End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 40.0,
|
3 |
+
"eval_cer": 0.941403838019438,
|
4 |
+
"eval_loss": 3.0389626026153564,
|
5 |
+
"eval_runtime": 61.1643,
|
6 |
+
"eval_samples": 2551,
|
7 |
+
"eval_samples_per_second": 41.707,
|
8 |
+
"eval_steps_per_second": 2.616,
|
9 |
+
"eval_wer": 0.9994627877185007,
|
10 |
+
"train_loss": 1.875457509039636,
|
11 |
+
"train_runtime": 14726.9257,
|
12 |
+
"train_samples": 12748,
|
13 |
+
"train_samples_per_second": 34.625,
|
14 |
+
"train_steps_per_second": 2.165
|
15 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 40.0,
|
3 |
+
"eval_cer": 0.941403838019438,
|
4 |
+
"eval_loss": 3.0389626026153564,
|
5 |
+
"eval_runtime": 61.1643,
|
6 |
+
"eval_samples": 2551,
|
7 |
+
"eval_samples_per_second": 41.707,
|
8 |
+
"eval_steps_per_second": 2.616,
|
9 |
+
"eval_wer": 0.9994627877185007
|
10 |
+
}
|
runs/Feb20_11-52-23_khanhpluto-Z10PE-D8-WS/events.out.tfevents.1645347873.khanhpluto-Z10PE-D8-WS.40112.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8b6c18ed63dfae218c91ebf424b61d5ae07db260f0cc9c517591d60b14909991
|
3 |
+
size 412
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 40.0,
|
3 |
+
"train_loss": 1.875457509039636,
|
4 |
+
"train_runtime": 14726.9257,
|
5 |
+
"train_samples": 12748,
|
6 |
+
"train_samples_per_second": 34.625,
|
7 |
+
"train_steps_per_second": 2.165
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,403 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 40.0,
|
5 |
+
"global_step": 31880,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.63,
|
12 |
+
"learning_rate": 2.976e-05,
|
13 |
+
"loss": 7.9128,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 1.25,
|
18 |
+
"learning_rate": 2.9525812619502866e-05,
|
19 |
+
"loss": 3.5609,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.88,
|
24 |
+
"learning_rate": 2.9048757170172085e-05,
|
25 |
+
"loss": 3.3615,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 2.51,
|
30 |
+
"learning_rate": 2.857074569789675e-05,
|
31 |
+
"loss": 2.8318,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 3.14,
|
36 |
+
"learning_rate": 2.8093690248565966e-05,
|
37 |
+
"loss": 1.7371,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 3.76,
|
42 |
+
"learning_rate": 2.7615678776290633e-05,
|
43 |
+
"loss": 1.3703,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 4.39,
|
48 |
+
"learning_rate": 2.71395793499044e-05,
|
49 |
+
"loss": 1.1862,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 5.02,
|
54 |
+
"learning_rate": 2.6661567877629064e-05,
|
55 |
+
"loss": 1.0873,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 5.65,
|
60 |
+
"learning_rate": 2.618451242829828e-05,
|
61 |
+
"loss": 0.981,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 6.27,
|
66 |
+
"learning_rate": 2.5707456978967497e-05,
|
67 |
+
"loss": 0.9185,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 6.9,
|
72 |
+
"learning_rate": 2.522944550669216e-05,
|
73 |
+
"loss": 0.8622,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 7.53,
|
78 |
+
"learning_rate": 2.4752390057361376e-05,
|
79 |
+
"loss": 0.8181,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 8.16,
|
84 |
+
"learning_rate": 2.4274378585086042e-05,
|
85 |
+
"loss": 0.7863,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 8.78,
|
90 |
+
"learning_rate": 2.3797323135755258e-05,
|
91 |
+
"loss": 0.7409,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 9.41,
|
96 |
+
"learning_rate": 2.3319311663479924e-05,
|
97 |
+
"loss": 0.7223,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 10.04,
|
102 |
+
"learning_rate": 2.284225621414914e-05,
|
103 |
+
"loss": 0.7005,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 10.66,
|
108 |
+
"learning_rate": 2.2365200764818355e-05,
|
109 |
+
"loss": 0.6662,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 11.29,
