Erik W
commited on
Commit
•
e0d496d
1
Parent(s):
50f966a
End of training
Browse files- all_results.json +12 -12
- eval_results.json +8 -8
- runs/Sep25_01-35-39_08ff2b062b9b/events.out.tfevents.1664072766.08ff2b062b9b.67.17 +3 -0
- train_results.json +4 -4
- trainer_state.json +92 -92
all_results.json
CHANGED
@@ -1,16 +1,16 @@
|
|
1 |
{
|
2 |
"epoch": 5.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_f1": 0.
|
5 |
-
"eval_loss": 0.
|
6 |
-
"eval_precision": 0.
|
7 |
-
"eval_recall": 0.
|
8 |
-
"eval_runtime": 22.
|
9 |
-
"eval_samples_per_second": 117.
|
10 |
-
"eval_steps_per_second": 1.
|
11 |
"total_flos": 3.953802662903808e+18,
|
12 |
-
"train_loss": 0.
|
13 |
-
"train_runtime":
|
14 |
-
"train_samples_per_second":
|
15 |
-
"train_steps_per_second": 0.
|
16 |
}
|
|
|
1 |
{
|
2 |
"epoch": 5.0,
|
3 |
+
"eval_accuracy": 0.9825925925925926,
|
4 |
+
"eval_f1": 0.9825868474705166,
|
5 |
+
"eval_loss": 0.0510183610022068,
|
6 |
+
"eval_precision": 0.9828193476192771,
|
7 |
+
"eval_recall": 0.9825925925925926,
|
8 |
+
"eval_runtime": 22.9683,
|
9 |
+
"eval_samples_per_second": 117.553,
|
10 |
+
"eval_steps_per_second": 1.872,
|
11 |
"total_flos": 3.953802662903808e+18,
|
12 |
+
"train_loss": 0.4062625066857589,
|
13 |
+
"train_runtime": 2798.1175,
|
14 |
+
"train_samples_per_second": 43.422,
|
15 |
+
"train_steps_per_second": 0.17
|
16 |
}
|
eval_results.json
CHANGED
@@ -1,11 +1,11 @@
|
|
1 |
{
|
2 |
"epoch": 5.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_f1": 0.
|
5 |
-
"eval_loss": 0.
|
6 |
-
"eval_precision": 0.
|
7 |
-
"eval_recall": 0.
|
8 |
-
"eval_runtime": 22.
|
9 |
-
"eval_samples_per_second": 117.
|
10 |
-
"eval_steps_per_second": 1.
|
11 |
}
|
|
|
1 |
{
|
2 |
"epoch": 5.0,
|
3 |
+
"eval_accuracy": 0.9825925925925926,
|
4 |
+
"eval_f1": 0.9825868474705166,
|
5 |
+
"eval_loss": 0.0510183610022068,
|
6 |
+
"eval_precision": 0.9828193476192771,
|
7 |
+
"eval_recall": 0.9825925925925926,
|
8 |
+
"eval_runtime": 22.9683,
|
9 |
+
"eval_samples_per_second": 117.553,
|
10 |
+
"eval_steps_per_second": 1.872
|
11 |
}
|
runs/Sep25_01-35-39_08ff2b062b9b/events.out.tfevents.1664072766.08ff2b062b9b.67.17
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5499bf0c0c4f05e961a73af14093e1000a91c08c0fe566ca9fbc3e2c9e4a6320
|
3 |
+
size 512
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 5.0,
|
3 |
"total_flos": 3.953802662903808e+18,
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 5.0,
|
3 |
"total_flos": 3.953802662903808e+18,
|
4 |
+
"train_loss": 0.4062625066857589,
|
5 |
+
"train_runtime": 2798.1175,
|
6 |
+
"train_samples_per_second": 43.422,
|
7 |
+
"train_steps_per_second": 0.17
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-eurosat/checkpoint-475",
|
4 |
"epoch": 5.0,
|
5 |
"global_step": 475,
|
@@ -10,353 +10,353 @@
|
|
10 |
{
|
11 |
"epoch": 0.11,
|
12 |
"learning_rate": 1.0526315789473684e-05,
|
13 |
-
"loss": 2.
|
14 |
"step": 10
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.21,
|
18 |
"learning_rate": 2.105263157894737e-05,
|
19 |
-
"loss": 2.
