddceb161543af8a1e4504bb6a7661d8acb089c7bc15580120
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 16.13,
|
3 |
+
"train_loss": 0.032722438401542604,
|
4 |
+
"train_runtime": 36967.2072,
|
5 |
+
"train_samples_per_second": 1.731,
|
6 |
+
"train_steps_per_second": 0.054
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,541 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 15.541324622856406,
|
3 |
+
"best_model_checkpoint": "/scratch/project_465000370/asr_children/models/whisper-large-et-clinic/checkpoint-2000",
|
4 |
+
"epoch": 16.12806403201601,
|
5 |
+
"global_step": 2000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.2,
|
12 |
+
"learning_rate": 1.1500000000000002e-06,
|
13 |
+
"loss": 0.495,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.4,
|
18 |
+
"learning_rate": 2.4000000000000003e-06,
|
19 |
+
"loss": 0.2703,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.6,
|
24 |
+
"learning_rate": 3.65e-06,
|
25 |
+
"loss": 0.2068,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.8,
|
30 |
+
"learning_rate": 4.9000000000000005e-06,
|
31 |
+
"loss": 0.2183,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.01,
|
36 |
+
"learning_rate": 6.15e-06,
|
37 |
+
"loss": 0.1988,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.21,
|
42 |
+
"learning_rate": 7.4e-06,
|
43 |
+
"loss": 0.1218,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.41,
|
48 |
+
"learning_rate": 8.65e-06,
|
49 |
+
"loss": 0.1199,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 1.61,
|
54 |
+
"learning_rate": 9.9e-06,
|
55 |
+
"loss": 0.1196,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 1.81,
|
60 |
+
"learning_rate": 9.872222222222223e-06,
|
61 |
+
"loss": 0.1177,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.02,
|
66 |
+
"learning_rate": 9.733333333333334e-06,
|
67 |
+
"loss": 0.1277,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.22,
|
72 |
+
"learning_rate": 9.594444444444445e-06,
|
73 |
+
"loss": 0.0481,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 2.42,
|
78 |
+
"learning_rate": 9.455555555555557e-06,
|
79 |
+
"loss": 0.0542,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 2.62,
|
84 |
+
"learning_rate": 9.316666666666667e-06,
|
85 |
+
"loss": 0.06,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 2.82,
|
90 |
+
"learning_rate": 9.17777777777778e-06,
|
91 |
+
"loss": 0.0543,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 3.02,
|
96 |
+
"learning_rate": 9.038888888888889e-06,
|
97 |
+
"loss": 0.0481,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 3.22,
|
102 |
+
"learning_rate": 8.900000000000001e-06,
|
103 |
+
"loss": 0.0238,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 3.42,
|
108 |
+
"learning_rate": 8.761111111111112e-06,
|
109 |
+
"loss": 0.0236,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 3.62,
|
114 |
+
"learning_rate": 8.627777777777778e-06,
|
115 |
+
"loss": 0.0265,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 3.82,
|
120 |
+
"learning_rate": 8.48888888888889e-06,
|
121 |
+
"loss": 0.0277,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 4.03,
|
126 |
+
"learning_rate": 8.35e-06,
|
127 |
+
"loss": 0.0278,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 4.03,
|
132 |
+
"eval_loss": 0.2985570430755615,
|
133 |
+
"eval_runtime": 2329.8998,
|
134 |
+
"eval_samples_per_second": 0.552,
|
135 |
+
"eval_steps_per_second": 0.552,
|
136 |
+
"eval_wer": 16.99402587355482,
|
137 |
+
"step": 500
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 4.23,
|
141 |
+
"learning_rate": 8.211111111111112e-06,
|
142 |
+
"loss": 0.013,
|
