Model save
Browse files- README.md +12 -12
- all_results.json +11 -11
- eval_results.json +6 -6
- model.safetensors +1 -1
- runs/Dec15_03-46-10_d2a76fcee09b/events.out.tfevents.1702612927.d2a76fcee09b.36815.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +437 -122
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 [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) 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 |
|
@@ -65,15 +65,15 @@ The following hyperparameters were used during training:
|
|
65 |
|
66 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
67 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
68 |
-
| 0.
|
69 |
-
| 0.
|
70 |
-
| 0.
|
71 |
-
| 0.
|
72 |
-
| 0.
|
73 |
-
| 0.
|
74 |
-
| 0.
|
75 |
-
| 0.
|
76 |
-
| 0.
|
77 |
|
78 |
|
79 |
### Framework versions
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.9018518518518519
|
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 [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.4588
|
36 |
+
- Accuracy: 0.9019
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
65 |
|
66 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
67 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
68 |
+
| 0.1739 | 1.11 | 100 | 0.5771 | 0.8407 |
|
69 |
+
| 0.0677 | 2.22 | 200 | 0.5907 | 0.8519 |
|
70 |
+
| 0.0699 | 3.33 | 300 | 0.4160 | 0.8870 |
|
71 |
+
| 0.0598 | 4.44 | 400 | 0.7336 | 0.8380 |
|
72 |
+
| 0.0108 | 5.56 | 500 | 0.5133 | 0.8898 |
|
73 |
+
| 0.0082 | 6.67 | 600 | 0.4786 | 0.8981 |
|
74 |
+
| 0.0031 | 7.78 | 700 | 0.4624 | 0.9009 |
|
75 |
+
| 0.0046 | 8.89 | 800 | 0.4594 | 0.9 |
|
76 |
+
| 0.0052 | 10.0 | 900 | 0.4588 | 0.9019 |
|
77 |
|
78 |
|
79 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 9.
|
8 |
-
"total_flos":
|
9 |
-
"train_loss": 0.
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 1.
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.9175925925925926,
|
4 |
+
"eval_loss": 0.3201213479042053,
|
5 |
+
"eval_runtime": 14.3842,
|
6 |
+
"eval_samples_per_second": 75.083,
|
7 |
+
"eval_steps_per_second": 9.385,
|
8 |
+
"total_flos": 3.347863979094835e+18,
|
9 |
+
"train_loss": 0.24984459381964472,
|
10 |
+
"train_runtime": 897.5432,
|
11 |
+
"train_samples_per_second": 48.131,
|
12 |
+
"train_steps_per_second": 1.003
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 9.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.9175925925925926,
|
4 |
+
"eval_loss": 0.3201213479042053,
|
5 |
+
"eval_runtime": 14.3842,
|
6 |
+
"eval_samples_per_second": 75.083,
|
7 |
+
"eval_steps_per_second": 9.385
|
8 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 343245508
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:35885972c930164a157c2ac6031f4f0bbb728ae3cfb7310362498b088b84a04e
|
3 |
size 343245508
|
runs/Dec15_03-46-10_d2a76fcee09b/events.out.tfevents.1702612927.d2a76fcee09b.36815.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:16d757943267b444b38a2eb6d28bd4471684cd5e9855597fe224d7013bf97463
|
3 |
+
size 22193
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 1.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"total_flos": 3.347863979094835e+18,
|
4 |
+
"train_loss": 0.24984459381964472,
|
5 |
+
"train_runtime": 897.5432,
|
6 |
+
"train_samples_per_second": 48.131,
|
7 |
+
"train_steps_per_second": 1.003
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,334 +1,649 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
"eval_steps": 100,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.11,
|
13 |
-
"learning_rate": 0.
|
14 |
-
"loss":
|
15 |
"step": 10
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.22,
|
19 |
-
"learning_rate": 0.
|
20 |
-
"loss":
|
21 |
"step": 20
|
22 |
},
|
23 |
{
|
24 |
"epoch": 0.33,
|
25 |
-
"learning_rate": 0.
|
26 |
-
"loss":
|
27 |
"step": 30
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.44,
|
31 |
-
"learning_rate": 0.
