End of training
Browse files- README.md +3 -3
- all_results.json +18 -0
- eval_results.json +8 -0
- test_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +448 -0
README.md
CHANGED
@@ -22,7 +22,7 @@ model-index:
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
-
value: 0.
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [facebook/deit-small-patch16-224](https://huggingface.co/facebook/deit-small-patch16-224) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss: 0.
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.7808219178082192
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [facebook/deit-small-patch16-224](https://huggingface.co/facebook/deit-small-patch16-224) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.5296
|
36 |
+
- Accuracy: 0.7808
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.51,
|
3 |
+
"eval_accuracy": 0.7808219178082192,
|
4 |
+
"eval_loss": 0.5295854806900024,
|
5 |
+
"eval_runtime": 1.2479,
|
6 |
+
"eval_samples_per_second": 175.494,
|
7 |
+
"eval_steps_per_second": 22.438,
|
8 |
+
"test_accuracy": 0.771689497716895,
|
9 |
+
"test_loss": 0.6326195001602173,
|
10 |
+
"test_runtime": 1.2468,
|
11 |
+
"test_samples_per_second": 175.648,
|
12 |
+
"test_steps_per_second": 22.457,
|
13 |
+
"total_flos": 2.493693637268521e+17,
|
14 |
+
"train_loss": 0.3432186938822269,
|
15 |
+
"train_runtime": 208.3396,
|
16 |
+
"train_samples_per_second": 62.686,
|
17 |
+
"train_steps_per_second": 1.92
|
18 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.51,
|
3 |
+
"eval_accuracy": 0.7808219178082192,
|
4 |
+
"eval_loss": 0.5295854806900024,
|
5 |
+
"eval_runtime": 1.2479,
|
6 |
+
"eval_samples_per_second": 175.494,
|
7 |
+
"eval_steps_per_second": 22.438
|
8 |
+
}
|
test_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"test_accuracy": 0.771689497716895,
|
3 |
+
"test_loss": 0.6326195001602173,
|
4 |
+
"test_runtime": 1.2468,
|
5 |
+
"test_samples_per_second": 175.648,
|
6 |
+
"test_steps_per_second": 22.457
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.51,
|
3 |
+
"total_flos": 2.493693637268521e+17,
|
4 |
+
"train_loss": 0.3432186938822269,
|
5 |
+
"train_runtime": 208.3396,
|
6 |
+
"train_samples_per_second": 62.686,
|
7 |
+
"train_steps_per_second": 1.92
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,448 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7808219178082192,
|
3 |
+
"best_model_checkpoint": "deit-small-patch16-224-finetuned-piid/checkpoint-307",
|
4 |
+
"epoch": 19.51219512195122,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 400,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.49,
|
13 |
+
"learning_rate": 1.25e-05,
|
14 |
+
"loss": 1.4297,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.98,
|
19 |
+
"learning_rate": 2.5e-05,
|
20 |
+
"loss": 1.1537,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.98,
|
25 |
+
"eval_accuracy": 0.547945205479452,
|
26 |
+
"eval_loss": 1.0005042552947998,
|
27 |
+
"eval_runtime": 1.0515,
|
28 |
+
"eval_samples_per_second": 208.282,
|
29 |
+
"eval_steps_per_second": 26.63,
|
30 |
+
"step": 20
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 1.46,
|
34 |
+
"learning_rate": 3.7500000000000003e-05,
|
35 |
+
"loss": 0.8255,
|
36 |
+
"step": 30
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 1.95,
|
40 |
+
"learning_rate": 5e-05,
|
41 |
+
"loss": 0.7025,
|
42 |
+
"step": 40
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 2.0,
|
46 |
+
"eval_accuracy": 0.593607305936073,
|
47 |
+
"eval_loss": 0.8481215834617615,
|
48 |
+
"eval_runtime": 1.2994,
|
49 |
+
"eval_samples_per_second": 168.538,
|
50 |
+
"eval_steps_per_second": 21.548,
|
51 |
+
"step": 41
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 2.44,
|
55 |
+
"learning_rate": 4.8611111111111115e-05,
|
56 |
+
"loss": 0.6057,
