marinone94
commited on
Commit
•
6012db9
1
Parent(s):
32a526a
Training in progress, step 1600
Browse files- checkpoint-1200/scaler.pt +0 -3
- checkpoint-1200/trainer_state.json +0 -484
- checkpoint-1300/scheduler.pt +0 -3
- checkpoint-1400/config.json +0 -115
- checkpoint-1400/optimizer.pt +0 -3
- checkpoint-1400/preprocessor_config.json +0 -9
- checkpoint-1400/pytorch_model.bin +0 -3
- checkpoint-1400/rng_state.pth +0 -3
- checkpoint-1400/training_args.bin +0 -3
- {checkpoint-1200 → checkpoint-1500}/config.json +0 -0
- {checkpoint-1200 → checkpoint-1500}/optimizer.pt +1 -1
- {checkpoint-1200 → checkpoint-1500}/preprocessor_config.json +0 -0
- {checkpoint-1300 → checkpoint-1500}/pytorch_model.bin +1 -1
- {checkpoint-1300 → checkpoint-1500}/rng_state.pth +1 -1
- {checkpoint-1400 → checkpoint-1500}/scaler.pt +1 -1
- {checkpoint-1200 → checkpoint-1500}/scheduler.pt +1 -1
- {checkpoint-1400 → checkpoint-1500}/trainer_state.json +42 -3
- {checkpoint-1200 → checkpoint-1500}/training_args.bin +0 -0
- {checkpoint-1300 → checkpoint-1600}/config.json +0 -0
- {checkpoint-1300 → checkpoint-1600}/optimizer.pt +1 -1
- {checkpoint-1300 → checkpoint-1600}/preprocessor_config.json +0 -0
- {checkpoint-1200 → checkpoint-1600}/pytorch_model.bin +1 -1
- {checkpoint-1200 → checkpoint-1600}/rng_state.pth +1 -1
- {checkpoint-1300 → checkpoint-1600}/scaler.pt +1 -1
- {checkpoint-1400 → checkpoint-1600}/scheduler.pt +1 -1
- {checkpoint-1300 → checkpoint-1600}/trainer_state.json +120 -3
- {checkpoint-1300 → checkpoint-1600}/training_args.bin +0 -0
checkpoint-1200/scaler.pt
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:19c7277eaca0850ae3e9b6790b3d002d820169cce0671185e672c28c8ae8e056
|
3 |
-
size 559
|
|
|
|
|
|
|
|
checkpoint-1200/trainer_state.json
DELETED
@@ -1,484 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"best_metric": null,
|
3 |
-
"best_model_checkpoint": null,
|
4 |
-
"epoch": 13.185792349726777,
|
5 |
-
"global_step": 1200,
|
6 |
-
"is_hyper_param_search": false,
|
7 |
-
"is_local_process_zero": true,
|
8 |
-
"is_world_process_zero": true,
|
9 |
-
"log_history": [
|
10 |
-
{
|
11 |
-
"epoch": 0.22,
|
12 |
-
"learning_rate": 2.5e-06,
|
13 |
-
"loss": 3.5867,
|
14 |
-
"step": 20
|
15 |
-
},
|
16 |
-
{
|
17 |
-
"epoch": 0.44,
|
18 |
-
"learning_rate": 5e-06,
|
19 |
-
"loss": 3.5457,
|
20 |
-
"step": 40
|
21 |
-
},
|
22 |
-
{
|
23 |
-
"epoch": 0.66,
|
24 |
-
"learning_rate": 7.5e-06,
|
25 |
-
"loss": 3.4513,
|
26 |
-
"step": 60
|
27 |
-
},
|
28 |
-
{
|
29 |
-
"epoch": 0.87,
|
30 |
-
"learning_rate": 1e-05,
|
31 |
-
"loss": 3.3432,
|
32 |
-
"step": 80
|
33 |
-
},
|
34 |
-
{
|
35 |
-
"epoch": 1.1,
|
36 |
-
"learning_rate": 1.25e-05,
|
37 |
-
"loss": 3.3533,
|
38 |
-
"step": 100
|
39 |
-
},
|
40 |
-
{
|
41 |
-
"epoch": 1.1,
|
42 |
-
"eval_loss": 3.2806732654571533,
|
43 |
-
"eval_runtime": 190.4728,
|
44 |
-
"eval_samples_per_second": 25.426,
|
45 |
-
"eval_steps_per_second": 0.798,
|
46 |
-
"eval_wer": 1.0,
|
47 |
-
"step": 100
|
48 |
-
},
|
49 |
-
{
|
50 |
-
"epoch": 1.32,
|
51 |
-
"learning_rate": 1.5e-05,
|
52 |
-
"loss": 3.2217,
|
53 |
-
"step": 120
|
54 |
-
},
|
55 |
-
{
|
56 |
-
"epoch": 1.54,
|
57 |
-
"learning_rate": 1.7500000000000002e-05,
|
58 |
-
"loss": 3.1765,
|
59 |
-
"step": 140
|
60 |
-
},
|
61 |
-
{
|
62 |
-
"epoch": 1.75,
|
63 |
-
"learning_rate": 2e-05,
|
64 |
-