|
32 |
-
"loss":
|
33 |
"step": 40
|
34 |
},
|
35 |
{
|
36 |
"epoch": 0.56,
|
37 |
-
"learning_rate": 0.
|
38 |
-
"loss":
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.67,
|
43 |
-
"learning_rate": 0.
|
44 |
-
"loss": 0.
|
45 |
"step": 60
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.78,
|
49 |
-
"learning_rate": 0.
|
50 |
-
"loss": 0.
|
51 |
"step": 70
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.89,
|
55 |
-
"learning_rate": 0.
|
56 |
-
"loss": 0.
|
57 |
"step": 80
|
58 |
},
|
59 |
{
|
60 |
"epoch": 1.0,
|
61 |
-
"learning_rate": 0.
|
62 |
-
"loss": 0.
|
63 |
"step": 90
|
64 |
},
|
65 |
{
|
66 |
"epoch": 1.11,
|
67 |
-
"learning_rate": 0.
|
68 |
-
"loss": 0.
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 1.11,
|
73 |
-
"eval_accuracy": 0.
|
74 |
-
"eval_loss": 0.
|
75 |
-
"eval_runtime":
|
76 |
-
"eval_samples_per_second":
|
77 |
-
"eval_steps_per_second":
|
78 |
"step": 100
|
79 |
},
|
80 |
{
|
81 |
"epoch": 1.22,
|
82 |
-
"learning_rate": 0.
|
83 |
-
"loss": 0.
|
84 |
"step": 110
|
85 |
},
|
86 |
{
|
87 |
"epoch": 1.33,
|
88 |
-
"learning_rate": 0.
|
89 |
-
"loss": 0.
|
90 |
"step": 120
|
91 |
},
|
92 |
{
|
93 |
"epoch": 1.44,
|
94 |
-
"learning_rate": 0.
|
95 |
-
"loss": 0.
|
96 |
"step": 130
|
97 |
},
|
98 |
{
|
99 |
"epoch": 1.56,
|
100 |
-
"learning_rate": 0.
|
101 |
-
"loss": 0.
|
102 |
"step": 140
|
103 |
},
|
104 |
{
|
105 |
"epoch": 1.67,
|
106 |
-
"learning_rate": 0.
|
107 |
-
"loss": 0.
|
108 |
"step": 150
|
109 |
},
|
110 |
{
|
111 |
"epoch": 1.78,
|
112 |
-
"learning_rate": 0.
|
113 |
-
"loss": 0.
|
114 |
"step": 160
|
115 |
},
|
116 |
{
|
117 |
"epoch": 1.89,
|
118 |
-
"learning_rate": 0.
|
119 |
-
"loss": 0.
|
120 |
"step": 170
|
121 |
},
|
122 |
{
|
123 |
"epoch": 2.0,
|
124 |
-
"learning_rate": 0.
|
125 |
-
"loss": 0.
|
126 |
"step": 180
|
127 |
},
|
128 |
{
|
129 |
"epoch": 2.11,
|
130 |
-
"learning_rate": 0.
|
131 |
-
"loss": 0.
|
132 |
"step": 190
|
133 |
},
|
134 |
{
|
135 |
"epoch": 2.22,
|
136 |
-
"learning_rate": 0.
|
137 |
-
"loss": 0.
|
138 |
"step": 200
|
139 |
},
|
140 |
{
|
141 |
"epoch": 2.22,
|
142 |
-
"eval_accuracy": 0.
|
143 |
-
"eval_loss": 0.
|
144 |
-
"eval_runtime": 12.
|
145 |
-
"eval_samples_per_second":
|
146 |
-
"eval_steps_per_second": 10.
|
147 |
"step": 200
|
148 |
},
|
149 |
{
|
150 |
"epoch": 2.33,
|
151 |
-
"learning_rate": 0.
|
152 |
-
"loss": 0.
|
153 |
"step": 210
|
154 |
},
|
155 |
{
|
156 |
"epoch": 2.44,
|
157 |
-
"learning_rate": 0.
|
158 |
-
"loss": 0.
|
159 |
"step": 220
|
160 |
},
|
161 |
{
|
162 |
"epoch": 2.56,
|
163 |
-
"learning_rate":
|
164 |
-
"loss": 0.