|
20 |
"step": 20
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.32,
|
24 |
"learning_rate": 3.157894736842105e-05,
|
25 |
-
"loss": 1.
|
26 |
"step": 30
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.42,
|
30 |
"learning_rate": 4.210526315789474e-05,
|
31 |
-
"loss":
|
32 |
"step": 40
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.53,
|
36 |
"learning_rate": 5.2631578947368424e-05,
|
37 |
-
"loss": 0.
|
38 |
"step": 50
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.63,
|
42 |
"learning_rate": 6.31578947368421e-05,
|
43 |
-
"loss": 0.
|
44 |
"step": 60
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.74,
|
48 |
"learning_rate": 7.368421052631579e-05,
|
49 |
-
"loss": 0.
|
50 |
"step": 70
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.84,
|
54 |
"learning_rate": 8.421052631578948e-05,
|
55 |
-
"loss": 0.
|
56 |
"step": 80
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.95,
|
60 |
"learning_rate": 9.473684210526316e-05,
|
61 |
-
"loss": 0.
|
62 |
"step": 90
|
63 |
},
|
64 |
{
|
65 |
"epoch": 1.0,
|
66 |
-
"eval_accuracy": 0.
|
67 |
-
"eval_f1": 0.
|
68 |
-
"eval_loss": 0.
|
69 |
-
"eval_precision": 0.
|
70 |
-
"eval_recall": 0.
|
71 |
-
"eval_runtime":
|
72 |
-
"eval_samples_per_second":
|
73 |
-
"eval_steps_per_second": 1.
|
74 |
"step": 95
|
75 |
},
|
76 |
{
|
77 |
"epoch": 1.05,
|
78 |
"learning_rate": 9.868421052631579e-05,
|
79 |
-
"loss": 0.
|
80 |
"step": 100
|
81 |
},
|
82 |
{
|
83 |
"epoch": 1.16,
|
84 |
"learning_rate": 9.605263157894737e-05,
|
85 |
-
"loss": 0.
|
86 |
"step": 110
|
87 |
},
|
88 |
{
|
89 |
"epoch": 1.26,
|
90 |
"learning_rate": 9.342105263157896e-05,
|
91 |
-
"loss": 0.
|
92 |
"step": 120
|
93 |
},
|
94 |
{
|
95 |
"epoch": 1.37,
|
96 |
"learning_rate": 9.078947368421054e-05,
|
97 |
-
"loss": 0.
|
98 |
"step": 130
|
99 |
},
|
100 |
{
|
101 |
"epoch": 1.47,
|
102 |
"learning_rate": 8.81578947368421e-05,
|
103 |
-
"loss": 0.
|
104 |
"step": 140
|
105 |
},
|
106 |
{
|
107 |
"epoch": 1.58,
|
108 |
"learning_rate": 8.552631578947369e-05,
|
109 |
-
"loss": 0.
|
110 |
"step": 150
|
111 |
},
|
112 |
{
|
113 |
"epoch": 1.68,
|
114 |
"learning_rate": 8.289473684210527e-05,
|
115 |
-
"loss": 0.
|
116 |
"step": 160
|
117 |
},
|
118 |
{
|
119 |
"epoch": 1.79,
|
120 |
"learning_rate": 8.026315789473685e-05,
|
121 |
-
"loss": 0.
|
122 |
"step": 170
|
123 |
},
|
124 |
{
|
125 |
"epoch": 1.89,
|
126 |
"learning_rate": 7.763157894736843e-05,
|
127 |
-
"loss": 0.
|
128 |
"step": 180
|
129 |
},
|
130 |
{
|
131 |
"epoch": 2.0,
|
132 |
"learning_rate": 7.500000000000001e-05,
|
133 |
-
"loss": 0.
|
134 |
"step": 190
|
135 |
},
|
136 |
{
|
137 |
"epoch": 2.0,
|
138 |
-
"eval_accuracy": 0.
|
139 |
-
"eval_f1": 0.
|
140 |
-
"eval_loss": 0.
|
141 |
-
"eval_precision": 0.
|
142 |
-
"eval_recall": 0.
|
143 |
-
"eval_runtime": 23.
|
144 |
-
"eval_samples_per_second": 115.
|
145 |
-
"eval_steps_per_second": 1.
|
146 |
"step": 190
|
147 |
},
|
148 |
{
|
149 |
"epoch": 2.11,
|
150 |
"learning_rate": 7.236842105263159e-05,
|
151 |
-
"loss": 0.