"loss": 3.1408,
|
65 |
-
"step": 160
|
66 |
-
},
|
67 |
-
{
|
68 |
-
"epoch": 1.97,
|
69 |
-
"learning_rate": 2.2499999999999998e-05,
|
70 |
-
"loss": 3.1165,
|
71 |
-
"step": 180
|
72 |
-
},
|
73 |
-
{
|
74 |
-
"epoch": 2.2,
|
75 |
-
"learning_rate": 2.5e-05,
|
76 |
-
"loss": 3.1709,
|
77 |
-
"step": 200
|
78 |
-
},
|
79 |
-
{
|
80 |
-
"epoch": 2.2,
|
81 |
-
"eval_loss": 3.1325438022613525,
|
82 |
-
"eval_runtime": 192.4978,
|
83 |
-
"eval_samples_per_second": 25.159,
|
84 |
-
"eval_steps_per_second": 0.79,
|
85 |
-
"eval_wer": 1.0,
|
86 |
-
"step": 200
|
87 |
-
},
|
88 |
-
{
|
89 |
-
"epoch": 2.42,
|
90 |
-
"learning_rate": 2.75e-05,
|
91 |
-
"loss": 3.079,
|
92 |
-
"step": 220
|
93 |
-
},
|
94 |
-
{
|
95 |
-
"epoch": 2.63,
|
96 |
-
"learning_rate": 3e-05,
|
97 |
-
"loss": 3.0677,
|
98 |
-
"step": 240
|
99 |
-
},
|
100 |
-
{
|
101 |
-
"epoch": 2.85,
|
102 |
-
"learning_rate": 3.2500000000000004e-05,
|
103 |
-
"loss": 3.0656,
|
104 |
-
"step": 260
|
105 |
-
},
|
106 |
-
{
|
107 |
-
"epoch": 3.08,
|
108 |
-
"learning_rate": 3.5000000000000004e-05,
|
109 |
-
"loss": 3.1463,
|
110 |
-
"step": 280
|
111 |
-
},
|
112 |
-
{
|
113 |
-
"epoch": 3.3,
|
114 |
-
"learning_rate": 3.75e-05,
|
115 |
-
"loss": 3.0573,
|
116 |
-
"step": 300
|
117 |
-
},
|
118 |
-
{
|
119 |
-
"epoch": 3.3,
|
120 |
-
"eval_loss": 3.0614514350891113,
|
121 |
-
"eval_runtime": 194.36,
|
122 |
-
"eval_samples_per_second": 24.918,
|
123 |
-
"eval_steps_per_second": 0.782,
|
124 |
-
"eval_wer": 1.0,
|
125 |
-
"step": 300
|
126 |
-
},
|
127 |
-
{
|
128 |
-
"epoch": 3.51,
|
129 |
-
"learning_rate": 4e-05,
|
130 |
-
"loss": 3.0511,
|
131 |
-
"step": 320
|
132 |
-
},
|
133 |
-
{
|
134 |
-
"epoch": 3.73,
|
135 |
-
"learning_rate": 4.25e-05,
|
136 |
-
"loss": 3.0358,
|
137 |
-
"step": 340
|
138 |
-
},
|
139 |
-
{
|
140 |
-
"epoch": 3.95,
|
141 |
-
"learning_rate": 4.4999999999999996e-05,
|
142 |
-
"loss": 3.0416,
|
143 |
-
"step": 360
|
144 |
-
},
|
145 |
-
{
|
146 |
-
"epoch": 4.17,
|
147 |
-
"learning_rate": 4.75e-05,
|
148 |
-
"loss": 3.1053,
|
149 |
-
"step": 380
|
150 |
-
},
|
151 |
-
{
|
152 |
-
"epoch": 4.39,
|
153 |
-
"learning_rate": 5e-05,
|
154 |
-
"loss": 3.0314,
|
155 |
-
"step": 400
|
156 |
-
},
|
157 |
-
{
|
158 |
-
"epoch": 4.39,
|
159 |
-
"eval_loss": 3.0990231037139893,
|
160 |
-
"eval_runtime": 198.3688,
|
161 |
-
"eval_samples_per_second": 24.414,
|
162 |
-
"eval_steps_per_second": 0.766,
|
163 |
-
"eval_wer": 1.0,
|
164 |
-
"step": 400
|
165 |
-
},
|
166 |
-
{
|
167 |
-
"epoch": 4.61,
|
168 |
-
"learning_rate": 5.25e-05,
|
169 |
-
"loss": 3.0309,
|
170 |
-
"step": 420
|
171 |
-
},
|
172 |
-
{
|
173 |
-
"epoch": 4.83,
|
174 |
-
"learning_rate": 5.5e-05,
|
175 |
-
"loss": 3.0259,
|
176 |
-
"step": 440
|
177 |
-
},
|
178 |
-
{
|
179 |
-
"epoch": 5.05,
|
180 |
-
"learning_rate": 5.75e-05,
|
181 |
-
"loss": 3.0998,
|
182 |
-
"step": 460
|
183 |
-
},
|
184 |
-
{
|
185 |
-
"epoch": 5.27,
|
186 |
-
"learning_rate": 6e-05,
|
187 |
-
"loss": 3.0152,
|
188 |
-
"step": 480
|
189 |
-
},
|
190 |
-
{
|
191 |
-
"epoch": 5.49,
|
192 |
-
"learning_rate": 6.25e-05,
|
193 |
-
"loss": 3.0129,
|
194 |
-
"step": 500
|
195 |
-
},
|
196 |
-
{
|
197 |
-
"epoch": 5.49,
|
198 |
-
"eval_loss": 3.039973497390747,
|
199 |
-
"eval_runtime": 190.8567,
|
200 |
-
"eval_samples_per_second": 25.375,
|
201 |
-
"eval_steps_per_second": 0.796,
