Commit
•
54d11d6
1
Parent(s):
ff97657
End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 30.0,
|
3 |
+
"eval_accuracy": 1.0,
|
4 |
+
"eval_loss": 0.007612039800733328,
|
5 |
+
"eval_runtime": 11.0324,
|
6 |
+
"eval_samples_per_second": 79.765,
|
7 |
+
"eval_steps_per_second": 4.985,
|
8 |
+
"total_flos": 2.4304114274567455e+19,
|
9 |
+
"train_loss": 0.19108782262513133,
|
10 |
+
"train_runtime": 4786.2205,
|
11 |
+
"train_samples_per_second": 22.051,
|
12 |
+
"train_steps_per_second": 0.345
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 30.0,
|
3 |
+
"eval_accuracy": 1.0,
|
4 |
+
"eval_loss": 0.007612039800733328,
|
5 |
+
"eval_runtime": 11.0324,
|
6 |
+
"eval_samples_per_second": 79.765,
|
7 |
+
"eval_steps_per_second": 4.985
|
8 |
+
}
|
runs/Jul04_20-03-48_2df985b450cf/events.out.tfevents.1688506871.2df985b450cf.27022.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dfdedd241bc0b6659f02f6cb33a3d97ab1b5f4c35669b5e72c120f85b058edc0
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 30.0,
|
3 |
+
"total_flos": 2.4304114274567455e+19,
|
4 |
+
"train_loss": 0.19108782262513133,
|
5 |
+
"train_runtime": 4786.2205,
|
6 |
+
"train_samples_per_second": 22.051,
|
7 |
+
"train_steps_per_second": 0.345
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,787 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.0,
|
3 |
+
"best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50/checkpoint-495",
|
4 |
+
"epoch": 30.0,
|
5 |
+
"global_step": 1650,
|
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": 7.228915662650602e-06,
|
13 |
+
"loss": 1.9251,
|
14 |
+
"step": 20
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.73,
|
18 |
+
"learning_rate": 1.4457831325301205e-05,
|
19 |
+
"loss": 1.5952,
|
20 |
+
"step": 40
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.0,
|
24 |
+
"eval_accuracy": 0.6693181818181818,
|
25 |
+
"eval_loss": 0.848971426486969,
|
26 |
+
"eval_runtime": 10.9926,
|
27 |
+
"eval_samples_per_second": 80.054,
|
28 |
+
"eval_steps_per_second": 5.003,
|
29 |
+
"step": 55
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 1.09,
|
33 |
+
"learning_rate": 2.168674698795181e-05,
|
34 |
+
"loss": 1.1596,
|
35 |
+
"step": 60
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 1.45,
|
39 |
+
"learning_rate": 2.891566265060241e-05,
|
40 |
+
"loss": 0.8796,
|
41 |
+
"step": 80
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 1.82,
|
45 |
+
"learning_rate": 2.967453733248245e-05,
|
46 |
+
"loss": 0.7582,
|
47 |
+
"step": 100
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 2.0,
|
51 |
+
"eval_accuracy": 0.8386363636363636,
|
52 |
+
"eval_loss": 0.45608994364738464,
|
53 |
+
"eval_runtime": 11.0458,
|
54 |
+
"eval_samples_per_second": 79.668,
|
55 |
+
"eval_steps_per_second": 4.979,
|
56 |
+
"step": 110
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 2.18,
|
60 |
+
"learning_rate": 2.929164007657945e-05,
|
61 |
+
"loss": 0.5899,
|
62 |
+
"step": 120
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.55,
|
66 |
+
"learning_rate": 2.8908742820676455e-05,
|
67 |
+
"loss": 0.4953,
|
68 |
+
"step": 140
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.91,
|
72 |
+
"learning_rate": 2.852584556477345e-05,
|
73 |
+
"loss": 0.4359,
|
74 |
+
"step": 160
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 3.0,
|
78 |
+
"eval_accuracy": 0.9227272727272727,
|
79 |
+