|
165 |
"step": 230
|
166 |
},
|
167 |
{
|
168 |
"epoch": 2.67,
|
169 |
-
"learning_rate":
|
170 |
-
"loss": 0.
|
171 |
"step": 240
|
172 |
},
|
173 |
{
|
174 |
"epoch": 2.78,
|
175 |
-
"learning_rate":
|
176 |
-
"loss": 0.
|
177 |
"step": 250
|
178 |
},
|
179 |
{
|
180 |
"epoch": 2.89,
|
181 |
-
"learning_rate":
|
182 |
-
"loss": 0.
|
183 |
"step": 260
|
184 |
},
|
185 |
{
|
186 |
"epoch": 3.0,
|
187 |
-
"learning_rate":
|
188 |
-
"loss": 0.
|
189 |
"step": 270
|
190 |
},
|
191 |
{
|
192 |
"epoch": 3.11,
|
193 |
-
"learning_rate":
|
194 |
-
"loss": 0.
|
195 |
"step": 280
|
196 |
},
|
197 |
{
|
198 |
"epoch": 3.22,
|
199 |
-
"learning_rate":
|
200 |
-
"loss": 0.
|
201 |
"step": 290
|
202 |
},
|
203 |
{
|
204 |
"epoch": 3.33,
|
205 |
-
"learning_rate":
|
206 |
-
"loss": 0.
|
207 |
"step": 300
|
208 |
},
|
209 |
{
|
210 |
"epoch": 3.33,
|
211 |
-
"eval_accuracy": 0.
|
212 |
-
"eval_loss": 0.
|
213 |
-
"eval_runtime": 12.
|
214 |
-
"eval_samples_per_second":
|
215 |
-
"eval_steps_per_second": 10.
|
216 |
"step": 300
|
217 |
},
|
218 |
{
|
219 |
"epoch": 3.44,
|
220 |
-
"learning_rate":
|
221 |
-
"loss": 0.
|
222 |
"step": 310
|
223 |
},
|
224 |
{
|
225 |
"epoch": 3.56,
|
226 |
-
"learning_rate":
|
227 |
-
"loss": 0.
|
228 |
"step": 320
|
229 |
},
|
230 |
{
|
231 |
"epoch": 3.67,
|
232 |
-
"learning_rate":
|
233 |
-
"loss": 0.
|
234 |
"step": 330
|
235 |
},
|
236 |
{
|
237 |
"epoch": 3.78,
|
238 |
-
"learning_rate":
|
239 |
-
"loss": 0.
|
240 |
"step": 340
|
241 |
},
|
242 |
{
|
243 |
"epoch": 3.89,
|
244 |
-
"learning_rate":
|
245 |
-
"loss": 0.
|
246 |
"step": 350
|
247 |
},
|
248 |
{
|
249 |
"epoch": 4.0,
|
250 |
-
"learning_rate":
|
251 |
-
"loss": 0.
|
252 |
"step": 360
|
253 |
},
|
254 |
{
|
255 |
"epoch": 4.11,
|
256 |
-
"learning_rate":
|
257 |
-
"loss": 0.
|
258 |
"step": 370
|
259 |
},
|
260 |
{
|
261 |
"epoch": 4.22,
|
262 |
-
"learning_rate":
|
263 |
-
"loss": 0.
|
264 |
"step": 380
|
265 |
},
|
266 |
{
|
267 |
"epoch": 4.33,
|
268 |
-
"learning_rate":
|
269 |
-
"loss": 0.
|
270 |
"step": 390
|
271 |
},
|
272 |
{
|
273 |
"epoch": 4.44,
|
274 |
-
"learning_rate":
|
275 |
-
"loss": 0.
|
276 |
"step": 400
|
277 |
},
|
278 |
{
|
279 |
"epoch": 4.44,
|
280 |
-
"eval_accuracy": 0.
|
281 |
-
"eval_loss": 0.
|
282 |
-
"eval_runtime":
|
283 |
-
"eval_samples_per_second":
|
284 |
-
"eval_steps_per_second":
|
285 |
"step": 400
|
286 |
},
|
287 |
{
|
288 |
"epoch": 4.56,
|
289 |
-
"learning_rate":
|
290 |
-
"loss": 0.
