Commit
•
a080dbd
1
Parent(s):
1b392e5
End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.07,
|
3 |
+
"eval_loss": 0.5947265625,
|
4 |
+
"eval_runtime": 1112.1793,
|
5 |
+
"eval_samples_per_second": 2.969,
|
6 |
+
"eval_steps_per_second": 0.186,
|
7 |
+
"eval_wer": 23.5020721174329,
|
8 |
+
"train_loss": 0.09166860008239747,
|
9 |
+
"train_runtime": 21210.666,
|
10 |
+
"train_samples_per_second": 3.017,
|
11 |
+
"train_steps_per_second": 0.047
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.07,
|
3 |
+
"eval_loss": 0.5947265625,
|
4 |
+
"eval_runtime": 1112.1793,
|
5 |
+
"eval_samples_per_second": 2.969,
|
6 |
+
"eval_steps_per_second": 0.186,
|
7 |
+
"eval_wer": 23.5020721174329
|
8 |
+
}
|
runs/Dec16_23-11-11_129-146-105-103/events.out.tfevents.1671254903.129-146-105-103.3099683.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8dc4e813b3b8b42433c25373e1277110ad89be409647c5d9b49d091a825c24da
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 9.07,
|
3 |
+
"train_loss": 0.09166860008239747,
|
4 |
+
"train_runtime": 21210.666,
|
5 |
+
"train_samples_per_second": 3.017,
|
6 |
+
"train_steps_per_second": 0.047
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 23.5020721174329,
|
3 |
+
"best_model_checkpoint": "./checkpoint-800",
|
4 |
+
"epoch": 9.073,
|
5 |
+
"global_step": 1000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 6.712113404111031e-06,
|
13 |
+
"loss": 0.5024,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 8.360489289678585e-06,
|
19 |
+
"loss": 0.3849,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.07,
|
24 |
+
"learning_rate": 9.28666248215634e-06,
|
25 |
+
"loss": 0.3349,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.1,
|
30 |
+
"learning_rate": 9.933858671331224e-06,
|
31 |
+
"loss": 0.3442,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.02,
|
36 |
+
"learning_rate": 9.766666666666667e-06,
|
37 |
+
"loss": 0.2877,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.05,
|
42 |
+
"learning_rate": 9.48888888888889e-06,
|
43 |
+
"loss": 0.2288,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.07,
|
48 |
+
"learning_rate": 9.211111111111111e-06,
|
49 |
+
"loss": 0.1896,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 1.1,
|
54 |
+
"learning_rate": 8.933333333333333e-06,
|
55 |
+
"loss": 0.1935,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 1.1,
|
60 |
+
"eval_loss": 0.424072265625,
|
61 |
+
"eval_runtime": 1295.5788,
|
62 |
+
"eval_samples_per_second": 2.549,
|
63 |
+
"eval_steps_per_second": 0.16,
|
64 |
+
"eval_wer": 29.652581432278552,
|
65 |
+
"step": 200
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 2.02,
|
69 |
+
"learning_rate": 8.644444444444445e-06,
|
70 |
+
"loss": 0.1725,
|
71 |
+
"step": 225
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 2.04,
|
75 |
+
"learning_rate": 8.366666666666667e-06,
|
76 |
+
"loss": 0.1351,
|
77 |
+
"step": 250
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 2.07,
|
81 |
+
"learning_rate": 8.08888888888889e-06,
|
82 |
+
"loss": 0.116,
|
83 |
+
"step": 275
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 2.09,
|
87 |
+
"learning_rate": 7.822222222222224e-06,
|
88 |
+
"loss": 0.1143,
|
89 |
+
"step": 300
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 3.02,
|
93 |
+
"learning_rate": 7.544444444444445e-06,
|
94 |
+
"loss": 0.1026,
|
95 |
+
"step": 325
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 3.04,
|
99 |
+
"learning_rate": 7.266666666666668e-06,
|
100 |
+
"loss": 0.0801,
|
101 |
+
"step": 350
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 3.07,
|
105 |
+
"learning_rate": 6.9888888888888895e-06,
|
106 |
+
"loss": 0.0655,
|
107 |
+
"step": 375
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 3.09,
|
111 |
+
"learning_rate": 6.711111111111111e-06,
|
112 |
+
"loss": 0.0649,
|
113 |
+
"step": 400
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 3.09,
|
117 |
+
"eval_loss": 0.46826171875,
|
118 |
+
"eval_runtime": 1242.5517,
|
119 |
+
"eval_samples_per_second": 2.657,
|
120 |
+
"eval_steps_per_second": 0.167,
|
121 |
+
"eval_wer": 26.062202254153917,
|
122 |
+
"step": 400
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 4.01,
|
126 |
+
"learning_rate": 6.422222222222223e-06,
|
127 |
+
"loss": 0.0575,
|
128 |
+
"step": 425
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 4.04,
|
132 |
+
"learning_rate": 6.155555555555556e-06,
|
133 |
+
"loss": 0.0485,
|
134 |
+
"step": 450
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 4.06,
|
138 |
+
"learning_rate": 5.877777777777778e-06,
|
139 |
+
"loss": 0.0347,
|
140 |
+
"step": 475
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 4.09,