|
202 |
-
"eval_wer": 1.0,
|
203 |
-
"step": 500
|
204 |
-
},
|
205 |
-
{
|
206 |
-
"epoch": 5.71,
|
207 |
-
"learning_rate": 6.500000000000001e-05,
|
208 |
-
"loss": 3.0088,
|
209 |
-
"step": 520
|
210 |
-
},
|
211 |
-
{
|
212 |
-
"epoch": 5.93,
|
213 |
-
"learning_rate": 6.75e-05,
|
214 |
-
"loss": 3.0051,
|
215 |
-
"step": 540
|
216 |
-
},
|
217 |
-
{
|
218 |
-
"epoch": 6.15,
|
219 |
-
"learning_rate": 7.000000000000001e-05,
|
220 |
-
"loss": 3.073,
|
221 |
-
"step": 560
|
222 |
-
},
|
223 |
-
{
|
224 |
-
"epoch": 6.37,
|
225 |
-
"learning_rate": 7.25e-05,
|
226 |
-
"loss": 3.0031,
|
227 |
-
"step": 580
|
228 |
-
},
|
229 |
-
{
|
230 |
-
"epoch": 6.59,
|
231 |
-
"learning_rate": 7.5e-05,
|
232 |
-
"loss": 2.9964,
|
233 |
-
"step": 600
|
234 |
-
},
|
235 |
-
{
|
236 |
-
"epoch": 6.59,
|
237 |
-
"eval_loss": 2.998962640762329,
|
238 |
-
"eval_runtime": 193.5213,
|
239 |
-
"eval_samples_per_second": 25.026,
|
240 |
-
"eval_steps_per_second": 0.785,
|
241 |
-
"eval_wer": 1.0,
|
242 |
-
"step": 600
|
243 |
-
},
|
244 |
-
{
|
245 |
-
"epoch": 6.81,
|
246 |
-
"learning_rate": 7.75e-05,
|
247 |
-
"loss": 2.9921,
|
248 |
-
"step": 620
|
249 |
-
},
|
250 |
-
{
|
251 |
-
"epoch": 7.03,
|
252 |
-
"learning_rate": 8e-05,
|
253 |
-
"loss": 3.0665,
|
254 |
-
"step": 640
|
255 |
-
},
|
256 |
-
{
|
257 |
-
"epoch": 7.25,
|
258 |
-
"learning_rate": 8.25e-05,
|
259 |
-
"loss": 2.9826,
|
260 |
-
"step": 660
|
261 |
-
},
|
262 |
-
{
|
263 |
-
"epoch": 7.47,
|
264 |
-
"learning_rate": 8.5e-05,
|
265 |
-
"loss": 2.9689,
|
266 |
-
"step": 680
|
267 |
-
},
|
268 |
-
{
|
269 |
-
"epoch": 7.69,
|
270 |
-
"learning_rate": 8.75e-05,
|
271 |
-
"loss": 2.9602,
|
272 |
-
"step": 700
|
273 |
-
},
|
274 |
-
{
|
275 |
-
"epoch": 7.69,
|
276 |
-
"eval_loss": 2.9620397090911865,
|
277 |
-
"eval_runtime": 193.5851,
|
278 |
-
"eval_samples_per_second": 25.017,
|
279 |
-
"eval_steps_per_second": 0.785,
|
280 |
-
"eval_wer": 1.0,
|
281 |
-
"step": 700
|
282 |
-
},
|
283 |
-
{
|
284 |
-
"epoch": 7.91,
|
285 |
-
"learning_rate": 8.999999999999999e-05,
|
286 |
-
"loss": 2.9639,
|
287 |
-
"step": 720
|
288 |
-
},
|
289 |
-
{
|
290 |
-
"epoch": 8.13,
|
291 |
-
"learning_rate": 9.25e-05,
|
292 |
-
"loss": 3.0215,
|
293 |
-
"step": 740
|
294 |
-
},
|
295 |
-
{
|
296 |
-
"epoch": 8.35,
|
297 |
-
"learning_rate": 9.5e-05,
|
298 |
-
"loss": 2.9454,
|
299 |
-
"step": 760
|
300 |
-
},
|
301 |
-
{
|
302 |
-
"epoch": 8.57,
|
303 |
-
"learning_rate": 9.750000000000001e-05,
|
304 |
-
"loss": 2.9239,
|
305 |
-
"step": 780
|
306 |
-
},
|
307 |
-
{
|
308 |
-
"epoch": 8.79,
|
309 |
-
"learning_rate": 0.0001,
|
310 |
-
"loss": 2.8756,
|
311 |
-
"step": 800
|
312 |
-
},
|
313 |
-
{
|
314 |
-
"epoch": 8.79,
|
315 |
-
"eval_loss": 2.7302000522613525,
|
316 |
-
"eval_runtime": 191.8065,
|
317 |
-
"eval_samples_per_second": 25.249,
|
318 |
-
"eval_steps_per_second": 0.792,
|
319 |
-
"eval_wer": 1.0,
|
320 |
-
"step": 800
|
321 |
-
},
|
322 |
-
{
|
323 |
-
"epoch": 9.01,
|
324 |
-
"learning_rate": 0.0001025,
|
325 |
-
"loss": 2.8933,
|
326 |
-
"step": 820
|
327 |
-
},
|
328 |
-
{
|
329 |
-
"epoch": 9.23,
|
330 |
-
"learning_rate": 0.000105,
|
331 |
-
"loss": 2.7318,
|
332 |
-
"step": 840
|
333 |
-
},
|
334 |
-
{
|
335 |
-
"epoch": 9.45,
|
336 |
-
"learning_rate": 0.0001075,
|
337 |
-
"loss": 2.5941,
|
338 |
-
"step": 860
|
339 |