"eval_loss": 0.24076080322265625,
|
80 |
+
"eval_runtime": 10.9212,
|
81 |
+
"eval_samples_per_second": 80.577,
|
82 |
+
"eval_steps_per_second": 5.036,
|
83 |
+
"step": 165
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 3.27,
|
87 |
+
"learning_rate": 2.8142948308870455e-05,
|
88 |
+
"loss": 0.4,
|
89 |
+
"step": 180
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 3.64,
|
93 |
+
"learning_rate": 2.7760051052967455e-05,
|
94 |
+
"loss": 0.361,
|
95 |
+
"step": 200
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 4.0,
|
99 |
+
"learning_rate": 2.7377153797064455e-05,
|
100 |
+
"loss": 0.318,
|
101 |
+
"step": 220
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 4.0,
|
105 |
+
"eval_accuracy": 0.9568181818181818,
|
106 |
+
"eval_loss": 0.12942154705524445,
|
107 |
+
"eval_runtime": 11.0819,
|
108 |
+
"eval_samples_per_second": 79.409,
|
109 |
+
"eval_steps_per_second": 4.963,
|
110 |
+
"step": 220
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 4.36,
|
114 |
+
"learning_rate": 2.6994256541161455e-05,
|
115 |
+
"loss": 0.2481,
|
116 |
+
"step": 240
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 4.73,
|
120 |
+
"learning_rate": 2.6611359285258455e-05,
|
121 |
+
"loss": 0.2414,
|
122 |
+
"step": 260
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 5.0,
|
126 |
+
"eval_accuracy": 0.990909090909091,
|
127 |
+
"eval_loss": 0.03464468568563461,
|
128 |
+
"eval_runtime": 11.2055,
|
129 |
+
"eval_samples_per_second": 78.533,
|
130 |
+
"eval_steps_per_second": 4.908,
|
131 |
+
"step": 275
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 5.09,
|
135 |
+
"learning_rate": 2.622846202935546e-05,
|
136 |
+
"loss": 0.2313,
|
137 |
+
"step": 280
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 5.45,
|
141 |
+
"learning_rate": 2.5845564773452456e-05,
|
142 |
+
"loss": 0.2074,
|
143 |
+
"step": 300
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 5.82,
|
147 |
+
"learning_rate": 2.546266751754946e-05,
|
148 |
+
"loss": 0.1888,
|
149 |
+
"step": 320
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 6.0,
|
153 |
+
"eval_accuracy": 0.9863636363636363,
|
154 |
+
"eval_loss": 0.04186202958226204,
|
155 |
+
"eval_runtime": 11.0353,
|
156 |
+
"eval_samples_per_second": 79.744,
|
157 |
+
"eval_steps_per_second": 4.984,
|
158 |
+
"step": 330
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 6.18,
|
162 |
+
"learning_rate": 2.507977026164646e-05,
|
163 |
+
"loss": 0.2057,
|
164 |
+
"step": 340
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 6.55,
|
168 |
+
"learning_rate": 2.469687300574346e-05,
|
169 |
+
"loss": 0.1693,
|
170 |
+
"step": 360
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 6.91,
|
174 |
+
"learning_rate": 2.431397574984046e-05,
|
175 |
+
"loss": 0.1717,
|
176 |
+
"step": 380
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 7.0,
|
180 |
+
"eval_accuracy": 0.9943181818181818,
|
181 |
+
"eval_loss": 0.023813609033823013,
|
182 |
+
"eval_runtime": 11.0829,
|
183 |
+
"eval_samples_per_second": 79.402,
|
184 |
+
"eval_steps_per_second": 4.963,
|
185 |
+
"step": 385
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 7.27,
|
189 |
+
"learning_rate": 2.393107849393746e-05,
|
190 |
+
"loss": 0.1614,
|
191 |
+
"step": 400
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 7.64,
|
195 |
+
"learning_rate": 2.3548181238034463e-05,
|
196 |
+
"loss": 0.1525,