143 |
+
"step": 525
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 4.43,
|
147 |
+
"learning_rate": 8.072222222222223e-06,
|
148 |
+
"loss": 0.0138,
|
149 |
+
"step": 550
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 4.63,
|
153 |
+
"learning_rate": 7.933333333333334e-06,
|
154 |
+
"loss": 0.0165,
|
155 |
+
"step": 575
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 4.83,
|
159 |
+
"learning_rate": 7.794444444444445e-06,
|
160 |
+
"loss": 0.0145,
|
161 |
+
"step": 600
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 5.04,
|
165 |
+
"learning_rate": 7.655555555555556e-06,
|
166 |
+
"loss": 0.0143,
|
167 |
+
"step": 625
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 5.24,
|
171 |
+
"learning_rate": 7.516666666666668e-06,
|
172 |
+
"loss": 0.008,
|
173 |
+
"step": 650
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 5.44,
|
177 |
+
"learning_rate": 7.377777777777778e-06,
|
178 |
+
"loss": 0.0103,
|
179 |
+
"step": 675
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 5.64,
|
183 |
+
"learning_rate": 7.23888888888889e-06,
|
184 |
+
"loss": 0.0081,
|
185 |
+
"step": 700
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 5.84,
|
189 |
+
"learning_rate": 7.100000000000001e-06,
|
190 |
+
"loss": 0.0083,
|
191 |
+
"step": 725
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 6.05,
|
195 |
+
"learning_rate": 6.9611111111111116e-06,
|
196 |
+
"loss": 0.0087,
|
197 |
+
"step": 750
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 6.25,
|
201 |
+
"learning_rate": 6.8222222222222225e-06,
|
202 |
+
"loss": 0.007,
|
203 |
+
"step": 775
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 6.45,
|
207 |
+
"learning_rate": 6.683333333333334e-06,
|
208 |
+
"loss": 0.0054,
|
209 |
+
"step": 800
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 6.65,
|
213 |
+
"learning_rate": 6.544444444444445e-06,
|
214 |
+
"loss": 0.0072,
|
215 |
+
"step": 825
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 6.85,
|
219 |
+
"learning_rate": 6.405555555555556e-06,
|
220 |
+
"loss": 0.0046,
|
221 |
+
"step": 850
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 7.06,
|
225 |
+
"learning_rate": 6.266666666666668e-06,
|
226 |
+
"loss": 0.0053,
|
227 |
+
"step": 875
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 7.26,
|
231 |
+
"learning_rate": 6.127777777777778e-06,
|
232 |
+
"loss": 0.0036,
|
233 |
+
"step": 900
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 7.46,
|
237 |
+
"learning_rate": 5.98888888888889e-06,
|
238 |
+
"loss": 0.0035,
|
239 |
+
"step": 925
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 7.66,
|
243 |
+
"learning_rate": 5.85e-06,
|
244 |
+
"loss": 0.0039,
|
245 |
+
"step": 950
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 7.86,
|
249 |
+
"learning_rate": 5.711111111111112e-06,
|
250 |
+
"loss": 0.004,
|
251 |
+
"step": 975
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 8.06,
|
255 |
+
"learning_rate": 5.572222222222223e-06,
|
256 |
+
"loss": 0.0037,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 8.06,
|
261 |
+
"eval_loss": 0.3382585346698761,
|
262 |
+
"eval_runtime": 2322.4054,
|
263 |
+
"eval_samples_per_second": 0.554,
|
264 |
+
"eval_steps_per_second": 0.554,
|
265 |
+
"eval_wer": 16.17741866162376,
|
266 |
+
"step": 1000
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 8.26,
|
270 |
+
"learning_rate": 5.4333333333333335e-06,
|
271 |
+
"loss": 0.002,
|
272 |
+
"step": 1025
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 8.46,
|
276 |
+
"learning_rate": 5.294444444444445e-06,
|
277 |
+
"loss": 0.0036,
|
278 |
+
"step": 1050