|
152 |
"step": 200
|
153 |
},
|
154 |
{
|
155 |
"epoch": 2.21,
|
156 |
"learning_rate": 6.973684210526315e-05,
|
157 |
-
"loss": 0.
|
158 |
"step": 210
|
159 |
},
|
160 |
{
|
161 |
"epoch": 2.32,
|
162 |
"learning_rate": 6.710526315789474e-05,
|
163 |
-
"loss": 0.
|
164 |
"step": 220
|
165 |
},
|
166 |
{
|
167 |
"epoch": 2.42,
|
168 |
"learning_rate": 6.447368421052632e-05,
|
169 |
-
"loss": 0.
|
170 |
"step": 230
|
171 |
},
|
172 |
{
|
173 |
"epoch": 2.53,
|
174 |
"learning_rate": 6.18421052631579e-05,
|
175 |
-
"loss": 0.
|
176 |
"step": 240
|
177 |
},
|
178 |
{
|
179 |
"epoch": 2.63,
|
180 |
"learning_rate": 5.921052631578947e-05,
|
181 |
-
"loss": 0.
|
182 |
"step": 250
|
183 |
},
|
184 |
{
|
185 |
"epoch": 2.74,
|
186 |
"learning_rate": 5.6578947368421056e-05,
|
187 |
-
"loss": 0.
|
188 |
"step": 260
|
189 |
},
|
190 |
{
|
191 |
"epoch": 2.84,
|
192 |
"learning_rate": 5.3947368421052635e-05,
|
193 |
-
"loss": 0.
|
194 |
"step": 270
|
195 |
},
|
196 |
{
|
197 |
"epoch": 2.95,
|
198 |
"learning_rate": 5.131578947368422e-05,
|
199 |
-
"loss": 0.
|
200 |
"step": 280
|
201 |
},
|
202 |
{
|
203 |
"epoch": 3.0,
|
204 |
-
"eval_accuracy": 0.
|
205 |
-
"eval_f1": 0.
|
206 |
-
"eval_loss": 0.
|
207 |
-
"eval_precision": 0.
|
208 |
-
"eval_recall": 0.
|
209 |
-
"eval_runtime": 23.
|
210 |
-
"eval_samples_per_second": 115.
|
211 |
-
"eval_steps_per_second": 1.
|
212 |
"step": 285
|
213 |
},
|
214 |
{
|
215 |
"epoch": 3.05,
|
216 |
"learning_rate": 4.868421052631579e-05,
|
217 |
-
"loss": 0.
|
218 |
"step": 290
|
219 |
},
|
220 |
{
|
221 |
"epoch": 3.16,
|
222 |
"learning_rate": 4.605263157894737e-05,
|
223 |
-
"loss": 0.
|
224 |
"step": 300
|
225 |
},
|
226 |
{
|
227 |
"epoch": 3.26,
|
228 |
"learning_rate": 4.342105263157895e-05,
|
229 |
-
"loss": 0.
|
230 |
"step": 310
|
231 |
},
|
232 |
{
|
233 |
"epoch": 3.37,
|
234 |
"learning_rate": 4.078947368421053e-05,
|
235 |
-
"loss": 0.
|
236 |
"step": 320
|
237 |
},
|
238 |
{
|
239 |
"epoch": 3.47,
|
240 |
"learning_rate": 3.815789473684211e-05,
|
241 |
-
"loss": 0.
|
242 |
"step": 330
|
243 |
},
|
244 |
{
|
245 |
"epoch": 3.58,
|
246 |
"learning_rate": 3.5526315789473684e-05,
|
247 |
-
"loss": 0.
|
248 |
"step": 340
|
249 |
},
|
250 |
{
|
251 |
"epoch": 3.68,
|
252 |
"learning_rate": 3.289473684210527e-05,
|
253 |
-
"loss": 0.
|
254 |
"step": 350
|
255 |
},
|
256 |
{
|
257 |
"epoch": 3.79,
|
258 |
"learning_rate": 3.0263157894736844e-05,
|
259 |
-
"loss": 0.
|
260 |
"step": 360
|
261 |
},
|
262 |
{
|
263 |
"epoch": 3.89,
|
264 |
"learning_rate": 2.7631578947368426e-05,
|
265 |
-
"loss": 0.
