jlondonobo
commited on
Commit
•
4f5d0d8
1
Parent(s):
b1970c3
End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.28,
|
3 |
+
"eval_loss": 0.1502685546875,
|
4 |
+
"eval_runtime": 2289.764,
|
5 |
+
"eval_samples_per_second": 3.796,
|
6 |
+
"eval_steps_per_second": 0.238,
|
7 |
+
"eval_wer": 4.8385198634858195,
|
8 |
+
"train_loss": 0.15757004801432292,
|
9 |
+
"train_runtime": 18540.0869,
|
10 |
+
"train_samples_per_second": 2.589,
|
11 |
+
"train_steps_per_second": 0.081
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.28,
|
3 |
+
"eval_loss": 0.1502685546875,
|
4 |
+
"eval_runtime": 2289.764,
|
5 |
+
"eval_samples_per_second": 3.796,
|
6 |
+
"eval_steps_per_second": 0.238,
|
7 |
+
"eval_wer": 4.8385198634858195
|
8 |
+
}
|
runs/Dec17_19-28-09_193-122-153-4/events.out.tfevents.1671326268.193-122-153-4.2618727.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7d8f1bd25e7b2b27ee8e2b06017471b50ad612d6abddd9799548ffb0ee8d6e7c
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.28,
|
3 |
+
"train_loss": 0.15757004801432292,
|
4 |
+
"train_runtime": 18540.0869,
|
5 |
+
"train_samples_per_second": 2.589,
|
6 |
+
"train_steps_per_second": 0.081
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,412 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 4.8385198634858195,
|
3 |
+
"best_model_checkpoint": "./checkpoint-1500",
|
4 |
+
"epoch": 2.276,
|
5 |
+
"global_step": 1500,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.02,
|
12 |
+
"learning_rate": 5.337725362513818e-07,
|
13 |
+
"loss": 1.0338,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.03,
|
18 |
+
"learning_rate": 6.712460395362235e-07,
|
19 |
+
"loss": 0.6138,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.05,
|
24 |
+
"learning_rate": 7.448557621064326e-07,
|
25 |
+
"loss": 0.3242,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.07,
|
30 |
+
"learning_rate": 7.98395961247633e-07,
|
31 |
+
"loss": 0.2734,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.08,
|
36 |
+
"learning_rate": 8.393538435335606e-07,
|
37 |
+
"loss": 0.2426,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.1,
|
42 |
+
"learning_rate": 8.725321774688733e-07,
|
43 |
+
"loss": 0.2288,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.12,
|
48 |
+
"learning_rate": 9.004197582692396e-07,
|
49 |
+
"loss": 0.2202,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.13,
|
54 |
+
"learning_rate": 9.244741681495453e-07,
|
55 |
+
"loss": 0.1864,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.15,
|
60 |
+
"learning_rate": 9.456229388687313e-07,
|
61 |
+
"loss": 0.2019,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.17,
|
66 |
+
"learning_rate": 9.644929895185596e-07,
|
67 |
+
"loss": 0.1834,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.18,
|
72 |
+
"learning_rate": 9.815279569082485e-07,
|
73 |
+
"loss": 0.1815,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.2,
|
78 |
+
"learning_rate": 9.970533381328862e-07,
|
79 |
+
"loss": 0.1761,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.22,
|
84 |
+
"learning_rate": 9.841666666666666e-07,
|
85 |
+
"loss": 0.185,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.23,
|
90 |
+
"learning_rate": 9.633333333333334e-07,
|
91 |
+
"loss": 0.1538,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.25,
|
96 |
+
"learning_rate": 9.425e-07,
|
97 |
+
"loss": 0.1623,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.27,
|
102 |
+
"learning_rate": 9.216666666666666e-07,
|
103 |
+
"loss": 0.1569,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.28,
|
108 |
+
"learning_rate": 9.008333333333333e-07,
|
109 |
+
"loss": 0.1465,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.3,
|
114 |
+
"learning_rate": 8.799999999999999e-07,
|
115 |
+
"loss": 0.1508,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.32,
|
120 |
+
"learning_rate": 8.591666666666666e-07,
|