|
291 |
"step": 410
|
292 |
},
|
293 |
{
|
294 |
"epoch": 4.67,
|
295 |
-
"learning_rate":
|
296 |
-
"loss": 0.
|
297 |
"step": 420
|
298 |
},
|
299 |
{
|
300 |
"epoch": 4.78,
|
301 |
-
"learning_rate":
|
302 |
-
"loss": 0.
|
303 |
"step": 430
|
304 |
},
|
305 |
{
|
306 |
"epoch": 4.89,
|
307 |
-
"learning_rate":
|
308 |
-
"loss": 0.
|
309 |
"step": 440
|
310 |
},
|
311 |
{
|
312 |
"epoch": 5.0,
|
313 |
-
"learning_rate": 0.
|
314 |
-
"loss": 0.
|
315 |
"step": 450
|
316 |
},
|
317 |
{
|
318 |
-
"epoch": 5.
|
319 |
-
"
|
320 |
-
"
|
321 |
-
"
|
322 |
-
|
323 |
-
|
324 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
325 |
}
|
326 |
],
|
327 |
"logging_steps": 10,
|
328 |
-
"max_steps":
|
329 |
-
"num_train_epochs":
|
330 |
"save_steps": 1000,
|
331 |
-
"total_flos":
|
332 |
"trial_name": null,
|
333 |
"trial_params": null
|
334 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 10.0,
|
5 |
"eval_steps": 100,
|
6 |
+
"global_step": 900,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.11,
|
13 |
+
"learning_rate": 0.00019777777777777778,
|
14 |
+
"loss": 1.9886,
|
15 |
"step": 10
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.22,
|
19 |
+
"learning_rate": 0.00019555555555555556,
|
20 |
+
"loss": 1.6024,
|
21 |
"step": 20
|
22 |
},
|
23 |
{
|
24 |
"epoch": 0.33,
|
25 |
+
"learning_rate": 0.00019333333333333333,
|
26 |
+
"loss": 1.3753,
|
27 |
"step": 30
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.44,
|
31 |
+
"learning_rate": 0.00019111111111111114,
|
32 |
+
"loss": 1.3624,
|
33 |
"step": 40
|
34 |
},
|
35 |
{
|
36 |
"epoch": 0.56,
|
37 |
+
"learning_rate": 0.00018888888888888888,
|
38 |
+
"loss": 1.3255,
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.67,
|
43 |
+
"learning_rate": 0.0001866666666666667,
|
44 |
+
"loss": 0.9955,
|
45 |
"step": 60
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.78,
|
49 |
+
"learning_rate": 0.00018444444444444446,
|
50 |
+
"loss": 0.9689,
|
51 |
"step": 70
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.89,
|
55 |
+
"learning_rate": 0.00018222222222222224,
|
56 |
+
"loss": 0.9057,
|
57 |
"step": 80
|
58 |
},
|
59 |
{
|
60 |
"epoch": 1.0,
|
61 |
+
"learning_rate": 0.00018,
|
62 |
+
"loss": 0.8525,
|
63 |
"step": 90
|
64 |
},
|
65 |
{
|
66 |
"epoch": 1.11,
|
67 |
+
"learning_rate": 0.00017777777777777779,
|
68 |
+
"loss": 0.6539,
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 1.11,
|
73 |
+
"eval_accuracy": 0.75,
|
74 |
+
"eval_loss": 0.8062787652015686,
|
75 |
+
"eval_runtime": 12.7041,
|
76 |
+
"eval_samples_per_second": 85.012,
|
77 |
+
"eval_steps_per_second": 10.626,
|
78 |
"step": 100
|
79 |
},
|
80 |
{
|
81 |
"epoch": 1.22,
|
82 |
+
"learning_rate": 0.00017555555555555556,
|
83 |
+
"loss": 0.6373,
|
84 |
"step": 110
|
85 |
},
|
86 |
{
|
87 |
"epoch": 1.33,
|
88 |
+
"learning_rate": 0.00017333333333333334,
|
89 |
+
"loss": 0.5875,
|
90 |
"step": 120
|
91 |
},
|
92 |
{
|
93 |
"epoch": 1.44,
|
94 |
+
"learning_rate": 0.0001711111111111111,
|
95 |
+
"loss": 0.5479,
|
96 |
"step": 130
|
97 |
},
|
98 |
{
|
99 |
"epoch": 1.56,
|
100 |
+
"learning_rate": 0.00016888888888888889,
|
101 |
+