|
197 |
+
"step": 420
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 8.0,
|
201 |
+
"learning_rate": 2.316528398213146e-05,
|
202 |
+
"loss": 0.1785,
|
203 |
+
"step": 440
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 8.0,
|
207 |
+
"eval_accuracy": 0.9943181818181818,
|
208 |
+
"eval_loss": 0.023027343675494194,
|
209 |
+
"eval_runtime": 11.2031,
|
210 |
+
"eval_samples_per_second": 78.55,
|
211 |
+
"eval_steps_per_second": 4.909,
|
212 |
+
"step": 440
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 8.36,
|
216 |
+
"learning_rate": 2.2782386726228464e-05,
|
217 |
+
"loss": 0.1256,
|
218 |
+
"step": 460
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 8.73,
|
222 |
+
"learning_rate": 2.2399489470325464e-05,
|
223 |
+
"loss": 0.1654,
|
224 |
+
"step": 480
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 9.0,
|
228 |
+
"eval_accuracy": 1.0,
|
229 |
+
"eval_loss": 0.007612039800733328,
|
230 |
+
"eval_runtime": 11.1318,
|
231 |
+
"eval_samples_per_second": 79.053,
|
232 |
+
"eval_steps_per_second": 4.941,
|
233 |
+
"step": 495
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 9.09,
|
237 |
+
"learning_rate": 2.2016592214422464e-05,
|
238 |
+
"loss": 0.1437,
|
239 |
+
"step": 500
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 9.45,
|
243 |
+
"learning_rate": 2.1633694958519464e-05,
|
244 |
+
"loss": 0.1363,
|
245 |
+
"step": 520
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 9.82,
|
249 |
+
"learning_rate": 2.1250797702616464e-05,
|
250 |
+
"loss": 0.1322,
|
251 |
+
"step": 540
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 10.0,
|
255 |
+
"eval_accuracy": 1.0,
|
256 |
+
"eval_loss": 0.004579578526318073,
|
257 |
+
"eval_runtime": 11.0656,
|
258 |
+
"eval_samples_per_second": 79.526,
|
259 |
+
"eval_steps_per_second": 4.97,
|
260 |
+
"step": 550
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 10.18,
|
264 |
+
"learning_rate": 2.0867900446713468e-05,
|
265 |
+
"loss": 0.1242,
|
266 |
+
"step": 560
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 10.55,
|
270 |
+
"learning_rate": 2.0485003190810464e-05,
|
271 |
+
"loss": 0.1121,
|
272 |
+
"step": 580
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 10.91,
|
276 |
+
"learning_rate": 2.0102105934907468e-05,
|
277 |
+
"loss": 0.1123,
|
278 |
+
"step": 600
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 11.0,
|
282 |
+
"eval_accuracy": 1.0,
|
283 |
+
"eval_loss": 0.003508554305881262,
|
284 |
+
"eval_runtime": 10.9833,
|
285 |
+
"eval_samples_per_second": 80.122,
|
286 |
+
"eval_steps_per_second": 5.008,
|
287 |
+
"step": 605
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 11.27,
|
291 |
+
"learning_rate": 1.9719208679004468e-05,
|
292 |
+
"loss": 0.1192,
|
293 |
+
"step": 620
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 11.64,
|
297 |
+
"learning_rate": 1.9336311423101468e-05,
|
298 |
+
"loss": 0.1179,
|
299 |
+
"step": 640
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 12.0,
|
303 |
+
"learning_rate": 1.8953414167198468e-05,
|
304 |
+
"loss": 0.0953,
|
305 |
+
"step": 660
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 12.0,
|
309 |
+
"eval_accuracy": 1.0,
|
310 |
+
"eval_loss": 0.002522848779335618,
|
311 |
+
"eval_runtime": 11.0556,
|
312 |
+
"eval_samples_per_second": 79.598,
|
313 |
+
"eval_steps_per_second": 4.975,
|
314 |
+
"step": 660