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 8.66,
|
282 |
+
"learning_rate": 5.155555555555556e-06,
|
283 |
+
"loss": 0.0029,
|
284 |
+
"step": 1075
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 8.86,
|
288 |
+
"learning_rate": 5.016666666666667e-06,
|
289 |
+
"loss": 0.0026,
|
290 |
+
"step": 1100
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 9.07,
|
294 |
+
"learning_rate": 4.877777777777778e-06,
|
295 |
+
"loss": 0.0023,
|
296 |
+
"step": 1125
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 9.27,
|
300 |
+
"learning_rate": 4.73888888888889e-06,
|
301 |
+
"loss": 0.0023,
|
302 |
+
"step": 1150
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 9.47,
|
306 |
+
"learning_rate": 4.600000000000001e-06,
|
307 |
+
"loss": 0.0023,
|
308 |
+
"step": 1175
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 9.67,
|
312 |
+
"learning_rate": 4.461111111111112e-06,
|
313 |
+
"loss": 0.0019,
|
314 |
+
"step": 1200
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 9.87,
|
318 |
+
"learning_rate": 4.322222222222223e-06,
|
319 |
+
"loss": 0.002,
|
320 |
+
"step": 1225
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 10.08,
|
324 |
+
"learning_rate": 4.183333333333334e-06,
|
325 |
+
"loss": 0.0017,
|
326 |
+
"step": 1250
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 10.28,
|
330 |
+
"learning_rate": 4.044444444444445e-06,
|
331 |
+
"loss": 0.0014,
|
332 |
+
"step": 1275
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 10.48,
|
336 |
+
"learning_rate": 3.9055555555555555e-06,
|
337 |
+
"loss": 0.0021,
|
338 |
+
"step": 1300
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 10.68,
|
342 |
+
"learning_rate": 3.766666666666667e-06,
|
343 |
+
"loss": 0.0017,
|
344 |
+
"step": 1325
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 10.88,
|
348 |
+
"learning_rate": 3.627777777777778e-06,
|
349 |
+
"loss": 0.0024,
|
350 |
+
"step": 1350
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 11.09,
|
354 |
+
"learning_rate": 3.4888888888888896e-06,
|
355 |
+
"loss": 0.0015,
|
356 |
+
"step": 1375
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 11.29,
|
360 |
+
"learning_rate": 3.3500000000000005e-06,
|
361 |
+
"loss": 0.0012,
|
362 |
+
"step": 1400
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 11.49,
|
366 |
+
"learning_rate": 3.2111111111111115e-06,
|
367 |
+
"loss": 0.0012,
|
368 |
+
"step": 1425
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 11.69,
|
372 |
+
"learning_rate": 3.0722222222222224e-06,
|
373 |
+
"loss": 0.0017,
|
374 |
+
"step": 1450
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 11.89,
|
378 |
+
"learning_rate": 2.9333333333333338e-06,
|
379 |
+
"loss": 0.0022,
|
380 |
+
"step": 1475
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 12.1,
|
384 |
+
"learning_rate": 2.7944444444444447e-06,
|
385 |
+
"loss": 0.0017,
|
386 |
+
"step": 1500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 12.1,
|
390 |
+
"eval_loss": 0.3618781566619873,
|
391 |
+
"eval_runtime": 2327.1444,
|
392 |
+
"eval_samples_per_second": 0.553,
|
393 |
+
"eval_steps_per_second": 0.553,
|
394 |
+
"eval_wer": 15.803498517213221,
|
395 |
+
"step": 1500
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 12.3,
|
399 |
+
"learning_rate": 2.6555555555555556e-06,
|
400 |
+
"loss": 0.0012,
|
401 |
+
"step": 1525
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 12.5,
|
405 |
+
"learning_rate": 2.5166666666666666e-06,
|
406 |
+
"loss": 0.0016,
|
407 |
+
"step": 1550
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 12.7,
|
411 |
+
"learning_rate": 2.377777777777778e-06,