"loss": 0.5463,
|
102 |
"step": 140
|
103 |
},
|
104 |
{
|
105 |
"epoch": 1.67,
|
106 |
+
"learning_rate": 0.0001666666666666667,
|
107 |
+
"loss": 0.5625,
|
108 |
"step": 150
|
109 |
},
|
110 |
{
|
111 |
"epoch": 1.78,
|
112 |
+
"learning_rate": 0.00016444444444444444,
|
113 |
+
"loss": 0.5033,
|
114 |
"step": 160
|
115 |
},
|
116 |
{
|
117 |
"epoch": 1.89,
|
118 |
+
"learning_rate": 0.00016222222222222224,
|
119 |
+
"loss": 0.5121,
|
120 |
"step": 170
|
121 |
},
|
122 |
{
|
123 |
"epoch": 2.0,
|
124 |
+
"learning_rate": 0.00016,
|
125 |
+
"loss": 0.5484,
|
126 |
"step": 180
|
127 |
},
|
128 |
{
|
129 |
"epoch": 2.11,
|
130 |
+
"learning_rate": 0.0001577777777777778,
|
131 |
+
"loss": 0.3358,
|
132 |
"step": 190
|
133 |
},
|
134 |
{
|
135 |
"epoch": 2.22,
|
136 |
+
"learning_rate": 0.00015555555555555556,
|
137 |
+
"loss": 0.2553,
|
138 |
"step": 200
|
139 |
},
|
140 |
{
|
141 |
"epoch": 2.22,
|
142 |
+
"eval_accuracy": 0.8351851851851851,
|
143 |
+
"eval_loss": 0.5555349588394165,
|
144 |
+
"eval_runtime": 12.687,
|
145 |
+
"eval_samples_per_second": 85.127,
|
146 |
+
"eval_steps_per_second": 10.641,
|
147 |
"step": 200
|
148 |
},
|
149 |
{
|
150 |
"epoch": 2.33,
|
151 |
+
"learning_rate": 0.00015333333333333334,
|
152 |
+
"loss": 0.3577,
|
153 |
"step": 210
|
154 |
},
|
155 |
{
|
156 |
"epoch": 2.44,
|
157 |
+
"learning_rate": 0.0001511111111111111,
|
158 |
+
"loss": 0.3141,
|
159 |
"step": 220
|
160 |
},
|
161 |
{
|
162 |
"epoch": 2.56,
|
163 |
+
"learning_rate": 0.0001488888888888889,
|
164 |
+
"loss": 0.3053,
|
165 |
"step": 230
|
166 |
},
|
167 |
{
|
168 |
"epoch": 2.67,
|
169 |
+
"learning_rate": 0.00014666666666666666,
|
170 |
+
"loss": 0.2487,
|
171 |
"step": 240
|
172 |
},
|
173 |
{
|
174 |
"epoch": 2.78,
|
175 |
+
"learning_rate": 0.00014444444444444444,
|
176 |
+
"loss": 0.3129,
|
177 |
"step": 250
|
178 |
},
|
179 |
{
|
180 |
"epoch": 2.89,
|
181 |
+
"learning_rate": 0.00014222222222222224,
|
182 |
+
"loss": 0.2975,
|
183 |
"step": 260
|
184 |
},
|
185 |
{
|
186 |
"epoch": 3.0,
|
187 |
+
"learning_rate": 0.00014,
|
188 |
+
"loss": 0.2661,
|
189 |
"step": 270
|
190 |
},
|
191 |
{
|
192 |
"epoch": 3.11,
|
193 |
+
"learning_rate": 0.0001377777777777778,
|
194 |
+
"loss": 0.1503,
|
195 |
"step": 280
|
196 |
},
|
197 |
{
|
198 |
"epoch": 3.22,
|
199 |
+
"learning_rate": 0.00013555555555555556,
|
200 |
+
"loss": 0.1649,
|
201 |
"step": 290
|
202 |
},
|
203 |
{
|
204 |
"epoch": 3.33,
|
205 |
+
"learning_rate": 0.00013333333333333334,
|
206 |
+
"loss": 0.1909,
|
207 |
"step": 300
|
208 |
},
|
209 |
{
|
210 |
"epoch": 3.33,
|
211 |
+
"eval_accuracy": 0.8453703703703703,
|
212 |
+
"eval_loss": 0.5217463970184326,
|
213 |
+
"eval_runtime": 12.8584,
|
214 |
+
"eval_samples_per_second": 83.992,
|
215 |
+
"eval_steps_per_second": 10.499,
|
216 |
"step": 300
|
217 |
},
|
218 |
{
|
219 |
"epoch": 3.44,
|
220 |
+
"learning_rate": 0.00013111111111111111,
|
221 |
+
"loss": 0.1387,
|
222 |
"step": 310
|
223 |
},
|
224 |
{
|
225 |
"epoch": 3.56,
|
226 |
+
"learning_rate": 0.00012888888888888892,
|
227 |
+
"loss": 0.1357,
|
228 |
"step": 320
|
229 |
},
|
230 |
{
|
231 |
"epoch": 3.67,
