Augusto777
commited on
Commit
•
3e4362d
1
Parent(s):
7c84eba
End of training
Browse files- README.md +2 -2
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Apr18_22-30-59_DESKTOP-SKBE9FB/events.out.tfevents.1713501643.DESKTOP-SKBE9FB.5296.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +516 -0
README.md
CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss:
|
21 |
-
- Accuracy: 0.
|
22 |
|
23 |
## Model description
|
24 |
|
|
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 1.0633
|
21 |
+
- Accuracy: 0.6333
|
22 |
|
23 |
## Model description
|
24 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 37.94,
|
3 |
+
"eval_accuracy": 0.6333333333333333,
|
4 |
+
"eval_loss": 1.0633025169372559,
|
5 |
+
"eval_runtime": 1.2452,
|
6 |
+
"eval_samples_per_second": 48.183,
|
7 |
+
"eval_steps_per_second": 1.606,
|
8 |
+
"train_loss": 1.8558632776039798,
|
9 |
+
"train_runtime": 524.2652,
|
10 |
+
"train_samples_per_second": 78.029,
|
11 |
+
"train_steps_per_second": 0.561
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 37.94,
|
3 |
+
"eval_accuracy": 0.6333333333333333,
|
4 |
+
"eval_loss": 1.0633025169372559,
|
5 |
+
"eval_runtime": 1.2452,
|
6 |
+
"eval_samples_per_second": 48.183,
|
7 |
+
"eval_steps_per_second": 1.606
|
8 |
+
}
|
runs/Apr18_22-30-59_DESKTOP-SKBE9FB/events.out.tfevents.1713501643.DESKTOP-SKBE9FB.5296.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e5c090786cbbac1b9b3597303fa98f7e854d423f6ed0177277fa26d55615aa96
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 37.94,
|
3 |
+
"train_loss": 1.8558632776039798,
|
4 |
+
"train_runtime": 524.2652,
|
5 |
+
"train_samples_per_second": 78.029,
|
6 |
+
"train_steps_per_second": 0.561
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,516 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.6333333333333333,
|
3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U5-42B\\checkpoint-193",
|
4 |
+
"epoch": 37.935483870967744,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 294,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.9,
|
13 |
+
"eval_accuracy": 0.11666666666666667,
|
14 |
+
"eval_loss": 7.866283893585205,
|
15 |
+
"eval_runtime": 1.0387,
|
16 |
+
"eval_samples_per_second": 57.764,
|
17 |
+
"eval_steps_per_second": 1.925,
|
18 |
+
"step": 7
|
19 |
+
},
|
20 |
+
{
|
21 |
+
"epoch": 1.55,
|
22 |
+
"learning_rate": 1.6000000000000003e-05,
|
23 |
+
"loss": 6.936,
|
24 |
+
"step": 12
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 1.94,
|
28 |
+
"eval_accuracy": 0.11666666666666667,
|
29 |
+
"eval_loss": 7.757172584533691,
|
30 |
+
"eval_runtime": 0.9676,
|
31 |
+
"eval_samples_per_second": 62.01,
|
32 |
+
"eval_steps_per_second": 2.067,
|
33 |
+
"step": 15
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 2.97,
|
37 |
+
"eval_accuracy": 0.11666666666666667,
|
38 |
+
"eval_loss": 7.17894983291626,
|
39 |
+
"eval_runtime": 1.0595,
|
40 |
+
"eval_samples_per_second": 56.633,
|
41 |
+
"eval_steps_per_second": 1.888,
|
42 |
+
"step": 23
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 3.1,
|
46 |
+
"learning_rate": 3.2000000000000005e-05,
|
47 |
+
"loss": 6.7016,
|
48 |
+
"step": 24
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 4.0,
|
52 |
+
"eval_accuracy": 0.11666666666666667,
|
53 |
+
"eval_loss": 5.914210796356201,
|
54 |
+
"eval_runtime": 1.1717,
|
55 |
+
"eval_samples_per_second": 51.21,
|
56 |
+
"eval_steps_per_second": 1.707,
|
57 |
+
"step": 31
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 4.65,