|
144 |
+
"learning_rate": 5.600000000000001e-06,
|
145 |
+
"loss": 0.0331,
|
146 |
+
"step": 500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 5.01,
|
150 |
+
"learning_rate": 5.322222222222223e-06,
|
151 |
+
"loss": 0.0307,
|
152 |
+
"step": 525
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 5.04,
|
156 |
+
"learning_rate": 5.044444444444445e-06,
|
157 |
+
"loss": 0.0264,
|
158 |
+
"step": 550
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 5.06,
|
162 |
+
"learning_rate": 4.766666666666667e-06,
|
163 |
+
"loss": 0.0192,
|
164 |
+
"step": 575
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 5.08,
|
168 |
+
"learning_rate": 4.488888888888889e-06,
|
169 |
+
"loss": 0.0156,
|
170 |
+
"step": 600
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 5.08,
|
174 |
+
"eval_loss": 0.54443359375,
|
175 |
+
"eval_runtime": 1180.1513,
|
176 |
+
"eval_samples_per_second": 2.798,
|
177 |
+
"eval_steps_per_second": 0.175,
|
178 |
+
"eval_wer": 25.810449668848523,
|
179 |
+
"step": 600
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 6.01,
|
183 |
+
"learning_rate": 4.2000000000000004e-06,
|
184 |
+
"loss": 0.0136,
|
185 |
+
"step": 625
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 6.03,
|
189 |
+
"learning_rate": 3.922222222222223e-06,
|
190 |
+
"loss": 0.0119,
|
191 |
+
"step": 650
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 6.06,
|
195 |
+
"learning_rate": 3.644444444444445e-06,
|
196 |
+
"loss": 0.0098,
|
197 |
+
"step": 675
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 6.08,
|
201 |
+
"learning_rate": 3.366666666666667e-06,
|
202 |
+
"loss": 0.0071,
|
203 |
+
"step": 700
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 7.0,
|
207 |
+
"learning_rate": 3.088888888888889e-06,
|
208 |
+
"loss": 0.0069,
|
209 |
+
"step": 725
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 7.03,
|
213 |
+
"learning_rate": 2.811111111111111e-06,
|
214 |
+
"loss": 0.0062,
|
215 |
+
"step": 750
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 7.05,
|
219 |
+
"learning_rate": 2.5333333333333338e-06,
|
220 |
+
"loss": 0.0046,
|
221 |
+
"step": 775
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 7.08,
|
225 |
+
"learning_rate": 2.2555555555555557e-06,
|
226 |
+
"loss": 0.0039,
|
227 |
+
"step": 800
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 7.08,
|
231 |
+
"eval_loss": 0.5947265625,
|
232 |
+
"eval_runtime": 1100.8609,
|
233 |
+
"eval_samples_per_second": 2.999,
|
234 |
+
"eval_steps_per_second": 0.188,
|
235 |
+
"eval_wer": 23.5020721174329,
|
236 |
+
"step": 800
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 8.0,
|
240 |
+
"learning_rate": 1.9666666666666668e-06,
|
241 |
+
"loss": 0.0039,
|
242 |
+
"step": 825
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 8.03,
|
246 |
+
"learning_rate": 1.688888888888889e-06,
|
247 |
+
"loss": 0.0031,
|
248 |
+
"step": 850
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 8.05,
|
252 |
+
"learning_rate": 1.4111111111111111e-06,
|
253 |
+
"loss": 0.0026,
|
254 |
+
"step": 875
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 8.08,
|
258 |
+
"learning_rate": 1.1333333333333334e-06,
|
259 |
+
"loss": 0.0023,
|
260 |
+
"step": 900
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 8.1,
|
264 |
+
"learning_rate": 8.555555555555556e-07,
|
265 |
+
"loss": 0.0023,
|
266 |
+
"step": 925
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 9.02,
|
270 |
+
"learning_rate": 5.777777777777778e-07,
|
271 |
+
"loss": 0.0021,
|
272 |
+
"step": 950
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 9.05,
|
276 |
+
"learning_rate": 3.0000000000000004e-07,
|
277 |
+
"loss": 0.002,
|
278 |
+
"step": 975
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 9.07,
|
282 |
+
"learning_rate": 2.2222222222222224e-08,
|
283 |
+
"loss": 0.0019,
|
284 |
+
"step": 1000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 9.07,
|
288 |
+
"eval_loss": 0.6123046875,
|
289 |
+
"eval_runtime": 1135.4108,
|
290 |
+
"eval_samples_per_second": 2.908,
|
291 |
+
"eval_steps_per_second": 0.182,
|
292 |
+
"eval_wer": 23.893256903830512,
|
293 |
+
"step": 1000
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 9.07,
|
297 |
+
"step": 1000,
|
298 |
+
"total_flos": 1.3615046685661279e+20,
|
299 |
+
"train_loss": 0.09166860008239747,
|
300 |
+
"train_runtime": 21210.666,
|
301 |
+
"train_samples_per_second": 3.017,
|
302 |
+
"train_steps_per_second": 0.047
|
303 |
+
}
|
304 |
+
],
|
305 |
+
"max_steps": 1000,
|
306 |
+
"num_train_epochs": 9223372036854775807,
|
307 |
+
"total_flos": 1.3615046685661279e+20,
|
308 |
+
"trial_name": null,
|
309 |
+
"trial_params": null
|
310 |
+
}
|