121 |
+
"loss": 0.1509,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.33,
|
126 |
+
"learning_rate": 8.383333333333334e-07,
|
127 |
+
"loss": 0.1526,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.33,
|
132 |
+
"eval_loss": 0.1588134765625,
|
133 |
+
"eval_runtime": 2273.9706,
|
134 |
+
"eval_samples_per_second": 3.823,
|
135 |
+
"eval_steps_per_second": 0.239,
|
136 |
+
"eval_wer": 4.907449437635548,
|
137 |
+
"step": 500
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.35,
|
141 |
+
"learning_rate": 8.175e-07,
|
142 |
+
"loss": 0.1533,
|
143 |
+
"step": 525
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 1.0,
|
147 |
+
"learning_rate": 7.966666666666666e-07,
|
148 |
+
"loss": 0.1471,
|
149 |
+
"step": 550
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.02,
|
153 |
+
"learning_rate": 7.758333333333334e-07,
|
154 |
+
"loss": 0.1451,
|
155 |
+
"step": 575
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.04,
|
159 |
+
"learning_rate": 7.55e-07,
|
160 |
+
"loss": 0.1406,
|
161 |
+
"step": 600
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.05,
|
165 |
+
"learning_rate": 7.341666666666666e-07,
|
166 |
+
"loss": 0.1306,
|
167 |
+
"step": 625
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.07,
|
171 |
+
"learning_rate": 7.133333333333333e-07,
|
172 |
+
"loss": 0.1226,
|
173 |
+
"step": 650
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 1.09,
|
177 |
+
"learning_rate": 6.924999999999999e-07,
|
178 |
+
"loss": 0.1152,
|
179 |
+
"step": 675
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.1,
|
183 |
+
"learning_rate": 6.716666666666666e-07,
|
184 |
+
"loss": 0.1251,
|
185 |
+
"step": 700
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.12,
|
189 |
+
"learning_rate": 6.508333333333334e-07,
|
190 |
+
"loss": 0.1169,
|
191 |
+
"step": 725
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.14,
|
195 |
+
"learning_rate": 6.3e-07,
|
196 |
+
"loss": 0.1092,
|
197 |
+
"step": 750
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 1.15,
|
201 |
+
"learning_rate": 6.091666666666666e-07,
|
202 |
+
"loss": 0.13,
|
203 |
+
"step": 775
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.17,
|
207 |
+
"learning_rate": 5.883333333333333e-07,
|
208 |
+
"loss": 0.1177,
|
209 |
+
"step": 800
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.19,
|
213 |
+
"learning_rate": 5.675e-07,
|
214 |
+
"loss": 0.1127,
|
215 |
+
"step": 825
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 1.2,
|
219 |
+
"learning_rate": 5.474999999999999e-07,
|
220 |
+
"loss": 0.1114,
|
221 |
+
"step": 850
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.22,
|
225 |
+
"learning_rate": 5.266666666666666e-07,
|
226 |
+
"loss": 0.1139,
|
227 |
+
"step": 875
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.24,
|
231 |
+
"learning_rate": 5.058333333333333e-07,
|
232 |
+
"loss": 0.1096,
|
233 |
+
"step": 900
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 1.25,
|
237 |
+
"learning_rate": 4.85e-07,
|
238 |
+
"loss": 0.1021,
|
239 |
+
"step": 925
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 1.27,
|
243 |
+
"learning_rate": 4.6416666666666663e-07,
|
244 |
+
"loss": 0.1037,
|
245 |
+
"step": 950
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 1.29,
|
249 |
+
"learning_rate": 4.4333333333333336e-07,
|
250 |
+
"loss": 0.1053,
|
251 |
+
"step": 975
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 1.3,
|
255 |
+
"learning_rate": 4.225e-07,
|
256 |
+
"loss": 0.1046,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.3,
|
261 |
+
"eval_loss": 0.1510009765625,
|
262 |
+
"eval_runtime": 2297.9318,
|
263 |
+
"eval_samples_per_second": 3.783,
|
264 |
+
"eval_steps_per_second": 0.237,
|
265 |
+
"eval_wer": 4.8805500916258975,
|
266 |
+
"step": 1000
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.32,
|
270 |
+
"learning_rate": 4.0166666666666665e-07,
|