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 12.36,
|
318 |
+
"learning_rate": 1.8570516911295472e-05,
|
319 |
+
"loss": 0.1302,
|
320 |
+
"step": 680
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 12.73,
|
324 |
+
"learning_rate": 1.818761965539247e-05,
|
325 |
+
"loss": 0.0864,
|
326 |
+
"step": 700
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 13.0,
|
330 |
+
"eval_accuracy": 1.0,
|
331 |
+
"eval_loss": 0.0033287114929407835,
|
332 |
+
"eval_runtime": 11.1101,
|
333 |
+
"eval_samples_per_second": 79.207,
|
334 |
+
"eval_steps_per_second": 4.95,
|
335 |
+
"step": 715
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 13.09,
|
339 |
+
"learning_rate": 1.780472239948947e-05,
|
340 |
+
"loss": 0.1334,
|
341 |
+
"step": 720
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 13.45,
|
345 |
+
"learning_rate": 1.7421825143586472e-05,
|
346 |
+
"loss": 0.0769,
|
347 |
+
"step": 740
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 13.82,
|
351 |
+
"learning_rate": 1.7038927887683472e-05,
|
352 |
+
"loss": 0.0984,
|
353 |
+
"step": 760
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 14.0,
|
357 |
+
"eval_accuracy": 0.9988636363636364,
|
358 |
+
"eval_loss": 0.003271339228376746,
|
359 |
+
"eval_runtime": 11.0383,
|
360 |
+
"eval_samples_per_second": 79.723,
|
361 |
+
"eval_steps_per_second": 4.983,
|
362 |
+
"step": 770
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 14.18,
|
366 |
+
"learning_rate": 1.6656030631780472e-05,
|
367 |
+
"loss": 0.0932,
|
368 |
+
"step": 780
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 14.55,
|
372 |
+
"learning_rate": 1.6273133375877472e-05,
|
373 |
+
"loss": 0.079,
|
374 |
+
"step": 800
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 14.91,
|
378 |
+
"learning_rate": 1.5890236119974476e-05,
|
379 |
+
"loss": 0.0952,
|
380 |
+
"step": 820
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 15.0,
|
384 |
+
"eval_accuracy": 1.0,
|
385 |
+
"eval_loss": 0.001456564525142312,
|
386 |
+
"eval_runtime": 10.9322,
|
387 |
+
"eval_samples_per_second": 80.496,
|
388 |
+
"eval_steps_per_second": 5.031,
|
389 |
+
"step": 825
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 15.27,
|
393 |
+
"learning_rate": 1.5507338864071473e-05,
|
394 |
+
"loss": 0.087,
|
395 |
+
"step": 840
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 15.64,
|
399 |
+
"learning_rate": 1.5124441608168476e-05,
|
400 |
+
"loss": 0.0879,
|
401 |
+
"step": 860
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 16.0,
|
405 |
+
"learning_rate": 1.4741544352265476e-05,
|
406 |
+
"loss": 0.0678,
|
407 |
+
"step": 880
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 16.0,
|
411 |
+
"eval_accuracy": 1.0,
|
412 |
+
"eval_loss": 0.0022302898578345776,
|
413 |
+
"eval_runtime": 11.0758,
|
414 |
+
"eval_samples_per_second": 79.453,
|
415 |
+
"eval_steps_per_second": 4.966,
|
416 |
+
"step": 880
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 16.36,
|
420 |
+
"learning_rate": 1.4358647096362477e-05,
|
421 |
+
"loss": 0.0687,
|
422 |
+
"step": 900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 16.73,
|
426 |
+
"learning_rate": 1.3975749840459477e-05,
|
427 |
+
"loss": 0.0592,
|
428 |
+
"step": 920
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 17.0,
|
432 |
+
"eval_accuracy": 1.0,
|
433 |
+
"eval_loss": 0.0012722605606541038,
|
434 |
+
"eval_runtime": 10.904,
|
435 |
+