|
232 |
+
"learning_rate": 0.00012666666666666666,
|
233 |
+
"loss": 0.1582,
|
234 |
"step": 330
|
235 |
},
|
236 |
{
|
237 |
"epoch": 3.78,
|
238 |
+
"learning_rate": 0.00012444444444444444,
|
239 |
+
"loss": 0.1676,
|
240 |
"step": 340
|
241 |
},
|
242 |
{
|
243 |
"epoch": 3.89,
|
244 |
+
"learning_rate": 0.00012222222222222224,
|
245 |
+
"loss": 0.156,
|
246 |
"step": 350
|
247 |
},
|
248 |
{
|
249 |
"epoch": 4.0,
|
250 |
+
"learning_rate": 0.00012,
|
251 |
+
"loss": 0.1711,
|
252 |
"step": 360
|
253 |
},
|
254 |
{
|
255 |
"epoch": 4.11,
|
256 |
+
"learning_rate": 0.00011777777777777779,
|
257 |
+
"loss": 0.1132,
|
258 |
"step": 370
|
259 |
},
|
260 |
{
|
261 |
"epoch": 4.22,
|
262 |
+
"learning_rate": 0.00011555555555555555,
|
263 |
+
"loss": 0.1259,
|
264 |
"step": 380
|
265 |
},
|
266 |
{
|
267 |
"epoch": 4.33,
|
268 |
+
"learning_rate": 0.00011333333333333334,
|
269 |
+
"loss": 0.1086,
|
270 |
"step": 390
|
271 |
},
|
272 |
{
|
273 |
"epoch": 4.44,
|
274 |
+
"learning_rate": 0.00011111111111111112,
|
275 |
+
"loss": 0.0999,
|
276 |
"step": 400
|
277 |
},
|
278 |
{
|
279 |
"epoch": 4.44,
|
280 |
+
"eval_accuracy": 0.8722222222222222,
|
281 |
+
"eval_loss": 0.5075178146362305,
|
282 |
+
"eval_runtime": 15.6251,
|
283 |
+
"eval_samples_per_second": 69.119,
|
284 |
+
"eval_steps_per_second": 8.64,
|
285 |
"step": 400
|
286 |
},
|
287 |
{
|
288 |
"epoch": 4.56,
|
289 |
+
"learning_rate": 0.00010888888888888889,
|
290 |
+
"loss": 0.0736,
|
291 |
"step": 410
|
292 |
},
|
293 |
{
|
294 |
"epoch": 4.67,
|
295 |
+
"learning_rate": 0.00010666666666666667,
|
296 |
+
"loss": 0.0755,
|
297 |
"step": 420
|
298 |
},
|
299 |
{
|
300 |
"epoch": 4.78,
|
301 |
+
"learning_rate": 0.00010444444444444445,
|
302 |
+
"loss": 0.0929,
|
303 |
"step": 430
|
304 |
},
|
305 |
{
|
306 |
"epoch": 4.89,
|
307 |
+
"learning_rate": 0.00010222222222222222,
|
308 |
+
"loss": 0.0873,
|
309 |
"step": 440
|
310 |
},
|
311 |
{
|
312 |
"epoch": 5.0,
|
313 |
+
"learning_rate": 0.0001,
|
314 |
+
"loss": 0.1017,
|
315 |
"step": 450
|
316 |
},
|
317 |
{
|
318 |
+
"epoch": 5.11,
|
319 |
+
"learning_rate": 9.777777777777778e-05,
|
320 |
+
"loss": 0.0312,
|
321 |
+
"step": 460
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 5.22,
|
325 |
+
"learning_rate": 9.555555555555557e-05,
|
326 |
+
"loss": 0.0707,
|
327 |
+
"step": 470
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 5.33,
|
331 |
+
"learning_rate": 9.333333333333334e-05,
|
332 |
+
"loss": 0.0412,
|
333 |
+
"step": 480
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 5.44,
|
337 |
+
"learning_rate": 9.111111111111112e-05,
|
338 |
+
"loss": 0.0315,
|
339 |
+
"step": 490
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 5.56,
|
343 |
+
"learning_rate": 8.888888888888889e-05,
|
344 |
+
"loss": 0.0666,
|
345 |
+
"step": 500
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 5.56,
|
349 |
+
"eval_accuracy": 0.8768518518518519,
|
350 |
+
"eval_loss": 0.4633159637451172,
|
351 |
+
"eval_runtime": 12.8293,
|
352 |
+
"eval_samples_per_second": 84.183,
|
353 |
+
"eval_steps_per_second": 10.523,
|
354 |
+
"step": 500