|
57 |
+
"step": 50
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 2.93,
|
61 |
+
"learning_rate": 4.722222222222222e-05,
|
62 |
+
"loss": 0.6581,
|
63 |
+
"step": 60
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 2.98,
|
67 |
+
"eval_accuracy": 0.7214611872146118,
|
68 |
+
"eval_loss": 0.6351255774497986,
|
69 |
+
"eval_runtime": 1.0449,
|
70 |
+
"eval_samples_per_second": 209.599,
|
71 |
+
"eval_steps_per_second": 26.798,
|
72 |
+
"step": 61
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 3.41,
|
76 |
+
"learning_rate": 4.5833333333333334e-05,
|
77 |
+
"loss": 0.5469,
|
78 |
+
"step": 70
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 3.9,
|
82 |
+
"learning_rate": 4.4444444444444447e-05,
|
83 |
+
"loss": 0.5019,
|
84 |
+
"step": 80
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 4.0,
|
88 |
+
"eval_accuracy": 0.7214611872146118,
|
89 |
+
"eval_loss": 0.6695594191551208,
|
90 |
+
"eval_runtime": 1.0755,
|
91 |
+
"eval_samples_per_second": 203.631,
|
92 |
+
"eval_steps_per_second": 26.035,
|
93 |
+
"step": 82
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 4.39,
|
97 |
+
"learning_rate": 4.305555555555556e-05,
|
98 |
+
"loss": 0.5202,
|
99 |
+
"step": 90
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 4.88,
|
103 |
+
"learning_rate": 4.166666666666667e-05,
|
104 |
+
"loss": 0.4708,
|
105 |
+
"step": 100
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 4.98,
|
109 |
+
"eval_accuracy": 0.7534246575342466,
|
110 |
+
"eval_loss": 0.5861416459083557,
|
111 |
+
"eval_runtime": 1.2022,
|
112 |
+
"eval_samples_per_second": 182.173,
|
113 |
+
"eval_steps_per_second": 23.292,
|
114 |
+
"step": 102
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 5.37,
|
118 |
+
"learning_rate": 4.027777777777778e-05,
|
119 |
+
"loss": 0.5432,
|
120 |
+
"step": 110
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 5.85,
|
124 |
+
"learning_rate": 3.888888888888889e-05,
|
125 |
+
"loss": 0.3647,
|
126 |
+
"step": 120
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 6.0,
|
130 |
+
"eval_accuracy": 0.776255707762557,
|
131 |
+
"eval_loss": 0.5583679676055908,
|
132 |
+
"eval_runtime": 1.0741,
|
133 |
+
"eval_samples_per_second": 203.885,
|
134 |
+
"eval_steps_per_second": 26.068,
|
135 |
+
"step": 123
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 6.34,
|
139 |
+
"learning_rate": 3.7500000000000003e-05,
|
140 |
+
"loss": 0.3456,
|
141 |
+
"step": 130
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 6.83,
|
145 |
+
"learning_rate": 3.611111111111111e-05,
|
146 |
+
"loss": 0.2973,
|
147 |
+
"step": 140
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 6.98,
|
151 |
+
"eval_accuracy": 0.7671232876712328,
|
152 |
+
"eval_loss": 0.5783523321151733,
|
153 |
+
"eval_runtime": 1.4218,
|
154 |
+
"eval_samples_per_second": 154.026,
|
155 |
+
"eval_steps_per_second": 19.693,
|
156 |
+
"step": 143
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 7.32,
|
160 |
+
"learning_rate": 3.472222222222222e-05,
|
161 |
+
"loss": 0.3172,
|
162 |
+
"step": 150
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 7.8,
|
166 |
+
"learning_rate": 3.3333333333333335e-05,
|
167 |
+
"loss": 0.2827,
|
168 |
+
"step": 160
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 8.0,
|
172 |
+
"eval_accuracy": 0.7671232876712328,
|
173 |
+
"eval_loss": 0.5850614309310913,
|
174 |
+
"eval_runtime": 1.1505,
|
175 |
+
"eval_samples_per_second": 190.352,
|
176 |
+
"eval_steps_per_second": 24.337,
|
177 |
+
"step": 164
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 8.29,
|
181 |
+
"learning_rate": 3.194444444444444e-05,
|
182 |
+
"loss": 0.2399,
|
183 |
+
"step": 170
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 8.78,
|
187 |
+
"learning_rate": 3.055555555555556e-05,
|
188 |
+
"loss": 0.237,
