End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_loss": 3.449990749359131,
|
4 |
+
"eval_runtime": 32.2707,
|
5 |
+
"eval_samples": 979,
|
6 |
+
"eval_samples_per_second": 30.337,
|
7 |
+
"eval_steps_per_second": 1.921,
|
8 |
+
"eval_wer": 0.9999107182715057,
|
9 |
+
"train_loss": 3.4646308027512136,
|
10 |
+
"train_runtime": 5488.5308,
|
11 |
+
"train_samples": 3934,
|
12 |
+
"train_samples_per_second": 10.752,
|
13 |
+
"train_steps_per_second": 5.376
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_loss": 3.449990749359131,
|
4 |
+
"eval_runtime": 32.2707,
|
5 |
+
"eval_samples": 979,
|
6 |
+
"eval_samples_per_second": 30.337,
|
7 |
+
"eval_steps_per_second": 1.921,
|
8 |
+
"eval_wer": 0.9999107182715057
|
9 |
+
}
|
runs/Feb17_16-07-46_khanhpluto-Z10PE-D8-WS/events.out.tfevents.1645094549.khanhpluto-Z10PE-D8-WS.20559.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fff4389ae62469961992320a7b2f3be4cff3a9e5806397ba13329d6a1d490c82
|
3 |
+
size 364
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"train_loss": 3.4646308027512136,
|
4 |
+
"train_runtime": 5488.5308,
|
5 |
+
"train_samples": 3934,
|
6 |
+
"train_samples_per_second": 10.752,
|
7 |
+
"train_steps_per_second": 5.376
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,379 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 15.0,
|
5 |
+
"global_step": 29505,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.25,
|
12 |
+
"learning_rate": 0.002976,
|
13 |
+
"loss": 4.0956,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.51,
|
18 |
+
"learning_rate": 0.0029488019307016035,
|
19 |
+
"loss": 3.522,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.76,
|
24 |
+
"learning_rate": 0.0028970867091880712,
|
25 |
+
"loss": 3.4575,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.02,
|
30 |
+
"learning_rate": 0.002845371487674539,
|
31 |
+
"loss": 3.4469,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.27,
|
36 |
+
"learning_rate": 0.0027936562661610067,
|
37 |
+
"loss": 3.4404,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.53,
|
42 |
+
"learning_rate": 0.0027419410446474744,
|
43 |
+
"loss": 3.4604,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.78,
|
48 |
+
"learning_rate": 0.002690225823133942,
|
49 |
+
"loss": 3.4632,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 2.03,
|
54 |
+
"learning_rate": 0.0026385106016204103,
|
55 |
+
"loss": 3.4906,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 2.29,
|
60 |
+
"learning_rate": 0.002586795380106878,
|
61 |
+
"loss": 3.4523,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.54,
|
66 |
+
"learning_rate": 0.002535080158593346,
|
67 |
+
"loss": 3.4521,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.8,
|
72 |
+
"learning_rate": 0.002483468367522841,
|
73 |
+
"loss": 3.4765,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 3.05,
|
78 |
+
"learning_rate": 0.0024318565764523356,
|
79 |
+
"loss": 3.4482,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 3.3,
|
84 |
+
"learning_rate": 0.0023801413549388037,
|
85 |
+
"loss": 3.4663,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 3.56,
|
90 |
+
"learning_rate": 0.0023284261334252715,
|
91 |
+
"loss": 3.4747,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 3.81,
|
96 |
+
"learning_rate": 0.002276710911911739,
|
97 |
+
"loss": 3.4471,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 4.07,
|
102 |
+
"learning_rate": 0.0022250991208412345,
|
103 |
+
"loss": 3.4668,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 4.32,
|
108 |
+
"learning_rate": 0.0021733838993277022,
|
109 |
+
"loss": 3.4726,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 4.58,
|
114 |
+
"learning_rate": 0.00212166867781417,
|
115 |
+
"loss": 3.4374,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 4.83,
|
120 |
+
"learning_rate": 0.0020699534563006377,
|
121 |
+
"loss": 3.4422,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 5.08,
|
126 |
+
"learning_rate": 0.002018238234787106,
|
127 |
+
"loss": 3.4557,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 5.34,
|
132 |
+
"learning_rate": 0.0019666264437166008,
|
133 |
+
"loss": 3.4643,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 5.59,
|
138 |
+
"learning_rate": 0.0019150146526460954,
|
139 |
+
"loss": 3.4428,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 5.85,
|
144 |
+
"learning_rate": 0.0018632994311325636,
|
145 |
+
"loss": 3.4795,
|
146 |
+
"step": 11500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 6.1,
|
150 |
+
"learning_rate": 0.0018115842096190313,
|
151 |
+
"loss": 3.4469,
|
152 |
+
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 6.35,
|
156 |
+
"learning_rate": 0.001759868988105499,
|
157 |
+
"loss": 3.452,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 6.61,
|
162 |
+
"learning_rate": 0.001708153766591967,
|
163 |
+
"loss": 3.4436,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 6.86,
|
168 |
+
"learning_rate": 0.0016564385450784347,
|
169 |
+
"loss": 3.4325,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 7.12,
|
174 |
+
"learning_rate": 0.0016048267540079296,
|
175 |
+
"loss": 3.4705,
|
176 |
+
"step": 14000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 7.37,
|
180 |
+
"learning_rate": 0.0015531115324943976,
|
181 |
+
"loss": 3.432,
|
182 |
+