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 5.67,
|
358 |
+
"learning_rate": 8.666666666666667e-05,
|
359 |
+
"loss": 0.0705,
|
360 |
+
"step": 510
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 5.78,
|
364 |
+
"learning_rate": 8.444444444444444e-05,
|
365 |
+
"loss": 0.0272,
|
366 |
+
"step": 520
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 5.89,
|
370 |
+
"learning_rate": 8.222222222222222e-05,
|
371 |
+
"loss": 0.0291,
|
372 |
+
"step": 530
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 6.0,
|
376 |
+
"learning_rate": 8e-05,
|
377 |
+
"loss": 0.0503,
|
378 |
+
"step": 540
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 6.11,
|
382 |
+
"learning_rate": 7.777777777777778e-05,
|
383 |
+
"loss": 0.031,
|
384 |
+
"step": 550
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 6.22,
|
388 |
+
"learning_rate": 7.555555555555556e-05,
|
389 |
+
"loss": 0.0474,
|
390 |
+
"step": 560
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 6.33,
|
394 |
+
"learning_rate": 7.333333333333333e-05,
|
395 |
+
"loss": 0.0203,
|
396 |
+
"step": 570
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 6.44,
|
400 |
+
"learning_rate": 7.111111111111112e-05,
|
401 |
+
"loss": 0.0235,
|
402 |
+
"step": 580
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 6.56,
|
406 |
+
"learning_rate": 6.88888888888889e-05,
|
407 |
+
"loss": 0.0203,
|
408 |
+
"step": 590
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 6.67,
|
412 |
+
"learning_rate": 6.666666666666667e-05,
|
413 |
+
"loss": 0.0392,
|
414 |
+
"step": 600
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 6.67,
|
418 |
+
"eval_accuracy": 0.8740740740740741,
|
419 |
+
"eval_loss": 0.4613787829875946,
|
420 |
+
"eval_runtime": 13.0592,
|
421 |
+
"eval_samples_per_second": 82.7,
|
422 |
+
"eval_steps_per_second": 10.338,
|
423 |
+
"step": 600
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 6.78,
|
427 |
+
"learning_rate": 6.444444444444446e-05,
|
428 |
+
"loss": 0.0242,
|
429 |
+
"step": 610
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 6.89,
|
433 |
+
"learning_rate": 6.222222222222222e-05,
|
434 |
+
"loss": 0.0277,
|
435 |
+
"step": 620
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 7.0,
|
439 |
+
"learning_rate": 6e-05,
|
440 |
+
"loss": 0.0151,
|
441 |
+
"step": 630
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 7.11,
|
445 |
+
"learning_rate": 5.7777777777777776e-05,
|
446 |
+
"loss": 0.0189,
|
447 |
+
"step": 640
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 7.22,
|
451 |
+
"learning_rate": 5.555555555555556e-05,
|
452 |
+
"loss": 0.0181,
|
453 |
+
"step": 650
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 7.33,
|
457 |
+
"learning_rate": 5.333333333333333e-05,
|
458 |
+
"loss": 0.0133,
|
459 |
+
"step": 660
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 7.44,
|
463 |
+
"learning_rate": 5.111111111111111e-05,
|
464 |
+
"loss": 0.0236,
|
465 |
+
"step": 670
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 7.56,
|
469 |
+
"learning_rate": 4.888888888888889e-05,
|
470 |
+
"loss": 0.015,
|
471 |
+
"step": 680
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 7.67,