|
61 |
+
"learning_rate": 3.909090909090909e-05,
|
62 |
+
"loss": 5.5418,
|
63 |
+
"step": 36
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 4.9,
|
67 |
+
"eval_accuracy": 0.11666666666666667,
|
68 |
+
"eval_loss": 4.606503486633301,
|
69 |
+
"eval_runtime": 0.9707,
|
70 |
+
"eval_samples_per_second": 61.813,
|
71 |
+
"eval_steps_per_second": 2.06,
|
72 |
+
"step": 38
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 5.94,
|
76 |
+
"eval_accuracy": 0.11666666666666667,
|
77 |
+
"eval_loss": 3.2782602310180664,
|
78 |
+
"eval_runtime": 1.0414,
|
79 |
+
"eval_samples_per_second": 57.616,
|
80 |
+
"eval_steps_per_second": 1.921,
|
81 |
+
"step": 46
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 6.19,
|
85 |
+
"learning_rate": 3.7272727272727276e-05,
|
86 |
+
"loss": 3.6439,
|
87 |
+
"step": 48
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 6.97,
|
91 |
+
"eval_accuracy": 0.11666666666666667,
|
92 |
+
"eval_loss": 2.1983864307403564,
|
93 |
+
"eval_runtime": 1.0791,
|
94 |
+
"eval_samples_per_second": 55.604,
|
95 |
+
"eval_steps_per_second": 1.853,
|
96 |
+
"step": 54
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 7.74,
|
100 |
+
"learning_rate": 3.545454545454546e-05,
|
101 |
+
"loss": 2.2477,
|
102 |
+
"step": 60
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 8.0,
|
106 |
+
"eval_accuracy": 0.45,
|
107 |
+
"eval_loss": 1.5630935430526733,
|
108 |
+
"eval_runtime": 1.2494,
|
109 |
+
"eval_samples_per_second": 48.021,
|
110 |
+
"eval_steps_per_second": 1.601,
|
111 |
+
"step": 62
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 8.9,
|
115 |
+
"eval_accuracy": 0.45,
|
116 |
+
"eval_loss": 1.315753698348999,
|
117 |
+
"eval_runtime": 1.0403,
|
118 |
+
"eval_samples_per_second": 57.675,
|
119 |
+
"eval_steps_per_second": 1.923,
|
120 |
+
"step": 69
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 9.29,
|
124 |
+
"learning_rate": 3.363636363636364e-05,
|
125 |
+
"loss": 1.5076,
|
126 |
+
"step": 72
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 9.94,
|
130 |
+
"eval_accuracy": 0.3333333333333333,
|
131 |
+
"eval_loss": 1.3664613962173462,
|
132 |
+
"eval_runtime": 1.076,
|
133 |
+
"eval_samples_per_second": 55.76,
|
134 |
+
"eval_steps_per_second": 1.859,
|
135 |
+
"step": 77
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 10.84,
|
139 |
+
"learning_rate": 3.181818181818182e-05,
|
140 |
+
"loss": 1.3865,
|
141 |
+
"step": 84
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 10.97,
|
145 |
+
"eval_accuracy": 0.35,
|
146 |
+
"eval_loss": 1.3560707569122314,
|
147 |
+
"eval_runtime": 1.2112,
|
148 |
+
"eval_samples_per_second": 49.538,
|
149 |
+
"eval_steps_per_second": 1.651,
|
150 |
+
"step": 85
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 12.0,
|
154 |
+
"eval_accuracy": 0.4666666666666667,
|
155 |
+
"eval_loss": 1.2673075199127197,
|
156 |
+
"eval_runtime": 0.9395,
|
157 |
+
"eval_samples_per_second": 63.867,
|
158 |
+
"eval_steps_per_second": 2.129,
|
159 |
+
"step": 93
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 12.39,
|
163 |
+
"learning_rate": 3.0000000000000004e-05,
|
164 |
+
"loss": 1.3436,
|
165 |
+
"step": 96
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 12.9,
|
169 |
+
"eval_accuracy": 0.2833333333333333,
|
170 |
+
"eval_loss": 1.5020203590393066,
|
171 |
+
"eval_runtime": 1.0828,
|
172 |
+
"eval_samples_per_second": 55.411,
|
173 |
+
"eval_steps_per_second": 1.847,
|
174 |
+
"step": 100