-
},
|
340 |
-
{
|
341 |
-
"epoch": 9.67,
|
342 |
-
"learning_rate": 0.00011,
|
343 |
-
"loss": 2.4441,
|
344 |
-
"step": 880
|
345 |
-
},
|
346 |
-
{
|
347 |
-
"epoch": 9.89,
|
348 |
-
"learning_rate": 0.00011250000000000001,
|
349 |
-
"loss": 2.2931,
|
350 |
-
"step": 900
|
351 |
-
},
|
352 |
-
{
|
353 |
-
"epoch": 9.89,
|
354 |
-
"eval_loss": 1.5057899951934814,
|
355 |
-
"eval_runtime": 196.4368,
|
356 |
-
"eval_samples_per_second": 24.654,
|
357 |
-
"eval_steps_per_second": 0.774,
|
358 |
-
"eval_wer": 0.9775759296054499,
|
359 |
-
"step": 900
|
360 |
-
},
|
361 |
-
{
|
362 |
-
"epoch": 10.11,
|
363 |
-
"learning_rate": 0.000115,
|
364 |
-
"loss": 2.1999,
|
365 |
-
"step": 920
|
366 |
-
},
|
367 |
-
{
|
368 |
-
"epoch": 10.33,
|
369 |
-
"learning_rate": 0.0001175,
|
370 |
-
"loss": 2.0574,
|
371 |
-
"step": 940
|
372 |
-
},
|
373 |
-
{
|
374 |
-
"epoch": 10.55,
|
375 |
-
"learning_rate": 0.00012,
|
376 |
-
"loss": 1.9672,
|
377 |
-
"step": 960
|
378 |
-
},
|
379 |
-
{
|
380 |
-
"epoch": 10.77,
|
381 |
-
"learning_rate": 0.0001225,
|
382 |
-
"loss": 1.9015,
|
383 |
-
"step": 980
|
384 |
-
},
|
385 |
-
{
|
386 |
-
"epoch": 10.98,
|
387 |
-
"learning_rate": 0.000125,
|
388 |
-
"loss": 1.8427,
|
389 |
-
"step": 1000
|
390 |
-
},
|
391 |
-
{
|
392 |
-
"epoch": 10.98,
|
393 |
-
"eval_loss": 0.9154536724090576,
|
394 |
-
"eval_runtime": 203.1168,
|
395 |
-
"eval_samples_per_second": 23.843,
|
396 |
-
"eval_steps_per_second": 0.748,
|
397 |
-
"eval_wer": 0.7832245245529378,
|
398 |
-
"step": 1000
|
399 |
-
},
|
400 |
-
{
|
401 |
-
"epoch": 11.21,
|
402 |
-
"learning_rate": 0.0001275,
|
403 |
-
"loss": 1.8155,
|
404 |
-
"step": 1020
|
405 |
-
},
|
406 |
-
{
|
407 |
-
"epoch": 11.43,
|
408 |
-
"learning_rate": 0.00013000000000000002,
|
409 |
-
"loss": 1.6767,
|
410 |
-
"step": 1040
|
411 |
-
},
|
412 |
-
{
|
413 |
-
"epoch": 11.64,
|
414 |
-
"learning_rate": 0.00013250000000000002,
|
415 |
-
"loss": 1.5184,
|
416 |
-
"step": 1060
|
417 |
-
},
|
418 |
-
{
|
419 |
-
"epoch": 11.86,
|
420 |
-
"learning_rate": 0.000135,
|
421 |
-
"loss": 1.4505,
|
422 |
-
"step": 1080
|
423 |
-
},
|
424 |
-
{
|
425 |
-
"epoch": 12.09,
|
426 |
-
"learning_rate": 0.0001375,
|
427 |
-
"loss": 1.4286,
|
428 |
-
"step": 1100
|
429 |
-
},
|
430 |
-
{
|
431 |
-
"epoch": 12.09,
|
432 |
-
"eval_loss": 0.4074769914150238,
|
433 |
-
"eval_runtime": 194.7798,
|
434 |
-
"eval_samples_per_second": 24.864,
|
435 |
-
"eval_steps_per_second": 0.78,
|
436 |
-
"eval_wer": 0.37964802724950325,
|
437 |
-
"step": 1100
|
438 |
-
},
|
439 |
-
{
|
440 |
-
"epoch": 12.31,
|
441 |
-
"learning_rate": 0.00014000000000000001,
|
442 |
-
"loss": 1.3478,
|
443 |
-
"step": 1120
|
444 |
-
},
|
445 |
-
{
|
446 |
-
"epoch": 12.52,
|
447 |
-
"learning_rate": 0.0001425,
|
448 |
-
"loss": 1.3164,
|
449 |
-
"step": 1140
|
450 |
-
},
|
451 |
-
{
|
452 |
-
"epoch": 12.74,
|
453 |
-
"learning_rate": 0.000145,
|
454 |
-
"loss": 1.269,
|
455 |
-
"step": 1160
|
456 |
-
},
|
457 |
-
{
|
458 |
-
"epoch": 12.96,
|
459 |
-
"learning_rate": 0.0001475,
|
460 |
-
"loss": 1.2292,
|
461 |
-
"step": 1180
|
462 |
-
},
|
463 |
-
{
|
464 |
-
"epoch": 13.19,
|
465 |
-
"learning_rate": 0.00015,
|
466 |
-
"loss": 1.2229,
|
467 |
-
"step": 1200
|
468 |
-
},
|
469 |
-
{
|
470 |
-
"epoch": 13.19,
|
471 |
-
"eval_loss": 0.28933778405189514,
|
472 |
-
"eval_runtime": 192.6398,