"eval_samples_per_second": 80.704,
|
436 |
+
"eval_steps_per_second": 5.044,
|
437 |
+
"step": 935
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 17.09,
|
441 |
+
"learning_rate": 1.3592852584556478e-05,
|
442 |
+
"loss": 0.0701,
|
443 |
+
"step": 940
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 17.45,
|
447 |
+
"learning_rate": 1.3209955328653479e-05,
|
448 |
+
"loss": 0.0952,
|
449 |
+
"step": 960
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 17.82,
|
453 |
+
"learning_rate": 1.2827058072750479e-05,
|
454 |
+
"loss": 0.0729,
|
455 |
+
"step": 980
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 18.0,
|
459 |
+
"eval_accuracy": 0.9988636363636364,
|
460 |
+
"eval_loss": 0.00370142818428576,
|
461 |
+
"eval_runtime": 11.1436,
|
462 |
+
"eval_samples_per_second": 78.969,
|
463 |
+
"eval_steps_per_second": 4.936,
|
464 |
+
"step": 990
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 18.18,
|
468 |
+
"learning_rate": 1.2444160816847479e-05,
|
469 |
+
"loss": 0.0841,
|
470 |
+
"step": 1000
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 18.55,
|
474 |
+
"learning_rate": 1.206126356094448e-05,
|
475 |
+
"loss": 0.0837,
|
476 |
+
"step": 1020
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 18.91,
|
480 |
+
"learning_rate": 1.167836630504148e-05,
|
481 |
+
"loss": 0.0672,
|
482 |
+
"step": 1040
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 19.0,
|
486 |
+
"eval_accuracy": 0.9988636363636364,
|
487 |
+
"eval_loss": 0.004141129087656736,
|
488 |
+
"eval_runtime": 11.0993,
|
489 |
+
"eval_samples_per_second": 79.284,
|
490 |
+
"eval_steps_per_second": 4.955,
|
491 |
+
"step": 1045
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 19.27,
|
495 |
+
"learning_rate": 1.1295469049138481e-05,
|
496 |
+
"loss": 0.0572,
|
497 |
+
"step": 1060
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 19.64,
|
501 |
+
"learning_rate": 1.0912571793235483e-05,
|
502 |
+
"loss": 0.0577,
|
503 |
+
"step": 1080
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 20.0,
|
507 |
+
"learning_rate": 1.0529674537332483e-05,
|
508 |
+
"loss": 0.0615,
|
509 |
+
"step": 1100
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 20.0,
|
513 |
+
"eval_accuracy": 1.0,
|
514 |
+
"eval_loss": 0.001032730215229094,
|
515 |
+
"eval_runtime": 11.0623,
|
516 |
+
"eval_samples_per_second": 79.55,
|
517 |
+
"eval_steps_per_second": 4.972,
|
518 |
+
"step": 1100
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 20.36,
|
522 |
+
"learning_rate": 1.0146777281429485e-05,
|
523 |
+
"loss": 0.0546,
|
524 |
+
"step": 1120
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 20.73,
|
528 |
+
"learning_rate": 9.763880025526483e-06,
|
529 |
+
"loss": 0.058,
|
530 |
+
"step": 1140
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 21.0,
|
534 |
+
"eval_accuracy": 1.0,
|
535 |
+
"eval_loss": 0.0008914543432183564,
|
536 |
+
"eval_runtime": 10.9202,
|
537 |
+
"eval_samples_per_second": 80.585,
|
538 |
+
"eval_steps_per_second": 5.037,
|
539 |
+
"step": 1155
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 21.09,
|
543 |
+
"learning_rate": 9.380982769623483e-06,
|
544 |
+
"loss": 0.0497,
|
545 |
+
"step": 1160
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 21.45,
|
549 |
+
"learning_rate": 8.998085513720485e-06,
|
550 |
+
"loss": 0.0538,
|
551 |
+
"step": 1180
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 21.82,