|
189 |
+
"step": 180
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 8.98,
|
193 |
+
"eval_accuracy": 0.7625570776255708,
|
194 |
+
"eval_loss": 0.6790792942047119,
|
195 |
+
"eval_runtime": 1.21,
|
196 |
+
"eval_samples_per_second": 180.993,
|
197 |
+
"eval_steps_per_second": 23.141,
|
198 |
+
"step": 184
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 9.27,
|
202 |
+
"learning_rate": 2.916666666666667e-05,
|
203 |
+
"loss": 0.1989,
|
204 |
+
"step": 190
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 9.76,
|
208 |
+
"learning_rate": 2.777777777777778e-05,
|
209 |
+
"loss": 0.2505,
|
210 |
+
"step": 200
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 10.0,
|
214 |
+
"eval_accuracy": 0.7625570776255708,
|
215 |
+
"eval_loss": 0.5550326108932495,
|
216 |
+
"eval_runtime": 1.0754,
|
217 |
+
"eval_samples_per_second": 203.65,
|
218 |
+
"eval_steps_per_second": 26.037,
|
219 |
+
"step": 205
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 10.24,
|
223 |
+
"learning_rate": 2.6388888888888892e-05,
|
224 |
+
"loss": 0.2188,
|
225 |
+
"step": 210
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 10.73,
|
229 |
+
"learning_rate": 2.5e-05,
|
230 |
+
"loss": 0.2018,
|
231 |
+
"step": 220
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 10.98,
|
235 |
+
"eval_accuracy": 0.7625570776255708,
|
236 |
+
"eval_loss": 0.5445629358291626,
|
237 |
+
"eval_runtime": 1.0807,
|
238 |
+
"eval_samples_per_second": 202.648,
|
239 |
+
"eval_steps_per_second": 25.909,
|
240 |
+
"step": 225
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 11.22,
|
244 |
+
"learning_rate": 2.361111111111111e-05,
|
245 |
+
"loss": 0.2028,
|
246 |
+
"step": 230
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 11.71,
|
250 |
+
"learning_rate": 2.2222222222222223e-05,
|
251 |
+
"loss": 0.1841,
|
252 |
+
"step": 240
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 12.0,
|
256 |
+
"eval_accuracy": 0.7442922374429224,
|
257 |
+
"eval_loss": 0.5496687889099121,
|
258 |
+
"eval_runtime": 1.1278,
|
259 |
+
"eval_samples_per_second": 194.175,
|
260 |
+
"eval_steps_per_second": 24.826,
|
261 |
+
"step": 246
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 12.2,
|
265 |
+
"learning_rate": 2.0833333333333336e-05,
|
266 |
+
"loss": 0.165,
|
267 |
+
"step": 250
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 12.68,
|
271 |
+
"learning_rate": 1.9444444444444445e-05,
|
272 |
+
"loss": 0.1692,
|
273 |
+
"step": 260
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 12.98,
|
277 |
+
"eval_accuracy": 0.771689497716895,
|
278 |
+
"eval_loss": 0.5916743278503418,
|
279 |
+
"eval_runtime": 1.1094,
|
280 |
+
"eval_samples_per_second": 197.411,
|
281 |
+
"eval_steps_per_second": 25.24,
|
282 |
+
"step": 266
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 13.17,
|
286 |
+
"learning_rate": 1.8055555555555555e-05,
|
287 |
+
"loss": 0.2192,
|
288 |
+
"step": 270
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 13.66,
|
292 |
+
"learning_rate": 1.6666666666666667e-05,
|
293 |
+
"loss": 0.1624,
|
294 |
+
"step": 280
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 14.0,
|
298 |
+
"eval_accuracy": 0.776255707762557,
|
299 |
+
"eval_loss": 0.5254391431808472,
|
300 |
+
"eval_runtime": 1.1036,
|
301 |
+
"eval_samples_per_second": 198.45,
|
302 |
+
"eval_steps_per_second": 25.373,
|
303 |
+
"step": 287
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 14.15,
|
307 |
+
"learning_rate": 1.527777777777778e-05,
|
308 |
+
"loss": 0.2206,
|
309 |
+
"step": 290
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 14.63,
|
313 |
+
"learning_rate": 1.388888888888889e-05,
|
314 |
+
"loss": 0.1518,
|
315 |
+
"step": 300
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 14.98,
|
319 |
+
"eval_accuracy": 0.7808219178082192,