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 13.94,
|
178 |
+
"learning_rate": 2.8181818181818185e-05,
|
179 |
+
"loss": 1.3187,
|
180 |
+
"step": 108
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 13.94,
|
184 |
+
"eval_accuracy": 0.4666666666666667,
|
185 |
+
"eval_loss": 1.3018145561218262,
|
186 |
+
"eval_runtime": 0.9637,
|
187 |
+
"eval_samples_per_second": 62.261,
|
188 |
+
"eval_steps_per_second": 2.075,
|
189 |
+
"step": 108
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 14.97,
|
193 |
+
"eval_accuracy": 0.48333333333333334,
|
194 |
+
"eval_loss": 1.2581952810287476,
|
195 |
+
"eval_runtime": 0.968,
|
196 |
+
"eval_samples_per_second": 61.985,
|
197 |
+
"eval_steps_per_second": 2.066,
|
198 |
+
"step": 116
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 15.48,
|
202 |
+
"learning_rate": 2.6363636363636365e-05,
|
203 |
+
"loss": 1.2132,
|
204 |
+
"step": 120
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 16.0,
|
208 |
+
"eval_accuracy": 0.5,
|
209 |
+
"eval_loss": 1.2250442504882812,
|
210 |
+
"eval_runtime": 1.0265,
|
211 |
+
"eval_samples_per_second": 58.453,
|
212 |
+
"eval_steps_per_second": 1.948,
|
213 |
+
"step": 124
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 16.9,
|
217 |
+
"eval_accuracy": 0.5166666666666667,
|
218 |
+
"eval_loss": 1.2070510387420654,
|
219 |
+
"eval_runtime": 0.8885,
|
220 |
+
"eval_samples_per_second": 67.53,
|
221 |
+
"eval_steps_per_second": 2.251,
|
222 |
+
"step": 131
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 17.03,
|
226 |
+
"learning_rate": 2.454545454545455e-05,
|
227 |
+
"loss": 1.2041,
|
228 |
+
"step": 132
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 17.94,
|
232 |
+
"eval_accuracy": 0.5166666666666667,
|
233 |
+
"eval_loss": 1.1806073188781738,
|
234 |
+
"eval_runtime": 1.0077,
|
235 |
+
"eval_samples_per_second": 59.543,
|
236 |
+
"eval_steps_per_second": 1.985,
|
237 |
+
"step": 139
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 18.58,
|
241 |
+
"learning_rate": 2.2727272727272733e-05,
|
242 |
+
"loss": 1.1756,
|
243 |
+
"step": 144
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 18.97,
|
247 |
+
"eval_accuracy": 0.55,
|
248 |
+
"eval_loss": 1.1720372438430786,
|
249 |
+
"eval_runtime": 1.0797,
|
250 |
+
"eval_samples_per_second": 55.573,
|
251 |
+
"eval_steps_per_second": 1.852,
|
252 |
+
"step": 147
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 20.0,
|
256 |
+
"eval_accuracy": 0.5333333333333333,
|
257 |
+
"eval_loss": 1.1318727731704712,
|
258 |
+
"eval_runtime": 0.8755,
|
259 |
+
"eval_samples_per_second": 68.529,
|
260 |
+
"eval_steps_per_second": 2.284,
|
261 |
+
"step": 155
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 20.13,
|
265 |
+
"learning_rate": 2.090909090909091e-05,
|
266 |
+
"loss": 1.1107,
|
267 |
+
"step": 156
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 20.9,
|
271 |
+
"eval_accuracy": 0.5833333333333334,
|
272 |
+
"eval_loss": 1.0850528478622437,
|
273 |
+
"eval_runtime": 0.9417,
|
274 |
+
"eval_samples_per_second": 63.716,
|
275 |
+
"eval_steps_per_second": 2.124,
|
276 |
+
"step": 162
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 21.68,
|
280 |
+
"learning_rate": 1.9090909090909094e-05,
|
281 |
+
"loss": 1.0651,
|
282 |
+
"step": 168
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 21.94,
|
286 |
+
"eval_accuracy": 0.5833333333333334,
|
287 |
+
"eval_loss": 1.0553667545318604,
|
288 |
+
"eval_runtime": 0.9498,
|
289 |
+
"eval_samples_per_second": 63.168,