"step": 14500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 7.63,
|
186 |
+
"learning_rate": 0.0015013963109808653,
|
187 |
+
"loss": 3.4689,
|
188 |
+
"step": 15000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 7.88,
|
192 |
+
"learning_rate": 0.0014496810894673333,
|
193 |
+
"loss": 3.4853,
|
194 |
+
"step": 15500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 8.13,
|
198 |
+
"learning_rate": 0.001397965867953801,
|
199 |
+
"loss": 3.4548,
|
200 |
+
"step": 16000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 8.39,
|
204 |
+
"learning_rate": 0.001346354076883296,
|
205 |
+
"loss": 3.4377,
|
206 |
+
"step": 16500
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 8.64,
|
210 |
+
"learning_rate": 0.0012946388553697638,
|
211 |
+
"loss": 3.4421,
|
212 |
+
"step": 17000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 8.9,
|
216 |
+
"learning_rate": 0.0012429236338562318,
|
217 |
+
"loss": 3.437,
|
218 |
+
"step": 17500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 9.15,
|
222 |
+
"learning_rate": 0.0011912084123426995,
|
223 |
+
"loss": 3.474,
|
224 |
+
"step": 18000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 9.41,
|
228 |
+
"learning_rate": 0.0011394931908291675,
|
229 |
+
"loss": 3.4269,
|
230 |
+
"step": 18500
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 9.66,
|
234 |
+
"learning_rate": 0.0010877779693156354,
|
235 |
+
"loss": 3.4404,
|
236 |
+
"step": 19000
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 9.91,
|
240 |
+
"learning_rate": 0.0010361661782451303,
|
241 |
+
"loss": 3.4725,
|
242 |
+
"step": 19500
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 10.17,
|
246 |
+
"learning_rate": 0.000984450956731598,
|
247 |
+
"loss": 3.4633,
|
248 |
+
"step": 20000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 10.42,
|
252 |
+
"learning_rate": 0.0009327357352180659,
|
253 |
+
"loss": 3.4325,
|
254 |
+
"step": 20500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 10.68,
|
258 |
+
"learning_rate": 0.0008810205137045338,
|
259 |
+
"loss": 3.4446,
|
260 |
+
"step": 21000
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 10.93,
|
264 |
+
"learning_rate": 0.0008293052921910016,
|
265 |
+
"loss": 3.4433,
|
266 |
+
"step": 21500
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 11.18,
|
270 |
+
"learning_rate": 0.0007775900706774695,
|
271 |
+
"loss": 3.4662,
|
272 |
+
"step": 22000
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 11.44,
|
276 |
+
"learning_rate": 0.0007258748491639372,
|
277 |
+
"loss": 3.4421,
|
278 |
+
"step": 22500
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 11.69,
|
282 |
+
"learning_rate": 0.0006742630580934321,
|
283 |
+
"loss": 3.4358,
|
284 |
+
"step": 23000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 11.95,
|
288 |
+
"learning_rate": 0.0006225478365799,
|
289 |
+
"loss": 3.456,
|
290 |
+
"step": 23500
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 12.2,
|
294 |
+
"learning_rate": 0.0005708326150663679,
|
295 |
+
"loss": 3.4475,
|
296 |
+
"step": 24000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 12.46,
|
300 |
+
"learning_rate": 0.0005191173935528358,
|
301 |
+
"loss": 3.452,
|
302 |
+
"step": 24500
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 12.71,
|
306 |
+
"learning_rate": 0.0004674021720393036,
|
307 |
+
"loss": 3.4259,
|
308 |
+
"step": 25000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 12.96,
|
312 |
+
"learning_rate": 0.0004157903809687985,
|
313 |
+
"loss": 3.4611,
|
314 |
+
"step": 25500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 13.22,
|
318 |
+
"learning_rate": 0.0003640751594552663,
|
319 |
+
"loss": 3.4556,
|
320 |
+
"step": 26000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 13.47,
|
324 |
+
"learning_rate": 0.0003123599379417342,
|
325 |
+
"loss": 3.4278,
|
326 |
+
"step": 26500
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 13.73,
|
330 |
+
"learning_rate": 0.000260644716428202,
|
331 |
+
"loss": 3.4602,
|
332 |
+
"step": 27000
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 13.98,
|
336 |
+
"learning_rate": 0.0002089294949146699,
|
337 |
+
"loss": 3.4503,
|
338 |
+
"step": 27500
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 14.23,
|
342 |
+
"learning_rate": 0.00015721427340113775,
|
343 |
+
"loss": 3.4479,
|
344 |
+
"step": 28000
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 14.49,
|
348 |
+
"learning_rate": 0.00010560248233063266,
|
349 |
+
"loss": 3.4541,
|
350 |
+
"step": 28500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 14.74,
|
354 |
+
"learning_rate": 5.38872608171005e-05,
|
355 |
+
"loss": 3.4546,
|
356 |
+
"step": 29000
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 15.0,
|
360 |
+
"learning_rate": 2.1720393035683503e-06,
|
361 |
+
"loss": 3.4205,
|
362 |
+
"step": 29500
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 15.0,
|
366 |
+
"step": 29505,
|
367 |
+
"total_flos": 3.264158561621508e+18,
|
368 |
+
"train_loss": 3.4646308027512136,
|
369 |
+
"train_runtime": 5488.5308,
|
370 |
+
"train_samples_per_second": 10.752,
|
371 |
+
"train_steps_per_second": 5.376
|
372 |
+
}
|
373 |
+
],
|
374 |
+
"max_steps": 29505,
|
375 |
+
"num_train_epochs": 15,
|
376 |
+
"total_flos": 3.264158561621508e+18,
|
377 |
+
"trial_name": null,
|
378 |
+
"trial_params": null
|
379 |
+
}
|