|
555 |
+
"learning_rate": 8.615188257817485e-06,
|
556 |
+
"loss": 0.0571,
|
557 |
+
"step": 1200
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 22.0,
|
561 |
+
"eval_accuracy": 0.9988636363636364,
|
562 |
+
"eval_loss": 0.002123360289260745,
|
563 |
+
"eval_runtime": 10.9798,
|
564 |
+
"eval_samples_per_second": 80.147,
|
565 |
+
"eval_steps_per_second": 5.009,
|
566 |
+
"step": 1210
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 22.18,
|
570 |
+
"learning_rate": 8.232291001914487e-06,
|
571 |
+
"loss": 0.0636,
|
572 |
+
"step": 1220
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 22.55,
|
576 |
+
"learning_rate": 7.849393746011487e-06,
|
577 |
+
"loss": 0.0365,
|
578 |
+
"step": 1240
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 22.91,
|
582 |
+
"learning_rate": 7.466496490108488e-06,
|
583 |
+
"loss": 0.0755,
|
584 |
+
"step": 1260
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 23.0,
|
588 |
+
"eval_accuracy": 0.9988636363636364,
|
589 |
+
"eval_loss": 0.002230195328593254,
|
590 |
+
"eval_runtime": 11.0651,
|
591 |
+
"eval_samples_per_second": 79.529,
|
592 |
+
"eval_steps_per_second": 4.971,
|
593 |
+
"step": 1265
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 23.27,
|
597 |
+
"learning_rate": 7.083599234205488e-06,
|
598 |
+
"loss": 0.0763,
|
599 |
+
"step": 1280
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 23.64,
|
603 |
+
"learning_rate": 6.700701978302489e-06,
|
604 |
+
"loss": 0.0547,
|
605 |
+
"step": 1300
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 24.0,
|
609 |
+
"learning_rate": 6.317804722399489e-06,
|
610 |
+
"loss": 0.0688,
|
611 |
+
"step": 1320
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 24.0,
|
615 |
+
"eval_accuracy": 0.9988636363636364,
|
616 |
+
"eval_loss": 0.0025376665871590376,
|
617 |
+
"eval_runtime": 10.9568,
|
618 |
+
"eval_samples_per_second": 80.315,
|
619 |
+
"eval_steps_per_second": 5.02,
|
620 |
+
"step": 1320
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 24.36,
|
624 |
+
"learning_rate": 5.93490746649649e-06,
|
625 |
+
"loss": 0.0726,
|
626 |
+
"step": 1340
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 24.73,
|
630 |
+
"learning_rate": 5.5520102105934905e-06,
|
631 |
+
"loss": 0.0417,
|
632 |
+
"step": 1360
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 25.0,
|
636 |
+
"eval_accuracy": 1.0,
|
637 |
+
"eval_loss": 0.00029166368767619133,
|
638 |
+
"eval_runtime": 11.0108,
|
639 |
+
"eval_samples_per_second": 79.922,
|
640 |
+
"eval_steps_per_second": 4.995,
|
641 |
+
"step": 1375
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 25.09,
|
645 |
+
"learning_rate": 5.1691129546904915e-06,
|
646 |
+
"loss": 0.0472,
|
647 |
+
"step": 1380
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 25.45,
|
651 |
+
"learning_rate": 4.7862156987874925e-06,
|
652 |
+
"loss": 0.0565,
|
653 |
+
"step": 1400
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 25.82,
|
657 |
+
"learning_rate": 4.4033184428844934e-06,
|
658 |
+
"loss": 0.0589,
|
659 |
+
"step": 1420
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 26.0,
|
663 |
+
"eval_accuracy": 1.0,
|
664 |
+
"eval_loss": 0.0006550709367729723,
|
665 |
+
"eval_runtime": 10.9639,
|
666 |
+
"eval_samples_per_second": 80.264,
|
667 |
+
"eval_steps_per_second": 5.016,
|
668 |
+
"step": 1430
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 26.18,