|
320 |
+
"eval_loss": 0.5295854806900024,
|
321 |
+
"eval_runtime": 1.059,
|
322 |
+
"eval_samples_per_second": 206.808,
|
323 |
+
"eval_steps_per_second": 26.441,
|
324 |
+
"step": 307
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 15.12,
|
328 |
+
"learning_rate": 1.25e-05,
|
329 |
+
"loss": 0.1465,
|
330 |
+
"step": 310
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 15.61,
|
334 |
+
"learning_rate": 1.1111111111111112e-05,
|
335 |
+
"loss": 0.1275,
|
336 |
+
"step": 320
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 16.0,
|
340 |
+
"eval_accuracy": 0.7625570776255708,
|
341 |
+
"eval_loss": 0.585834801197052,
|
342 |
+
"eval_runtime": 1.2635,
|
343 |
+
"eval_samples_per_second": 173.334,
|
344 |
+
"eval_steps_per_second": 22.161,
|
345 |
+
"step": 328
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 16.1,
|
349 |
+
"learning_rate": 9.722222222222223e-06,
|
350 |
+
"loss": 0.1237,
|
351 |
+
"step": 330
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 16.59,
|
355 |
+
"learning_rate": 8.333333333333334e-06,
|
356 |
+
"loss": 0.1107,
|
357 |
+
"step": 340
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 16.98,
|
361 |
+
"eval_accuracy": 0.776255707762557,
|
362 |
+
"eval_loss": 0.591891348361969,
|
363 |
+
"eval_runtime": 1.0639,
|
364 |
+
"eval_samples_per_second": 205.84,
|
365 |
+
"eval_steps_per_second": 26.317,
|
366 |
+
"step": 348
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 17.07,
|
370 |
+
"learning_rate": 6.944444444444445e-06,
|
371 |
+
"loss": 0.1422,
|
372 |
+
"step": 350
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 17.56,
|
376 |
+
"learning_rate": 5.555555555555556e-06,
|
377 |
+
"loss": 0.1192,
|
378 |
+
"step": 360
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 18.0,
|
382 |
+
"eval_accuracy": 0.771689497716895,
|
383 |
+
"eval_loss": 0.6027428507804871,
|
384 |
+
"eval_runtime": 1.0536,
|
385 |
+
"eval_samples_per_second": 207.851,
|
386 |
+
"eval_steps_per_second": 26.575,
|
387 |
+
"step": 369
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 18.05,
|
391 |
+
"learning_rate": 4.166666666666667e-06,
|
392 |
+
"loss": 0.1806,
|
393 |
+
"step": 370
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 18.54,
|
397 |
+
"learning_rate": 2.777777777777778e-06,
|
398 |
+
"loss": 0.0842,
|
399 |
+
"step": 380
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 18.98,
|
403 |
+
"eval_accuracy": 0.771689497716895,
|
404 |
+
"eval_loss": 0.6435291171073914,
|
405 |
+
"eval_runtime": 1.0647,
|
406 |
+
"eval_samples_per_second": 205.689,
|
407 |
+
"eval_steps_per_second": 26.298,
|
408 |
+
"step": 389
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 19.02,
|
412 |
+
"learning_rate": 1.388888888888889e-06,
|
413 |
+
"loss": 0.1594,
|
414 |
+
"step": 390
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 19.51,
|
418 |
+
"learning_rate": 0.0,
|
419 |
+
"loss": 0.1472,
|
420 |
+
"step": 400
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 19.51,
|
424 |
+
"eval_accuracy": 0.7671232876712328,
|
425 |
+
"eval_loss": 0.6201618313789368,
|
426 |
+
"eval_runtime": 1.0892,
|
427 |
+
"eval_samples_per_second": 201.069,
|
428 |
+
"eval_steps_per_second": 25.707,
|
429 |
+
"step": 400
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 19.51,
|
433 |
+
"step": 400,
|
434 |
+
"total_flos": 2.493693637268521e+17,
|
435 |
+
"train_loss": 0.3432186938822269,
|
436 |
+
"train_runtime": 208.3396,
|
437 |
+
"train_samples_per_second": 62.686,
|
438 |
+
"train_steps_per_second": 1.92
|
439 |
+
}
|
440 |
+
],
|
441 |
+
"logging_steps": 10,
|
442 |
+
"max_steps": 400,
|
443 |
+
"num_train_epochs": 20,
|
444 |
+
"save_steps": 500,
|
445 |
+
"total_flos": 2.493693637268521e+17,
|
446 |
+
"trial_name": null,
|
447 |
+
"trial_params": null
|
448 |
+
}
|