|
290 |
+
"eval_steps_per_second": 2.106,
|
291 |
+
"step": 170
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 22.97,
|
295 |
+
"eval_accuracy": 0.5,
|
296 |
+
"eval_loss": 1.0579936504364014,
|
297 |
+
"eval_runtime": 0.9211,
|
298 |
+
"eval_samples_per_second": 65.141,
|
299 |
+
"eval_steps_per_second": 2.171,
|
300 |
+
"step": 178
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 23.23,
|
304 |
+
"learning_rate": 1.7272727272727274e-05,
|
305 |
+
"loss": 1.0419,
|
306 |
+
"step": 180
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 24.0,
|
310 |
+
"eval_accuracy": 0.5833333333333334,
|
311 |
+
"eval_loss": 1.1471267938613892,
|
312 |
+
"eval_runtime": 0.9539,
|
313 |
+
"eval_samples_per_second": 62.901,
|
314 |
+
"eval_steps_per_second": 2.097,
|
315 |
+
"step": 186
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 24.77,
|
319 |
+
"learning_rate": 1.5454545454545454e-05,
|
320 |
+
"loss": 0.9804,
|
321 |
+
"step": 192
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 24.9,
|
325 |
+
"eval_accuracy": 0.6333333333333333,
|
326 |
+
"eval_loss": 1.0633025169372559,
|
327 |
+
"eval_runtime": 0.9613,
|
328 |
+
"eval_samples_per_second": 62.413,
|
329 |
+
"eval_steps_per_second": 2.08,
|
330 |
+
"step": 193
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 25.94,
|
334 |
+
"eval_accuracy": 0.5833333333333334,
|
335 |
+
"eval_loss": 1.0597257614135742,
|
336 |
+
"eval_runtime": 1.9689,
|
337 |
+
"eval_samples_per_second": 30.474,
|
338 |
+
"eval_steps_per_second": 1.016,
|
339 |
+
"step": 201
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 26.32,
|
343 |
+
"learning_rate": 1.3636363636363637e-05,
|
344 |
+
"loss": 0.9195,
|
345 |
+
"step": 204
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 26.97,
|
349 |
+
"eval_accuracy": 0.6333333333333333,
|
350 |
+
"eval_loss": 0.9563372135162354,
|
351 |
+
"eval_runtime": 0.9494,
|
352 |
+
"eval_samples_per_second": 63.199,
|
353 |
+
"eval_steps_per_second": 2.107,
|
354 |
+
"step": 209
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 27.87,
|
358 |
+
"learning_rate": 1.181818181818182e-05,
|
359 |
+
"loss": 0.9053,
|
360 |
+
"step": 216
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 28.0,
|
364 |
+
"eval_accuracy": 0.5166666666666667,
|
365 |
+
"eval_loss": 0.9838621020317078,
|
366 |
+
"eval_runtime": 0.8939,
|
367 |
+
"eval_samples_per_second": 67.124,
|
368 |
+
"eval_steps_per_second": 2.237,
|
369 |
+
"step": 217
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 28.9,
|
373 |
+
"eval_accuracy": 0.5166666666666667,
|
374 |
+
"eval_loss": 0.9913681745529175,
|
375 |
+
"eval_runtime": 0.8926,
|
376 |
+
"eval_samples_per_second": 67.223,
|
377 |
+
"eval_steps_per_second": 2.241,
|
378 |
+
"step": 224
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 29.42,
|
382 |
+
"learning_rate": 1e-05,
|
383 |
+
"loss": 0.8645,
|
384 |
+
"step": 228
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 29.94,
|
388 |
+
"eval_accuracy": 0.5833333333333334,
|
389 |
+
"eval_loss": 0.9519664645195007,
|
390 |
+
"eval_runtime": 0.9221,
|
391 |
+
"eval_samples_per_second": 65.066,
|
392 |
+
"eval_steps_per_second": 2.169,
|
393 |
+
"step": 232
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 30.97,
|
397 |
+
"learning_rate": 8.181818181818183e-06,
|
398 |
+
"loss": 0.8139,
|
399 |
+
"step": 240
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 30.97,
|
403 |
+
"eval_accuracy": 0.5333333333333333,
|
404 |
+
"eval_loss": 0.9572514295578003,
