End of training
Browse files- trainer_state.json +337 -0
trainer_state.json
ADDED
@@ -0,0 +1,337 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.9642857142857143,
|
3 |
+
"best_model_checkpoint": "ast-finetuned-audioset-10-10-0.4593_ft_env_aug_0-2\\checkpoint-84",
|
4 |
+
"epoch": 10.0,
|
5 |
+
"global_step": 280,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.36,
|
12 |
+
"learning_rate": 7.142857142857143e-07,
|
13 |
+
"loss": 2.1288,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.71,
|
18 |
+
"learning_rate": 1.4285714285714286e-06,
|
19 |
+
"loss": 2.0165,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.0,
|
24 |
+
"eval_accuracy": 0.4642857142857143,
|
25 |
+
"eval_f1": 0.4710884353741497,
|
26 |
+
"eval_loss": 1.6251839399337769,
|
27 |
+
"eval_precision": 0.5373376623376623,
|
28 |
+
"eval_recall": 0.4642857142857143,
|
29 |
+
"eval_runtime": 2.5452,
|
30 |
+
"eval_samples_per_second": 11.001,
|
31 |
+
"eval_steps_per_second": 5.501,
|
32 |
+
"step": 28
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.07,
|
36 |
+
"learning_rate": 1.984126984126984e-06,
|
37 |
+
"loss": 1.816,
|
38 |
+
"step": 30
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.43,
|
42 |
+
"learning_rate": 1.9047619047619045e-06,
|
43 |
+
"loss": 1.5466,
|
44 |
+
"step": 40
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.79,
|
48 |
+
"learning_rate": 1.8253968253968254e-06,
|
49 |
+
"loss": 1.3702,
|
50 |
+
"step": 50
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 2.0,
|
54 |
+
"eval_accuracy": 0.8571428571428571,
|
55 |
+
"eval_f1": 0.8535714285714285,
|
56 |
+
"eval_loss": 1.0553056001663208,
|
57 |
+
"eval_precision": 0.8928571428571429,
|
58 |
+
"eval_recall": 0.8571428571428571,
|
59 |
+
"eval_runtime": 2.5524,
|
60 |
+
"eval_samples_per_second": 10.97,
|
61 |
+
"eval_steps_per_second": 5.485,
|
62 |
+
"step": 56
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.14,
|
66 |
+
"learning_rate": 1.746031746031746e-06,
|
67 |
+
"loss": 1.2834,
|
68 |
+
"step": 60
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.5,
|
72 |
+
"learning_rate": 1.6666666666666667e-06,
|
73 |
+
"loss": 0.9612,
|
74 |
+
"step": 70
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 2.86,
|
78 |
+
"learning_rate": 1.587301587301587e-06,
|
79 |
+
"loss": 0.8861,
|
80 |
+
"step": 80
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 3.0,
|
84 |
+
"eval_accuracy": 0.9642857142857143,
|
85 |
+
"eval_f1": 0.9631083202511775,
|
86 |
+
"eval_loss": 0.6899059414863586,
|
87 |
+
"eval_precision": 0.9693877551020408,
|
88 |
+
"eval_recall": 0.9642857142857143,
|
89 |
+
"eval_runtime": 2.5679,
|
90 |
+
"eval_samples_per_second": 10.904,
|
91 |
+
"eval_steps_per_second": 5.452,
|
92 |
+
"step": 84
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 3.21,
|
96 |
+
"learning_rate": 1.5079365079365078e-06,
|
97 |
+
"loss": 0.7961,
|
98 |
+
"step": 90
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 3.57,
|
102 |
+
"learning_rate": 1.4285714285714286e-06,
|
103 |
+
"loss": 0.7187,
|
104 |
+
"step": 100
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 3.93,
|
108 |
+
"learning_rate": 1.3492063492063493e-06,
|
109 |
+
"loss": 0.5655,
|
110 |
+
"step": 110
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 4.0,
|
114 |
+
"eval_accuracy": 0.9642857142857143,
|
115 |
+
"eval_f1": 0.9631083202511775,
|
116 |
+
"eval_loss": 0.47662997245788574,
|