271 |
+
"loss": 0.1072,
|
272 |
+
"step": 1025
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 1.34,
|
276 |
+
"learning_rate": 3.808333333333333e-07,
|
277 |
+
"loss": 0.1089,
|
278 |
+
"step": 1050
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 1.35,
|
282 |
+
"learning_rate": 3.6e-07,
|
283 |
+
"loss": 0.1082,
|
284 |
+
"step": 1075
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 2.01,
|
288 |
+
"learning_rate": 3.3916666666666667e-07,
|
289 |
+
"loss": 0.1089,
|
290 |
+
"step": 1100
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 2.03,
|
294 |
+
"learning_rate": 3.1833333333333334e-07,
|
295 |
+
"loss": 0.1109,
|
296 |
+
"step": 1125
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 2.04,
|
300 |
+
"learning_rate": 2.9749999999999996e-07,
|
301 |
+
"loss": 0.0989,
|
302 |
+
"step": 1150
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 2.06,
|
306 |
+
"learning_rate": 2.766666666666667e-07,
|
307 |
+
"loss": 0.0959,
|
308 |
+
"step": 1175
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 2.08,
|
312 |
+
"learning_rate": 2.5583333333333335e-07,
|
313 |
+
"loss": 0.0985,
|
314 |
+
"step": 1200
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 2.09,
|
318 |
+
"learning_rate": 2.3499999999999997e-07,
|
319 |
+
"loss": 0.0862,
|
320 |
+
"step": 1225
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 2.11,
|
324 |
+
"learning_rate": 2.1416666666666667e-07,
|
325 |
+
"loss": 0.092,
|
326 |
+
"step": 1250
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 2.13,
|
330 |
+
"learning_rate": 1.9333333333333332e-07,
|
331 |
+
"loss": 0.0926,
|
332 |
+
"step": 1275
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 2.14,
|
336 |
+
"learning_rate": 1.725e-07,
|
337 |
+
"loss": 0.0893,
|
338 |
+
"step": 1300
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 2.16,
|
342 |
+
"learning_rate": 1.5166666666666666e-07,
|
343 |
+
"loss": 0.0984,
|
344 |
+
"step": 1325
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 2.18,
|
348 |
+
"learning_rate": 1.3083333333333333e-07,
|
349 |
+
"loss": 0.0902,
|
350 |
+
"step": 1350
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 2.19,
|
354 |
+
"learning_rate": 1.0999999999999999e-07,
|
355 |
+
"loss": 0.0972,
|
356 |
+
"step": 1375
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 2.21,
|
360 |
+
"learning_rate": 8.916666666666666e-08,
|
361 |
+
"loss": 0.0938,
|
362 |
+
"step": 1400
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 2.23,
|
366 |
+
"learning_rate": 6.833333333333332e-08,
|
367 |
+
"loss": 0.0847,
|
368 |
+
"step": 1425
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 2.24,
|
372 |
+
"learning_rate": 4.7499999999999995e-08,
|
373 |
+
"loss": 0.091,
|
374 |
+
"step": 1450
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 2.26,
|
378 |
+
"learning_rate": 2.6666666666666667e-08,
|
379 |
+
"loss": 0.0808,
|
380 |
+
"step": 1475
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 2.28,
|
384 |
+
"learning_rate": 5.8333333333333335e-09,
|
385 |
+
"loss": 0.079,
|
386 |
+
"step": 1500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 2.28,
|
390 |
+
"eval_loss": 0.1502685546875,
|
391 |
+
"eval_runtime": 2273.3398,
|
392 |
+
"eval_samples_per_second": 3.824,
|
393 |
+
"eval_steps_per_second": 0.239,
|
394 |
+
"eval_wer": 4.8385198634858195,
|
395 |
+
"step": 1500
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 2.28,
|
399 |
+
"step": 1500,
|
400 |
+
"total_flos": 1.0191220995495297e+20,
|
401 |
+
"train_loss": 0.15757004801432292,
|
402 |
+
"train_runtime": 18540.0869,
|
403 |
+
"train_samples_per_second": 2.589,
|
404 |
+
"train_steps_per_second": 0.081
|
405 |
+
}
|
406 |
+
],
|
407 |
+
"max_steps": 1500,
|
408 |
+
"num_train_epochs": 9223372036854775807,
|
409 |
+
"total_flos": 1.0191220995495297e+20,
|
410 |
+
"trial_name": null,
|
411 |
+
"trial_params": null
|
412 |
+
}
|