|
473 |
-
"eval_samples_per_second": 25.14,
|
474 |
-
"eval_steps_per_second": 0.789,
|
475 |
-
"eval_wer": 0.26520011353959694,
|
476 |
-
"step": 1200
|
477 |
-
}
|
478 |
-
],
|
479 |
-
"max_steps": 4550,
|
480 |
-
"num_train_epochs": 50,
|
481 |
-
"total_flos": 1.8579484602378154e+19,
|
482 |
-
"trial_name": null,
|
483 |
-
"trial_params": null
|
484 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1300/scheduler.pt
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:49a74fb1217961243834b43dc5bb531167031d0d67fb9ade238f561b7c566d57
|
3 |
-
size 623
|
|
|
|
|
|
|
|
checkpoint-1400/config.json
DELETED
@@ -1,115 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"_name_or_path": "marinone94/xls-r-300m-sv-robust",
|
3 |
-
"activation_dropout": 0.1,
|
4 |
-
"adapter_kernel_size": 3,
|
5 |
-
"adapter_stride": 2,
|
6 |
-
"add_adapter": false,
|
7 |
-
"apply_spec_augment": true,
|
8 |
-
"architectures": [
|
9 |
-
"Wav2Vec2ForCTC"
|
10 |
-
],
|
11 |
-
"attention_dropout": 0.0,
|
12 |
-
"bos_token_id": 1,
|
13 |
-
"classifier_proj_size": 256,
|
14 |
-
"codevector_dim": 768,
|
15 |
-
"contrastive_logits_temperature": 0.1,
|
16 |
-
"conv_bias": true,
|
17 |
-
"conv_dim": [
|
18 |
-
512,
|
19 |
-
512,
|
20 |
-
512,
|
21 |
-
512,
|
22 |
-
512,
|
23 |
-
512,
|
24 |
-
512
|
25 |
-
],
|
26 |
-
"conv_kernel": [
|
27 |
-
10,
|
28 |
-
3,
|
29 |
-
3,
|
30 |
-
3,
|
31 |
-
3,
|
32 |
-
2,
|
33 |
-
2
|
34 |
-
],
|
35 |
-
"conv_stride": [
|
36 |
-
5,
|
37 |
-
2,
|
38 |
-
2,
|
39 |
-
2,
|
40 |
-
2,
|
41 |
-
2,
|
42 |
-
2
|
43 |
-
],
|
44 |
-
"ctc_loss_reduction": "mean",
|
45 |
-
"ctc_zero_infinity": false,
|
46 |
-
"diversity_loss_weight": 0.1,
|
47 |
-
"do_stable_layer_norm": true,
|
48 |
-
"eos_token_id": 2,
|
49 |
-
"feat_extract_activation": "gelu",
|
50 |
-
"feat_extract_dropout": 0.0,
|
51 |
-
"feat_extract_norm": "layer",
|
52 |
-
"feat_proj_dropout": 0.0,
|
53 |
-
"feat_quantizer_dropout": 0.0,
|
54 |
-
"final_dropout": 0.0,
|
55 |
-
"hidden_act": "gelu",
|
56 |
-
"hidden_dropout": 0.0,
|
57 |
-
"hidden_size": 1024,
|
58 |
-
"initializer_range": 0.02,
|
59 |
-
"intermediate_size": 4096,
|
60 |
-
"layer_norm_eps": 1e-05,
|
61 |
-
"layerdrop": 0.0,
|
62 |
-
"mask_channel_length": 10,
|
63 |
-
"mask_channel_min_space": 1,
|
64 |
-
"mask_channel_other": 0.0,
|
65 |
-
"mask_channel_prob": 0.0,
|
66 |
-
"mask_channel_selection": "static",
|
67 |
-
"mask_feature_length": 64,
|
68 |
-
"mask_feature_min_masks": 0,
|
69 |
-
"mask_feature_prob": 0.25,
|
70 |
-
"mask_time_length": 10,
|
71 |
-
"mask_time_min_masks": 2,
|
72 |
-
"mask_time_min_space": 1,
|
73 |
-
"mask_time_other": 0.0,
|
74 |
-
"mask_time_prob": 0.75,
|
75 |
-
"mask_time_selection": "static",
|
76 |
-
"model_type": "wav2vec2",
|
77 |
-
"num_adapter_layers": 3,
|
78 |
-
"num_attention_heads": 16,
|
79 |
-
"num_codevector_groups": 2,
|
80 |
-
"num_codevectors_per_group": 320,
|
81 |
-
"num_conv_pos_embedding_groups": 16,
|
82 |
-
"num_conv_pos_embeddings": 128,
|
83 |
-
"num_feat_extract_layers": 7,
|
84 |
-
"num_hidden_layers": 24,
|
85 |
-
"num_negatives": 100,
|
86 |
-
"output_hidden_size": 1024,
|
87 |
-
"pad_token_id": 31,
|
88 |
-
"proj_codevector_dim": 768,
|
89 |
-
"tdnn_dilation": [
|
90 |
-
1,
|
91 |
-
2,
|
92 |
-
3,
|
93 |
-
1,
|
94 |
-
1
|
95 |
-
],
|
96 |
-
"tdnn_dim": [
|
97 |
-
512,
|
98 |
-
512,
|
99 |