117 |
+
"eval_precision": 0.9693877551020408,
|
118 |
+
"eval_recall": 0.9642857142857143,
|
119 |
+
"eval_runtime": 2.6044,
|
120 |
+
"eval_samples_per_second": 10.751,
|
121 |
+
"eval_steps_per_second": 5.376,
|
122 |
+
"step": 112
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 4.29,
|
126 |
+
"learning_rate": 1.2698412698412697e-06,
|
127 |
+
"loss": 0.4864,
|
128 |
+
"step": 120
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 4.64,
|
132 |
+
"learning_rate": 1.1904761904761904e-06,
|
133 |
+
"loss": 0.4195,
|
134 |
+
"step": 130
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 5.0,
|
138 |
+
"learning_rate": 1.111111111111111e-06,
|
139 |
+
"loss": 0.4232,
|
140 |
+
"step": 140
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 5.0,
|
144 |
+
"eval_accuracy": 0.9642857142857143,
|
145 |
+
"eval_f1": 0.9631083202511775,
|
146 |
+
"eval_loss": 0.3402732312679291,
|
147 |
+
"eval_precision": 0.9693877551020408,
|
148 |
+
"eval_recall": 0.9642857142857143,
|
149 |
+
"eval_runtime": 2.6311,
|
150 |
+
"eval_samples_per_second": 10.642,
|
151 |
+
"eval_steps_per_second": 5.321,
|
152 |
+
"step": 140
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 5.36,
|
156 |
+
"learning_rate": 1.031746031746032e-06,
|
157 |
+
"loss": 0.3117,
|
158 |
+
"step": 150
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 5.71,
|
162 |
+
"learning_rate": 9.523809523809522e-07,
|
163 |
+
"loss": 0.3148,
|
164 |
+
"step": 160
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 6.0,
|
168 |
+
"eval_accuracy": 0.9642857142857143,
|
169 |
+
"eval_f1": 0.9631083202511775,
|
170 |
+
"eval_loss": 0.2678673565387726,
|
171 |
+
"eval_precision": 0.9693877551020408,
|
172 |
+
"eval_recall": 0.9642857142857143,
|
173 |
+
"eval_runtime": 2.5813,
|
174 |
+
"eval_samples_per_second": 10.847,
|
175 |
+
"eval_steps_per_second": 5.424,
|
176 |
+
"step": 168
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 6.07,
|
180 |
+
"learning_rate": 8.73015873015873e-07,
|
181 |
+
"loss": 0.3124,
|
182 |
+
"step": 170
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 6.43,
|
186 |
+
"learning_rate": 7.936507936507936e-07,
|
187 |
+
"loss": 0.2345,
|
188 |
+
"step": 180
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 6.79,
|
192 |
+
"learning_rate": 7.142857142857143e-07,
|
193 |
+
"loss": 0.2335,
|
194 |
+
"step": 190
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 7.0,
|
198 |
+
"eval_accuracy": 0.9642857142857143,
|
199 |
+
"eval_f1": 0.9631083202511775,
|
200 |
+
"eval_loss": 0.22391526401042938,
|
201 |
+
"eval_precision": 0.9693877551020408,
|
202 |
+
"eval_recall": 0.9642857142857143,
|
203 |
+
"eval_runtime": 2.5943,
|
204 |
+
"eval_samples_per_second": 10.793,
|
205 |
+
"eval_steps_per_second": 5.397,
|
206 |
+
"step": 196
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 7.14,
|
210 |
+
"learning_rate": 6.349206349206349e-07,
|
211 |
+
"loss": 0.2206,
|
212 |
+
"step": 200
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 7.5,
|
216 |
+
"learning_rate": 5.555555555555555e-07,
|
217 |
+
"loss": 0.1858,
|
218 |
+
"step": 210
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 7.86,
|
222 |
+
"learning_rate": 4.761904761904761e-07,
|
223 |
+
"loss": 0.176,
|
224 |
+
"step": 220
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 8.0,
|
228 |
+
"eval_accuracy": 0.9642857142857143,
|
229 |
+
"eval_f1": 0.9631083202511775,
|
230 |
+