|
266 |
"step": 370
|
267 |
},
|
268 |
{
|
269 |
"epoch": 4.0,
|
270 |
"learning_rate": 2.5e-05,
|
271 |
-
"loss": 0.
|
272 |
"step": 380
|
273 |
},
|
274 |
{
|
275 |
"epoch": 4.0,
|
276 |
-
"eval_accuracy": 0.
|
277 |
-
"eval_f1": 0.
|
278 |
-
"eval_loss": 0.
|
279 |
-
"eval_precision": 0.
|
280 |
-
"eval_recall": 0.
|
281 |
-
"eval_runtime": 23.
|
282 |
-
"eval_samples_per_second": 115.
|
283 |
-
"eval_steps_per_second": 1.
|
284 |
"step": 380
|
285 |
},
|
286 |
{
|
287 |
"epoch": 4.11,
|
288 |
"learning_rate": 2.236842105263158e-05,
|
289 |
-
"loss": 0.
|
290 |
"step": 390
|
291 |
},
|
292 |
{
|
293 |
"epoch": 4.21,
|
294 |
"learning_rate": 1.9736842105263158e-05,
|
295 |
-
"loss": 0.
|
296 |
"step": 400
|
297 |
},
|
298 |
{
|
299 |
"epoch": 4.32,
|
300 |
"learning_rate": 1.7105263157894737e-05,
|
301 |
-
"loss": 0.
|
302 |
"step": 410
|
303 |
},
|
304 |
{
|
305 |
"epoch": 4.42,
|
306 |
"learning_rate": 1.4473684210526317e-05,
|
307 |
-
"loss": 0.
|
308 |
"step": 420
|
309 |
},
|
310 |
{
|
311 |
"epoch": 4.53,
|
312 |
"learning_rate": 1.1842105263157895e-05,
|
313 |
-
"loss": 0.
|
314 |
"step": 430
|
315 |
},
|
316 |
{
|
317 |
"epoch": 4.63,
|
318 |
"learning_rate": 9.210526315789474e-06,
|
319 |
-
"loss": 0.
|
320 |
"step": 440
|
321 |
},
|
322 |
{
|
323 |
"epoch": 4.74,
|
324 |
"learning_rate": 6.578947368421053e-06,
|
325 |
-
"loss": 0.
|
326 |
"step": 450
|
327 |
},
|
328 |
{
|
329 |
"epoch": 4.84,
|
330 |
"learning_rate": 3.9473684210526315e-06,
|
331 |
-
"loss": 0.
|
332 |
"step": 460
|
333 |
},
|
334 |
{
|
335 |
"epoch": 4.95,
|
336 |
"learning_rate": 1.3157894736842106e-06,
|
337 |
-
"loss": 0.
|
338 |
"step": 470
|
339 |
},
|
340 |
{
|
341 |
"epoch": 5.0,
|
342 |
-
"eval_accuracy": 0.
|
343 |
-
"eval_f1": 0.
|
344 |
-
"eval_loss": 0.
|
345 |
-
"eval_precision": 0.
|
346 |
-
"eval_recall": 0.
|
347 |
-
"eval_runtime": 23.
|
348 |
-
"eval_samples_per_second": 115.
|
349 |
-
"eval_steps_per_second": 1.
|
350 |
"step": 475
|
351 |
},
|
352 |
{
|
353 |
"epoch": 5.0,
|
354 |
"step": 475,
|
355 |
"total_flos": 3.953802662903808e+18,
|
356 |
-
"train_loss": 0.
|
357 |
-
"train_runtime":
|
358 |
-
"train_samples_per_second":
|
359 |
-
"train_steps_per_second": 0.