|
405 |
+
"eval_runtime": 0.8817,
|
406 |
+
"eval_samples_per_second": 68.051,
|
407 |
+
"eval_steps_per_second": 2.268,
|
408 |
+
"step": 240
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 32.0,
|
412 |
+
"eval_accuracy": 0.5166666666666667,
|
413 |
+
"eval_loss": 0.9509603977203369,
|
414 |
+
"eval_runtime": 0.9232,
|
415 |
+
"eval_samples_per_second": 64.991,
|
416 |
+
"eval_steps_per_second": 2.166,
|
417 |
+
"step": 248
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 32.52,
|
421 |
+
"learning_rate": 6.363636363636364e-06,
|
422 |
+
"loss": 0.8151,
|
423 |
+
"step": 252
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 32.9,
|
427 |
+
"eval_accuracy": 0.5166666666666667,
|
428 |
+
"eval_loss": 0.9468602538108826,
|
429 |
+
"eval_runtime": 0.8657,
|
430 |
+
"eval_samples_per_second": 69.308,
|
431 |
+
"eval_steps_per_second": 2.31,
|
432 |
+
"step": 255
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 33.94,
|
436 |
+
"eval_accuracy": 0.5333333333333333,
|
437 |
+
"eval_loss": 0.9872100353240967,
|
438 |
+
"eval_runtime": 0.9376,
|
439 |
+
"eval_samples_per_second": 63.99,
|
440 |
+
"eval_steps_per_second": 2.133,
|
441 |
+
"step": 263
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 34.06,
|
445 |
+
"learning_rate": 4.5454545454545455e-06,
|
446 |
+
"loss": 0.7837,
|
447 |
+
"step": 264
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 34.97,
|
451 |
+
"eval_accuracy": 0.5833333333333334,
|
452 |
+
"eval_loss": 0.9278779625892639,
|
453 |
+
"eval_runtime": 1.3601,
|
454 |
+
"eval_samples_per_second": 44.113,
|
455 |
+
"eval_steps_per_second": 1.47,
|
456 |
+
"step": 271
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 35.61,
|
460 |
+
"learning_rate": 2.7272727272727272e-06,
|
461 |
+
"loss": 0.7659,
|
462 |
+
"step": 276
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 36.0,
|
466 |
+
"eval_accuracy": 0.5666666666666667,
|
467 |
+
"eval_loss": 0.9173651337623596,
|
468 |
+
"eval_runtime": 0.9252,
|
469 |
+
"eval_samples_per_second": 64.854,
|
470 |
+
"eval_steps_per_second": 2.162,
|
471 |
+
"step": 279
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 36.9,
|
475 |
+
"eval_accuracy": 0.5666666666666667,
|
476 |
+
"eval_loss": 0.934675931930542,
|
477 |
+
"eval_runtime": 1.2264,
|
478 |
+
"eval_samples_per_second": 48.923,
|
479 |
+
"eval_steps_per_second": 1.631,
|
480 |
+
"step": 286
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 37.16,
|
484 |
+
"learning_rate": 9.090909090909091e-07,
|
485 |
+
"loss": 0.7835,
|
486 |
+
"step": 288
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 37.94,
|
490 |
+
"eval_accuracy": 0.5666666666666667,
|
491 |
+
"eval_loss": 0.9372119903564453,
|
492 |
+
"eval_runtime": 0.9199,
|
493 |
+
"eval_samples_per_second": 65.224,
|
494 |
+
"eval_steps_per_second": 2.174,
|
495 |
+
"step": 294
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 37.94,
|
499 |
+
"step": 294,
|
500 |
+
"total_flos": 1.2027310550050406e+18,
|
501 |
+
"train_loss": 1.8558632776039798,
|
502 |
+
"train_runtime": 524.2652,
|
503 |
+
"train_samples_per_second": 78.029,
|
504 |
+
"train_steps_per_second": 0.561
|
505 |
+
}
|
506 |
+
],
|
507 |
+
"logging_steps": 12,
|
508 |
+
"max_steps": 294,
|
509 |
+
"num_input_tokens_seen": 0,
|
510 |
+
"num_train_epochs": 42,
|
511 |
+
"save_steps": 500,
|
512 |
+
"total_flos": 1.2027310550050406e+18,
|
513 |
+
"train_batch_size": 32,
|
514 |
+
"trial_name": null,
|
515 |
+
"trial_params": null
|
516 |
+
}
|