-
512,
|
100 |
-
512,
|
101 |
-
1500
|
102 |
-
],
|
103 |
-
"tdnn_kernel": [
|
104 |
-
5,
|
105 |
-
3,
|
106 |
-
3,
|
107 |
-
1,
|
108 |
-
1
|
109 |
-
],
|
110 |
-
"torch_dtype": "float32",
|
111 |
-
"transformers_version": "4.17.0.dev0",
|
112 |
-
"use_weighted_layer_sum": false,
|
113 |
-
"vocab_size": 34,
|
114 |
-
"xvector_output_dim": 512
|
115 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1400/optimizer.pt
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:066471e5b1e42d95f381420c5eb456280eaebbcce841ac5387ae7bfde1fc9de6
|
3 |
-
size 2490337809
|
|
|
|
|
|
|
|
checkpoint-1400/preprocessor_config.json
DELETED
@@ -1,9 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"do_normalize": true,
|
3 |
-
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
-
"feature_size": 1,
|
5 |
-
"padding_side": "right",
|
6 |
-
"padding_value": 0,
|
7 |
-
"return_attention_mask": true,
|
8 |
-
"sampling_rate": 16000
|
9 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1400/pytorch_model.bin
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:1a459f72ab853af3d30be6993d7c2796ebd50d28c9cae60928258ccc702d6e46
|
3 |
-
size 1262063089
|
|
|
|
|
|
|
|
checkpoint-1400/rng_state.pth
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:f1f4e020badfa95b92140e362fadf8866d696357e6c8984f31eba89adcf57045
|
3 |
-
size 14631
|
|
|
|
|
|
|
|
checkpoint-1400/training_args.bin
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:61c53503eb760641941023d2dcab7bb32b620bf895e31db5c3910b2760135b07
|
3 |
-
size 3055
|
|
|
|
|
|
|
|
{checkpoint-1200 → checkpoint-1500}/config.json
RENAMED
File without changes
|
{checkpoint-1200 → checkpoint-1500}/optimizer.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2490337809
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:59d9e961a47ab1f00dd325d343d6b3a2c56a477a890edfff182fab52ad7d1961
|
3 |
size 2490337809
|
{checkpoint-1200 → checkpoint-1500}/preprocessor_config.json
RENAMED
File without changes
|
{checkpoint-1300 → checkpoint-1500}/pytorch_model.bin
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262063089
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:22c881807ec99ef3592933c11ac25ff336dab472ee8669960c08f6bb055cb11c
|
3 |
size 1262063089
|
{checkpoint-1300 → checkpoint-1500}/rng_state.pth
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14567
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:26a16fb06bd95550585922193ec31f3a9a5878ee3e77efc87c1b8bdb4a21d142
|
3 |
size 14567
|
{checkpoint-1400 → checkpoint-1500}/scaler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 559
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:18b77bf231929dcf1b0885a05986daead51666518206e8005c99cd124832d9f9
|
3 |
size 559
|
{checkpoint-1200 → checkpoint-1500}/scheduler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a6d0ac6fcb7f6a54e2016654eacc790bd572cb9149b97999e246f807a8519d7d
|
3 |
size 623
|
{checkpoint-1400 → checkpoint-1500}/trainer_state.json
RENAMED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -552,11 +552,50 @@
|
|
552 |
"eval_steps_per_second": 0.802,
|
553 |
"eval_wer": 0.19727504967357365,
|
554 |
"step": 1400
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
555 |
}
|
556 |
],
|
557 |
"max_steps": 4550,
|
558 |
"num_train_epochs": 50,
|
559 |
-
"total_flos": 2.
|
560 |
"trial_name": null,
|
561 |
"trial_params": null