|
360 |
}
|
361 |
],
|
362 |
"max_steps": 475,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.9825925925925926,
|
3 |
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-eurosat/checkpoint-475",
|
4 |
"epoch": 5.0,
|
5 |
"global_step": 475,
|
|
|
10 |
{
|
11 |
"epoch": 0.11,
|
12 |
"learning_rate": 1.0526315789473684e-05,
|
13 |
+
"loss": 2.3113,
|
14 |
"step": 10
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.21,
|
18 |
"learning_rate": 2.105263157894737e-05,
|
19 |
+
"loss": 2.1411,
|
20 |
"step": 20
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.32,
|
24 |
"learning_rate": 3.157894736842105e-05,
|
25 |
+
"loss": 1.715,
|
26 |
"step": 30
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.42,
|
30 |
"learning_rate": 4.210526315789474e-05,
|
31 |
+
"loss": 1.0832,
|
32 |
"step": 40
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.53,
|
36 |
"learning_rate": 5.2631578947368424e-05,
|
37 |
+
"loss": 0.7166,
|
38 |
"step": 50
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.63,
|
42 |
"learning_rate": 6.31578947368421e-05,
|
43 |
+
"loss": 0.5886,
|
44 |
"step": 60
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.74,
|
48 |
"learning_rate": 7.368421052631579e-05,
|
49 |
+
"loss": 0.4958,
|
50 |
"step": 70
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.84,
|
54 |
"learning_rate": 8.421052631578948e-05,
|
55 |
+
"loss": 0.4257,
|
56 |
"step": 80
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.95,
|
60 |
"learning_rate": 9.473684210526316e-05,
|
61 |
+
"loss": 0.4479,
|
62 |
"step": 90
|
63 |
},
|
64 |
{
|
65 |
"epoch": 1.0,
|
66 |
+
"eval_accuracy": 0.9477777777777778,
|
67 |
+
"eval_f1": 0.9478323359637986,
|
68 |
+
"eval_loss": 0.15920470654964447,
|
69 |
+
"eval_precision": 0.9500303173429833,
|
70 |
+
"eval_recall": 0.9477777777777778,
|
71 |
+
"eval_runtime": 23.3166,
|
72 |
+
"eval_samples_per_second": 115.798,
|
73 |
+
"eval_steps_per_second": 1.844,
|
74 |
"step": 95
|
75 |
},
|
76 |
{
|
77 |
"epoch": 1.05,
|
78 |
"learning_rate": 9.868421052631579e-05,
|
79 |
+
"loss": 0.4166,
|
80 |
"step": 100
|
81 |
},
|
82 |
{
|
83 |
"epoch": 1.16,
|
84 |
"learning_rate": 9.605263157894737e-05,
|
85 |
+
"loss": 0.3648,
|
86 |
"step": 110
|
87 |
},
|
88 |
{
|
89 |
"epoch": 1.26,
|
90 |
"learning_rate": 9.342105263157896e-05,
|
91 |
+
"loss": 0.3384,
|
92 |
"step": 120
|
93 |
},
|
94 |
{
|
95 |
"epoch": 1.37,
|
96 |
"learning_rate": 9.078947368421054e-05,
|
97 |
+
"loss": 0.3095,
|
98 |
"step": 130
|
99 |
},
|
100 |
{
|
101 |
"epoch": 1.47,
|
102 |
"learning_rate": 8.81578947368421e-05,
|
103 |
+
"loss": 0.3676,
|
104 |
"step": 140
|
105 |
},
|
106 |
{
|
107 |
"epoch": 1.58,
|
108 |
"learning_rate": 8.552631578947369e-05,
|
109 |
+
"loss": 0.3062,
|
110 |
"step": 150
|
111 |
},
|
112 |
{
|
113 |
"epoch": 1.68,
|
114 |
"learning_rate": 8.289473684210527e-05,
|
115 |
+
"loss": 0.3251,
|
116 |
"step": 160
|
117 |
},
|
118 |
{
|
119 |
"epoch": 1.79,
|
120 |
"learning_rate": 8.026315789473685e-05,
|
121 |
+
"loss": 0.2973,
|
122 |
"step": 170
|
123 |
},