"eval_loss": 0.19793787598609924,
|
231 |
+
"eval_precision": 0.9693877551020408,
|
232 |
+
"eval_recall": 0.9642857142857143,
|
233 |
+
"eval_runtime": 2.6109,
|
234 |
+
"eval_samples_per_second": 10.724,
|
235 |
+
"eval_steps_per_second": 5.362,
|
236 |
+
"step": 224
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 8.21,
|
240 |
+
"learning_rate": 3.968253968253968e-07,
|
241 |
+
"loss": 0.1796,
|
242 |
+
"step": 230
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 8.57,
|
246 |
+
"learning_rate": 3.1746031746031743e-07,
|
247 |
+
"loss": 0.1475,
|
248 |
+
"step": 240
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 8.93,
|
252 |
+
"learning_rate": 2.3809523809523806e-07,
|
253 |
+
"loss": 0.1624,
|
254 |
+
"step": 250
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 9.0,
|
258 |
+
"eval_accuracy": 0.9642857142857143,
|
259 |
+
"eval_f1": 0.9631083202511775,
|
260 |
+
"eval_loss": 0.1823665350675583,
|
261 |
+
"eval_precision": 0.9693877551020408,
|
262 |
+
"eval_recall": 0.9642857142857143,
|
263 |
+
"eval_runtime": 2.6135,
|
264 |
+
"eval_samples_per_second": 10.714,
|
265 |
+
"eval_steps_per_second": 5.357,
|
266 |
+
"step": 252
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 9.29,
|
270 |
+
"learning_rate": 1.5873015873015872e-07,
|
271 |
+
"loss": 0.1532,
|
272 |
+
"step": 260
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 9.64,
|
276 |
+
"learning_rate": 7.936507936507936e-08,
|
277 |
+
"loss": 0.1549,
|
278 |
+
"step": 270
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 10.0,
|
282 |
+
"learning_rate": 0.0,
|
283 |
+
"loss": 0.1466,
|
284 |
+
"step": 280
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 10.0,
|
288 |
+
"eval_accuracy": 0.9642857142857143,
|
289 |
+
"eval_f1": 0.9631083202511775,
|
290 |
+
"eval_loss": 0.17811881005764008,
|
291 |
+
"eval_precision": 0.9693877551020408,
|
292 |
+
"eval_recall": 0.9642857142857143,
|
293 |
+
"eval_runtime": 2.5988,
|
294 |
+
"eval_samples_per_second": 10.774,
|
295 |
+
"eval_steps_per_second": 5.387,
|
296 |
+
"step": 280
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 10.0,
|
300 |
+
"step": 280,
|
301 |
+
"total_flos": 1.5184033474412544e+17,
|
302 |
+
"train_loss": 0.6554145208426885,
|
303 |
+
"train_runtime": 635.0581,
|
304 |
+
"train_samples_per_second": 3.527,
|
305 |
+
"train_steps_per_second": 0.441
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 10.0,
|
309 |
+
"eval_accuracy": 0.9642857142857143,
|
310 |
+
"eval_f1": 0.9631083202511775,
|
311 |
+
"eval_loss": 0.6899059414863586,
|
312 |
+
"eval_precision": 0.9693877551020408,
|
313 |
+
"eval_recall": 0.9642857142857143,
|
314 |
+
"eval_runtime": 2.7776,
|
315 |
+
"eval_samples_per_second": 10.081,
|
316 |
+
"eval_steps_per_second": 5.04,
|
317 |
+
"step": 280
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 10.0,
|
321 |
+
"step": 280,
|
322 |
+
"test_accuracy": 0.8928571428571429,
|
323 |
+
"test_f1": 0.8869047619047619,
|
324 |
+
"test_loss": 0.8998350501060486,
|
325 |
+
"test_precision": 0.9345238095238095,
|
326 |
+
"test_recall": 0.8928571428571429,
|
327 |
+
"test_runtime": 2.9018,
|
328 |
+
"test_samples_per_second": 9.649,
|
329 |
+
"test_steps_per_second": 4.825
|
330 |
+
}
|
331 |
+
],
|
332 |
+
"max_steps": 280,
|
333 |
+
"num_train_epochs": 10,
|
334 |
+
"total_flos": 1.5184033474412544e+17,
|
335 |
+
"trial_name": null,
|
336 |
+
"trial_params": null
|
337 |
+
}
|