|
114 |
+
"learning_rate": 2.188718929254302e-05,
|
115 |
+
"loss": 0.6458,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 11.92,
|
120 |
+
"learning_rate": 2.1410133843212236e-05,
|
121 |
+
"loss": 0.6246,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 12.55,
|
126 |
+
"learning_rate": 2.093307839388145e-05,
|
127 |
+
"loss": 0.5987,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 13.17,
|
132 |
+
"learning_rate": 2.0455066921606118e-05,
|
133 |
+
"loss": 0.5974,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 13.8,
|
138 |
+
"learning_rate": 1.9978011472275333e-05,
|
139 |
+
"loss": 0.5831,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 14.43,
|
144 |
+
"learning_rate": 1.9500956022944552e-05,
|
145 |
+
"loss": 0.5549,
|
146 |
+
"step": 11500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 15.06,
|
150 |
+
"learning_rate": 1.902294455066922e-05,
|
151 |
+
"loss": 0.561,
|
152 |
+
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 15.68,
|
156 |
+
"learning_rate": 1.8545889101338434e-05,
|
157 |
+
"loss": 0.5462,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 16.31,
|
162 |
+
"learning_rate": 1.806883365200765e-05,
|
163 |
+
"loss": 0.5491,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 16.94,
|
168 |
+
"learning_rate": 1.7590822179732315e-05,
|
169 |
+
"loss": 0.5594,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 17.57,
|
174 |
+
"learning_rate": 1.7112810707456982e-05,
|
175 |
+
"loss": 0.5565,
|
176 |
+
"step": 14000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 18.19,
|
180 |
+
"learning_rate": 1.6636711281070746e-05,
|
181 |
+
"loss": 0.5745,
|
182 |
+
"step": 14500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 18.82,
|
186 |
+
"learning_rate": 1.6158699808795412e-05,
|
187 |
+
"loss": 0.6096,
|
188 |
+
"step": 15000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 19.45,
|
192 |
+
"learning_rate": 1.568068833652008e-05,
|
193 |
+
"loss": 0.67,
|
194 |
+
"step": 15500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 20.08,
|
198 |
+
"learning_rate": 1.5203632887189294e-05,
|
199 |
+
"loss": 0.6566,
|
200 |
+
"step": 16000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 20.7,
|
204 |
+
"learning_rate": 1.4725621414913959e-05,
|
205 |
+
"loss": 0.6627,
|
206 |
+
"step": 16500
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 21.33,
|
210 |
+
"learning_rate": 1.4249521988527725e-05,
|
211 |
+
"loss": 0.6647,
|
212 |
+
"step": 17000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 21.96,
|
216 |
+
"learning_rate": 1.377151051625239e-05,
|
217 |
+
"loss": 0.9268,
|
218 |
+
"step": 17500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 22.58,
|
222 |
+
"learning_rate": 1.3293499043977056e-05,
|
223 |
+
"loss": 0.9965,
|
224 |
+
"step": 18000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 23.21,
|
228 |
+
"learning_rate": 1.2816443594646271e-05,
|
229 |
+
"loss": 0.9737,
|
230 |
+
"step": 18500
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 23.84,
|
234 |
+
"learning_rate": 1.2339388145315486e-05,
|
235 |
+
"loss": 1.4449,
|
236 |
+
"step": 19000
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 24.47,
|
240 |
+
"learning_rate": 1.1861376673040153e-05,
|
241 |
+
"loss": 2.0548,
|
242 |
+
"step": 19500
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 25.09,
|
246 |
+
"learning_rate": 1.138432122370937e-05,
|
247 |
+
"loss": 2.4726,
|
248 |
+
"step": 20000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 25.72,
|
252 |
+
"learning_rate": 1.0906309751434036e-05,
|
253 |
+
"loss": 2.7061,
|
254 |
+
"step": 20500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 26.35,
|
258 |
+
"learning_rate": 1.0430210325047802e-05,
|
259 |
+
"loss": 2.9211,
|
260 |
+
"step": 21000
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 26.98,
|