|
124 |
{
|
125 |
"epoch": 1.89,
|
126 |
"learning_rate": 7.763157894736843e-05,
|
127 |
+
"loss": 0.2848,
|
128 |
"step": 180
|
129 |
},
|
130 |
{
|
131 |
"epoch": 2.0,
|
132 |
"learning_rate": 7.500000000000001e-05,
|
133 |
+
"loss": 0.3078,
|
134 |
"step": 190
|
135 |
},
|
136 |
{
|
137 |
"epoch": 2.0,
|
138 |
+
"eval_accuracy": 0.9685185185185186,
|
139 |
+
"eval_f1": 0.9685839286713788,
|
140 |
+
"eval_loss": 0.09139837324619293,
|
141 |
+
"eval_precision": 0.9694937388736072,
|
142 |
+
"eval_recall": 0.9685185185185186,
|
143 |
+
"eval_runtime": 23.2994,
|
144 |
+
"eval_samples_per_second": 115.883,
|
145 |
+
"eval_steps_per_second": 1.846,
|
146 |
"step": 190
|
147 |
},
|
148 |
{
|
149 |
"epoch": 2.11,
|
150 |
"learning_rate": 7.236842105263159e-05,
|
151 |
+
"loss": 0.2692,
|
152 |
"step": 200
|
153 |
},
|
154 |
{
|
155 |
"epoch": 2.21,
|
156 |
"learning_rate": 6.973684210526315e-05,
|
157 |
+
"loss": 0.2636,
|
158 |
"step": 210
|
159 |
},
|
160 |
{
|
161 |
"epoch": 2.32,
|
162 |
"learning_rate": 6.710526315789474e-05,
|
163 |
+
"loss": 0.2537,
|
164 |
"step": 220
|
165 |
},
|
166 |
{
|
167 |
"epoch": 2.42,
|
168 |
"learning_rate": 6.447368421052632e-05,
|
169 |
+
"loss": 0.2531,
|
170 |
"step": 230
|
171 |
},
|
172 |
{
|
173 |
"epoch": 2.53,
|
174 |
"learning_rate": 6.18421052631579e-05,
|
175 |
+
"loss": 0.265,
|
176 |
"step": 240
|
177 |
},
|
178 |
{
|
179 |
"epoch": 2.63,
|
180 |
"learning_rate": 5.921052631578947e-05,
|
181 |
+
"loss": 0.2379,
|
182 |
"step": 250
|
183 |
},
|
184 |
{
|
185 |
"epoch": 2.74,
|
186 |
"learning_rate": 5.6578947368421056e-05,
|
187 |
+
"loss": 0.2156,
|
188 |
"step": 260
|
189 |
},
|
190 |
{
|
191 |
"epoch": 2.84,
|
192 |
"learning_rate": 5.3947368421052635e-05,
|
193 |
+
"loss": 0.2299,
|
194 |
"step": 270
|
195 |
},
|
196 |
{
|
197 |
"epoch": 2.95,
|
198 |
"learning_rate": 5.131578947368422e-05,
|
199 |
+
"loss": 0.2307,
|
200 |
"step": 280
|
201 |
},
|
202 |
{
|
203 |
"epoch": 3.0,
|
204 |
+
"eval_accuracy": 0.9785185185185186,
|
205 |
+
"eval_f1": 0.9784834463836688,
|
206 |
+
"eval_loss": 0.060329582542181015,
|
207 |
+
"eval_precision": 0.978983995320319,
|
208 |
+
"eval_recall": 0.9785185185185186,
|
209 |
+
"eval_runtime": 23.3466,
|
210 |
+
"eval_samples_per_second": 115.648,
|
211 |
+
"eval_steps_per_second": 1.842,
|
212 |
"step": 285
|
213 |
},
|
214 |
{
|
215 |
"epoch": 3.05,
|
216 |
"learning_rate": 4.868421052631579e-05,
|
217 |
+
"loss": 0.2046,
|
218 |
"step": 290
|
219 |
},
|
220 |
{
|
221 |
"epoch": 3.16,
|
222 |
"learning_rate": 4.605263157894737e-05,
|
223 |
+
"loss": 0.2185,
|
224 |
"step": 300
|
225 |
},
|
226 |
{
|
227 |
"epoch": 3.26,
|
228 |
"learning_rate": 4.342105263157895e-05,
|
229 |
+
"loss": 0.2206,
|
230 |
"step": 310
|
231 |
},
|
232 |
{
|
233 |
"epoch": 3.37,
|
234 |
"learning_rate": 4.078947368421053e-05,
|
235 |
+
"loss": 0.1843,
|
236 |
"step": 320
|
237 |
},
|
238 |
{
|
239 |
"epoch": 3.47,
|
240 |
"learning_rate": 3.815789473684211e-05,
|
241 |
+
"loss": 0.2162,
|
242 |
"step": 330
|
243 |
},
|
244 |
{
|
245 |