|
672 |
+
"learning_rate": 4.020421186981493e-06,
|
673 |
+
"loss": 0.0452,
|
674 |
+
"step": 1440
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 26.55,
|
678 |
+
"learning_rate": 3.6375239310784937e-06,
|
679 |
+
"loss": 0.0587,
|
680 |
+
"step": 1460
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 26.91,
|
684 |
+
"learning_rate": 3.2546266751754947e-06,
|
685 |
+
"loss": 0.0563,
|
686 |
+
"step": 1480
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 27.0,
|
690 |
+
"eval_accuracy": 1.0,
|
691 |
+
"eval_loss": 0.0006738207302987576,
|
692 |
+
"eval_runtime": 10.975,
|
693 |
+
"eval_samples_per_second": 80.182,
|
694 |
+
"eval_steps_per_second": 5.011,
|
695 |
+
"step": 1485
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 27.27,
|
699 |
+
"learning_rate": 2.8717294192724956e-06,
|
700 |
+
"loss": 0.0441,
|
701 |
+
"step": 1500
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 27.64,
|
705 |
+
"learning_rate": 2.4888321633694958e-06,
|
706 |
+
"loss": 0.0401,
|
707 |
+
"step": 1520
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 28.0,
|
711 |
+
"learning_rate": 2.1059349074664967e-06,
|
712 |
+
"loss": 0.0603,
|
713 |
+
"step": 1540
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 28.0,
|
717 |
+
"eval_accuracy": 0.9988636363636364,
|
718 |
+
"eval_loss": 0.0009842341532930732,
|
719 |
+
"eval_runtime": 10.951,
|
720 |
+
"eval_samples_per_second": 80.358,
|
721 |
+
"eval_steps_per_second": 5.022,
|
722 |
+
"step": 1540
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 28.36,
|
726 |
+
"learning_rate": 1.723037651563497e-06,
|
727 |
+
"loss": 0.0417,
|
728 |
+
"step": 1560
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 28.73,
|
732 |
+
"learning_rate": 1.3401403956604976e-06,
|
733 |
+
"loss": 0.0469,
|
734 |
+
"step": 1580
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 29.0,
|
738 |
+
"eval_accuracy": 1.0,
|
739 |
+
"eval_loss": 0.0004589582094922662,
|
740 |
+
"eval_runtime": 10.9725,
|
741 |
+
"eval_samples_per_second": 80.2,
|
742 |
+
"eval_steps_per_second": 5.013,
|
743 |
+
"step": 1595
|
744 |
+
},
|
745 |
+
{
|
746 |
+
"epoch": 29.09,
|
747 |
+
"learning_rate": 9.572431397574984e-07,
|
748 |
+
"loss": 0.0518,
|
749 |
+
"step": 1600
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 29.45,
|
753 |
+
"learning_rate": 5.743458838544991e-07,
|
754 |
+
"loss": 0.0446,
|
755 |
+
"step": 1620
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 29.82,
|
759 |
+
"learning_rate": 1.914486279514997e-07,
|
760 |
+
"loss": 0.0525,
|
761 |
+
"step": 1640
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 30.0,
|
765 |
+
"eval_accuracy": 1.0,
|
766 |
+
"eval_loss": 0.00037947672535665333,
|
767 |
+
"eval_runtime": 10.9402,
|
768 |
+
"eval_samples_per_second": 80.438,
|
769 |
+
"eval_steps_per_second": 5.027,
|
770 |
+
"step": 1650
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 30.0,
|
774 |
+
"step": 1650,
|
775 |
+
"total_flos": 2.4304114274567455e+19,
|
776 |
+
"train_loss": 0.19108782262513133,
|
777 |
+
"train_runtime": 4786.2205,
|
778 |
+
"train_samples_per_second": 22.051,
|
779 |
+
"train_steps_per_second": 0.345
|
780 |
+
}
|
781 |
+
],
|
782 |
+
"max_steps": 1650,
|
783 |
+
"num_train_epochs": 30,
|
784 |
+
"total_flos": 2.4304114274567455e+19,
|
785 |
+
"trial_name": null,
|
786 |
+
"trial_params": null
|
787 |
+
}
|