|
562 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 16.48087431693989,
|
5 |
+
"global_step": 1500,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
552 |
"eval_steps_per_second": 0.802,
|
553 |
"eval_wer": 0.19727504967357365,
|
554 |
"step": 1400
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 15.6,
|
558 |
+
"learning_rate": 0.0001775,
|
559 |
+
"loss": 1.0733,
|
560 |
+
"step": 1420
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 15.82,
|
564 |
+
"learning_rate": 0.00017999999999999998,
|
565 |
+
"loss": 1.0635,
|
566 |
+
"step": 1440
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 16.04,
|
570 |
+
"learning_rate": 0.0001825,
|
571 |
+
"loss": 1.1079,
|
572 |
+
"step": 1460
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 16.26,
|
576 |
+
"learning_rate": 0.000185,
|
577 |
+
"loss": 1.0467,
|
578 |
+
"step": 1480
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 16.48,
|
582 |
+
"learning_rate": 0.0001875,
|
583 |
+
"loss": 1.0667,
|
584 |
+
"step": 1500
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 16.48,
|
588 |
+
"eval_loss": 0.21288961172103882,
|
589 |
+
"eval_runtime": 189.0078,
|
590 |
+
"eval_samples_per_second": 25.623,
|
591 |
+
"eval_steps_per_second": 0.804,
|
592 |
+
"eval_wer": 0.18944081748509792,
|
593 |
+
"step": 1500
|
594 |
}
|
595 |
],
|
596 |
"max_steps": 4550,
|
597 |
"num_train_epochs": 50,
|
598 |
+
"total_flos": 2.31847014520748e+19,
|
599 |
"trial_name": null,
|
600 |
"trial_params": null
|
601 |
}
|
{checkpoint-1200 → checkpoint-1500}/training_args.bin
RENAMED
File without changes
|
{checkpoint-1300 → checkpoint-1600}/config.json
RENAMED
File without changes
|
{checkpoint-1300 → checkpoint-1600}/optimizer.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2490337809
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c01538eb4bd0c6517e58a239a0edad2ba490de930aaa3809478517bc233bd339
|
3 |
size 2490337809
|
{checkpoint-1300 → checkpoint-1600}/preprocessor_config.json
RENAMED
File without changes
|
{checkpoint-1200 → checkpoint-1600}/pytorch_model.bin
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262063089
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:577857a0d5063da507a1ec462e3b3478b15131b86a5760953cbe69453453fe49
|
3 |
size 1262063089
|
{checkpoint-1200 → checkpoint-1600}/rng_state.pth
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14567
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:32079a7ecfdaedcafafe628415e8789ccbf50b12b21038d2ca0bc5ad993d93dd
|
3 |
size 14567
|
{checkpoint-1300 → checkpoint-1600}/scaler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 559
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:45cc99d310862a4180e2b863b5866b132e05e88ff0bb7c56d68724e754f0549e
|
3 |
size 559
|
{checkpoint-1400 → checkpoint-1600}/scheduler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8a5627b7bd7a23e3bd273e778f370a9ce7118842c3616c776314b677675af7a5
|
3 |
size 623
|
{checkpoint-1300 → checkpoint-1600}/trainer_state.json
RENAMED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -513,11 +513,128 @@
|
|
513 |
"eval_steps_per_second": 0.766,
|
514 |
"eval_wer": 0.2253760999148453,
|
515 |
"step": 1300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
516 |
}
|
517 |
],
|
518 |
"max_steps": 4550,
|
519 |
"num_train_epochs": 50,
|
520 |
-
"total_flos": 2.