|
412 |
+
"loss": 0.0016,
|
413 |
+
"step": 1575
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 12.9,
|
417 |
+
"learning_rate": 2.238888888888889e-06,
|
418 |
+
"loss": 0.0012,
|
419 |
+
"step": 1600
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 13.1,
|
423 |
+
"learning_rate": 2.1000000000000002e-06,
|
424 |
+
"loss": 0.0012,
|
425 |
+
"step": 1625
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 13.3,
|
429 |
+
"learning_rate": 1.9611111111111116e-06,
|
430 |
+
"loss": 0.0011,
|
431 |
+
"step": 1650
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 13.5,
|
435 |
+
"learning_rate": 1.8222222222222225e-06,
|
436 |
+
"loss": 0.0014,
|
437 |
+
"step": 1675
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 13.7,
|
441 |
+
"learning_rate": 1.6833333333333335e-06,
|
442 |
+
"loss": 0.0012,
|
443 |
+
"step": 1700
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 13.9,
|
447 |
+
"learning_rate": 1.5444444444444446e-06,
|
448 |
+
"loss": 0.0013,
|
449 |
+
"step": 1725
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 14.11,
|
453 |
+
"learning_rate": 1.4055555555555555e-06,
|
454 |
+
"loss": 0.0013,
|
455 |
+
"step": 1750
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 14.31,
|
459 |
+
"learning_rate": 1.2666666666666669e-06,
|
460 |
+
"loss": 0.0012,
|
461 |
+
"step": 1775
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 14.51,
|
465 |
+
"learning_rate": 1.1277777777777778e-06,
|
466 |
+
"loss": 0.001,
|
467 |
+
"step": 1800
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 14.71,
|
471 |
+
"learning_rate": 9.88888888888889e-07,
|
472 |
+
"loss": 0.001,
|
473 |
+
"step": 1825
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 14.91,
|
477 |
+
"learning_rate": 8.500000000000001e-07,
|
478 |
+
"loss": 0.0011,
|
479 |
+
"step": 1850
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 15.12,
|
483 |
+
"learning_rate": 7.111111111111112e-07,
|
484 |
+
"loss": 0.0011,
|
485 |
+
"step": 1875
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 15.32,
|
489 |
+
"learning_rate": 5.722222222222223e-07,
|
490 |
+
"loss": 0.001,
|
491 |
+
"step": 1900
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 15.52,
|
495 |
+
"learning_rate": 4.333333333333334e-07,
|
496 |
+
"loss": 0.0009,
|
497 |
+
"step": 1925
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 15.72,
|
501 |
+
"learning_rate": 2.9444444444444444e-07,
|
502 |
+
"loss": 0.0012,
|
503 |
+
"step": 1950
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 15.92,
|
507 |
+
"learning_rate": 1.5555555555555556e-07,
|
508 |
+
"loss": 0.0012,
|
509 |
+
"step": 1975
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 16.13,
|
513 |
+
"learning_rate": 1.666666666666667e-08,
|
514 |
+
"loss": 0.0008,
|
515 |
+
"step": 2000
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 16.13,
|
519 |
+
"eval_loss": 0.3861992359161377,
|
520 |
+
"eval_runtime": 2318.6816,
|
521 |
+
"eval_samples_per_second": 0.555,
|
522 |
+
"eval_steps_per_second": 0.555,
|
523 |
+
"eval_wer": 15.541324622856406,
|
524 |
+
"step": 2000
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 16.13,
|
528 |
+
"step": 2000,
|
529 |
+
"total_flos": 1.36902067421184e+20,
|
530 |
+
"train_loss": 0.032722438401542604,
|
531 |
+
"train_runtime": 36967.2072,
|
532 |
+
"train_samples_per_second": 1.731,
|
533 |
+
"train_steps_per_second": 0.054
|
534 |
+
}
|
535 |
+
],
|
536 |
+
"max_steps": 2000,
|
537 |
+
"num_train_epochs": 17,
|
538 |
+
"total_flos": 1.36902067421184e+20,
|
539 |
+
"trial_name": null,
|
540 |
+
"trial_params": null
|
541 |
+
}
|