"epoch": 3.58,
|
246 |
"learning_rate": 3.5526315789473684e-05,
|
247 |
+
"loss": 0.1891,
|
248 |
"step": 340
|
249 |
},
|
250 |
{
|
251 |
"epoch": 3.68,
|
252 |
"learning_rate": 3.289473684210527e-05,
|
253 |
+
"loss": 0.2399,
|
254 |
"step": 350
|
255 |
},
|
256 |
{
|
257 |
"epoch": 3.79,
|
258 |
"learning_rate": 3.0263157894736844e-05,
|
259 |
+
"loss": 0.2171,
|
260 |
"step": 360
|
261 |
},
|
262 |
{
|
263 |
"epoch": 3.89,
|
264 |
"learning_rate": 2.7631578947368426e-05,
|
265 |
+
"loss": 0.1889,
|
266 |
"step": 370
|
267 |
},
|
268 |
{
|
269 |
"epoch": 4.0,
|
270 |
"learning_rate": 2.5e-05,
|
271 |
+
"loss": 0.227,
|
272 |
"step": 380
|
273 |
},
|
274 |
{
|
275 |
"epoch": 4.0,
|
276 |
+
"eval_accuracy": 0.9811111111111112,
|
277 |
+
"eval_f1": 0.9811027103905656,
|
278 |
+
"eval_loss": 0.053087469190359116,
|
279 |
+
"eval_precision": 0.9814045642951811,
|
280 |
+
"eval_recall": 0.9811111111111112,
|
281 |
+
"eval_runtime": 23.3477,
|
282 |
+
"eval_samples_per_second": 115.643,
|
283 |
+
"eval_steps_per_second": 1.842,
|
284 |
"step": 380
|
285 |
},
|
286 |
{
|
287 |
"epoch": 4.11,
|
288 |
"learning_rate": 2.236842105263158e-05,
|
289 |
+
"loss": 0.1926,
|
290 |
"step": 390
|
291 |
},
|
292 |
{
|
293 |
"epoch": 4.21,
|
294 |
"learning_rate": 1.9736842105263158e-05,
|
295 |
+
"loss": 0.1774,
|
296 |
"step": 400
|
297 |
},
|
298 |
{
|
299 |
"epoch": 4.32,
|
300 |
"learning_rate": 1.7105263157894737e-05,
|
301 |
+
"loss": 0.1884,
|
302 |
"step": 410
|
303 |
},
|
304 |
{
|
305 |
"epoch": 4.42,
|
306 |
"learning_rate": 1.4473684210526317e-05,
|
307 |
+
"loss": 0.1797,
|
308 |
"step": 420
|
309 |
},
|
310 |
{
|
311 |
"epoch": 4.53,
|
312 |
"learning_rate": 1.1842105263157895e-05,
|
313 |
+
"loss": 0.1612,
|
314 |
"step": 430
|
315 |
},
|
316 |
{
|
317 |
"epoch": 4.63,
|
318 |
"learning_rate": 9.210526315789474e-06,
|
319 |
+
"loss": 0.1793,
|
320 |
"step": 440
|
321 |
},
|
322 |
{
|
323 |
"epoch": 4.74,
|
324 |
"learning_rate": 6.578947368421053e-06,
|
325 |
+
"loss": 0.2003,
|
326 |
"step": 450
|
327 |
},
|
328 |
{
|
329 |
"epoch": 4.84,
|
330 |
"learning_rate": 3.9473684210526315e-06,
|
331 |
+
"loss": 0.1785,
|
332 |
"step": 460
|
333 |
},
|
334 |
{
|
335 |
"epoch": 4.95,
|
336 |
"learning_rate": 1.3157894736842106e-06,
|
337 |
+
"loss": 0.1674,
|
338 |
"step": 470
|
339 |
},
|
340 |
{
|
341 |
"epoch": 5.0,
|
342 |
+
"eval_accuracy": 0.9825925925925926,
|
343 |
+
"eval_f1": 0.9825868474705166,
|
344 |
+
"eval_loss": 0.0510183610022068,
|
345 |
+
"eval_precision": 0.9828193476192771,
|
346 |
+
"eval_recall": 0.9825925925925926,
|
347 |
+
"eval_runtime": 23.3012,
|
348 |
+
"eval_samples_per_second": 115.874,
|
349 |
+
"eval_steps_per_second": 1.845,
|
350 |
"step": 475
|
351 |
},
|
352 |
{
|
353 |
"epoch": 5.0,
|
354 |
"step": 475,
|
355 |
"total_flos": 3.953802662903808e+18,
|
356 |
+
"train_loss": 0.4062625066857589,
|
357 |
+
"train_runtime": 2798.1175,
|
358 |
+
"train_samples_per_second": 43.422,
|
359 |
+
"train_steps_per_second": 0.17
|
360 |
}
|
361 |
],
|
362 |
"max_steps": 475,
|