|
521 |
"trial_name": null,
|
522 |
"trial_params": null
|
523 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 17.579234972677597,
|
5 |
+
"global_step": 1600,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
513 |
"eval_steps_per_second": 0.766,
|
514 |
"eval_wer": 0.2253760999148453,
|
515 |
"step": 1300
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 14.5,
|
519 |
+
"learning_rate": 0.000165,
|
520 |
+
"loss": 1.1217,
|
521 |
+
"step": 1320
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 14.72,
|
525 |
+
"learning_rate": 0.0001675,
|
526 |
+
"loss": 1.1236,
|
527 |
+
"step": 1340
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 14.94,
|
531 |
+
"learning_rate": 0.00017,
|
532 |
+
"loss": 1.0969,
|
533 |
+
"step": 1360
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 15.16,
|
537 |
+
"learning_rate": 0.0001725,
|
538 |
+
"loss": 1.0971,
|
539 |
+
"step": 1380
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 15.38,
|
543 |
+
"learning_rate": 0.000175,
|
544 |
+
"loss": 1.0663,
|
545 |
+
"step": 1400
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 15.38,
|
549 |
+
"eval_loss": 0.22192780673503876,
|
550 |
+
"eval_runtime": 189.4566,
|
551 |
+
"eval_samples_per_second": 25.563,
|
552 |
+
"eval_steps_per_second": 0.802,
|
553 |
+
"eval_wer": 0.19727504967357365,
|
554 |
+
"step": 1400
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 15.6,
|
558 |
+
"learning_rate": 0.0001775,
|
559 |
+
"loss": 1.0733,
|
560 |
+
"step": 1420
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 15.82,
|
564 |
+
"learning_rate": 0.00017999999999999998,
|
565 |
+
"loss": 1.0635,
|
566 |
+
"step": 1440
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 16.04,
|
570 |
+
"learning_rate": 0.0001825,
|
571 |
+
"loss": 1.1079,
|
572 |
+
"step": 1460
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 16.26,
|
576 |
+
"learning_rate": 0.000185,
|
577 |
+
"loss": 1.0467,
|
578 |
+
"step": 1480
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 16.48,
|
582 |
+
"learning_rate": 0.0001875,
|
583 |
+
"loss": 1.0667,
|
584 |
+
"step": 1500
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 16.48,
|
588 |
+
"eval_loss": 0.21288961172103882,
|
589 |
+
"eval_runtime": 189.0078,
|
590 |
+
"eval_samples_per_second": 25.623,
|
591 |
+
"eval_steps_per_second": 0.804,
|
592 |
+
"eval_wer": 0.18944081748509792,
|
593 |
+
"step": 1500
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 16.7,
|
597 |
+
"learning_rate": 0.00019,
|
598 |
+
"loss": 1.0204,
|
599 |
+
"step": 1520
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 16.92,
|
603 |
+
"learning_rate": 0.00019250000000000002,
|
604 |
+
"loss": 1.0356,
|
605 |
+
"step": 1540
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 17.14,
|
609 |
+
"learning_rate": 0.00019500000000000002,
|
610 |
+
"loss": 1.0538,
|
611 |
+
"step": 1560
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 17.36,
|
615 |
+
"learning_rate": 0.0001975,
|
616 |
+
"loss": 1.0228,
|
617 |
+
"step": 1580
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 17.58,
|
621 |
+
"learning_rate": 0.0002,
|
622 |
+
"loss": 1.0193,
|
623 |
+
"step": 1600
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 17.58,
|
627 |
+
"eval_loss": 0.1991206556558609,
|
628 |
+
"eval_runtime": 191.1717,
|
629 |
+
"eval_samples_per_second": 25.333,
|
630 |
+
"eval_steps_per_second": 0.795,
|
631 |
+
"eval_wer": 0.17885325007096226,
|
632 |
+
"step": 1600
|
633 |
}
|
634 |
],
|
635 |
"max_steps": 4550,
|
636 |
"num_train_epochs": 50,
|
637 |
+
"total_flos": 2.4776551891108737e+19,
|
638 |
"trial_name": null,
|
639 |
"trial_params": null
|
640 |
}
|
{checkpoint-1300 → checkpoint-1600}/training_args.bin
RENAMED
File without changes
|