264 |
+
"learning_rate": 9.952198852772467e-06,
|
265 |
+
"loss": 2.9367,
|
266 |
+
"step": 21500
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 27.6,
|
270 |
+
"learning_rate": 9.474187380497133e-06,
|
271 |
+
"loss": 2.9015,
|
272 |
+
"step": 22000
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 28.23,
|
276 |
+
"learning_rate": 8.997131931166348e-06,
|
277 |
+
"loss": 2.9781,
|
278 |
+
"step": 22500
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 28.86,
|
282 |
+
"learning_rate": 8.520076481835564e-06,
|
283 |
+
"loss": 3.0789,
|
284 |
+
"step": 23000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 29.49,
|
288 |
+
"learning_rate": 8.042065009560229e-06,
|
289 |
+
"loss": 3.0514,
|
290 |
+
"step": 23500
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 30.11,
|
294 |
+
"learning_rate": 7.565009560229446e-06,
|
295 |
+
"loss": 3.0248,
|
296 |
+
"step": 24000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 30.74,
|
300 |
+
"learning_rate": 7.087954110898662e-06,
|
301 |
+
"loss": 2.9983,
|
302 |
+
"step": 24500
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 31.37,
|
306 |
+
"learning_rate": 6.609942638623327e-06,
|
307 |
+
"loss": 2.9716,
|
308 |
+
"step": 25000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 31.99,
|
312 |
+
"learning_rate": 6.1328871892925425e-06,
|
313 |
+
"loss": 2.9549,
|
314 |
+
"step": 25500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 32.62,
|
318 |
+
"learning_rate": 5.6558317399617595e-06,
|
319 |
+
"loss": 2.9551,
|
320 |
+
"step": 26000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 33.25,
|
324 |
+
"learning_rate": 5.177820267686425e-06,
|
325 |
+
"loss": 2.9571,
|
326 |
+
"step": 26500
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 33.88,
|
330 |
+
"learning_rate": 4.70076481835564e-06,
|
331 |
+
"loss": 2.9589,
|
332 |
+
"step": 27000
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 34.5,
|
336 |
+
"learning_rate": 4.2237093690248565e-06,
|
337 |
+
"loss": 2.9573,
|
338 |
+
"step": 27500
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 35.13,
|
342 |
+
"learning_rate": 3.7456978967495225e-06,
|
343 |
+
"loss": 2.9556,
|
344 |
+
"step": 28000
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 35.76,
|
348 |
+
"learning_rate": 3.268642447418738e-06,
|
349 |
+
"loss": 2.9545,
|
350 |
+
"step": 28500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 36.39,
|
354 |
+
"learning_rate": 2.7906309751434034e-06,
|
355 |
+
"loss": 2.9585,
|
356 |
+
"step": 29000
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 37.01,
|
360 |
+
"learning_rate": 2.3135755258126195e-06,
|
361 |
+
"loss": 2.9558,
|
362 |
+
"step": 29500
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 37.64,
|
366 |
+
"learning_rate": 1.8355640535372848e-06,
|
367 |
+
"loss": 2.9561,
|
368 |
+
"step": 30000
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 38.27,
|
372 |
+
"learning_rate": 1.358508604206501e-06,
|
373 |
+
"loss": 2.9597,
|
374 |
+
"step": 30500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 38.9,
|
378 |
+
"learning_rate": 8.81453154875717e-07,
|
379 |
+
"loss": 2.957,
|
380 |
+
"step": 31000
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 39.52,
|
384 |
+
"learning_rate": 4.0344168260038244e-07,
|
385 |
+
"loss": 2.9569,
|
386 |
+
"step": 31500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 40.0,
|
390 |
+
"step": 31880,
|
391 |
+
"total_flos": 2.3797340595067814e+19,
|
392 |
+
"train_loss": 1.875457509039636,
|
393 |
+
"train_runtime": 14726.9257,
|
394 |
+
"train_samples_per_second": 34.625,
|
395 |
+
"train_steps_per_second": 2.165
|
396 |
+
}
|
397 |
+
],
|
398 |
+
"max_steps": 31880,
|
399 |
+
"num_train_epochs": 40,
|
400 |
+
"total_flos": 2.3797340595067814e+19,
|
401 |
+
"trial_name": null,
|
402 |
+
"trial_params": null
|
403 |
+
}
|