|
475 |
+
"learning_rate": 4.666666666666667e-05,
|
476 |
+
"loss": 0.0137,
|
477 |
+
"step": 690
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 7.78,
|
481 |
+
"learning_rate": 4.4444444444444447e-05,
|
482 |
+
"loss": 0.0111,
|
483 |
+
"step": 700
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 7.78,
|
487 |
+
"eval_accuracy": 0.9101851851851852,
|
488 |
+
"eval_loss": 0.35738176107406616,
|
489 |
+
"eval_runtime": 12.567,
|
490 |
+
"eval_samples_per_second": 85.939,
|
491 |
+
"eval_steps_per_second": 10.742,
|
492 |
+
"step": 700
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 7.89,
|
496 |
+
"learning_rate": 4.222222222222222e-05,
|
497 |
+
"loss": 0.0182,
|
498 |
+
"step": 710
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 8.0,
|
502 |
+
"learning_rate": 4e-05,
|
503 |
+
"loss": 0.0135,
|
504 |
+
"step": 720
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 8.11,
|
508 |
+
"learning_rate": 3.777777777777778e-05,
|
509 |
+
"loss": 0.0121,
|
510 |
+
"step": 730
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 8.22,
|
514 |
+
"learning_rate": 3.555555555555556e-05,
|
515 |
+
"loss": 0.013,
|
516 |
+
"step": 740
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 8.33,
|
520 |
+
"learning_rate": 3.3333333333333335e-05,
|
521 |
+
"loss": 0.0111,
|
522 |
+
"step": 750
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 8.44,
|
526 |
+
"learning_rate": 3.111111111111111e-05,
|
527 |
+
"loss": 0.0104,
|
528 |
+
"step": 760
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 8.56,
|
532 |
+
"learning_rate": 2.8888888888888888e-05,
|
533 |
+
"loss": 0.0122,
|
534 |
+
"step": 770
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 8.67,
|
538 |
+
"learning_rate": 2.6666666666666667e-05,
|
539 |
+
"loss": 0.0155,
|
540 |
+
"step": 780
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 8.78,
|
544 |
+
"learning_rate": 2.4444444444444445e-05,
|
545 |
+
"loss": 0.0089,
|
546 |
+
"step": 790
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 8.89,
|
550 |
+
"learning_rate": 2.2222222222222223e-05,
|
551 |
+
"loss": 0.0122,
|
552 |
+
"step": 800
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 8.89,
|
556 |
+
"eval_accuracy": 0.9166666666666666,
|
557 |
+
"eval_loss": 0.3158920705318451,
|
558 |
+
"eval_runtime": 13.6141,
|
559 |
+
"eval_samples_per_second": 79.33,
|
560 |
+
"eval_steps_per_second": 9.916,
|
561 |
+
"step": 800
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 9.0,
|
565 |
+
"learning_rate": 2e-05,
|
566 |
+
"loss": 0.0082,
|
567 |
+
"step": 810
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 9.11,
|
571 |
+
"learning_rate": 1.777777777777778e-05,
|
572 |
+
"loss": 0.0118,
|
573 |
+
"step": 820
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 9.22,
|
577 |
+
"learning_rate": 1.5555555555555555e-05,
|
578 |
+
"loss": 0.009,
|
579 |
+
"step": 830
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 9.33,
|
583 |
+
"learning_rate": 1.3333333333333333e-05,
|
584 |
+
"loss": 0.0079,
|
585 |
+
"step": 840
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 9.44,
|
589 |
+
"learning_rate": 1.1111111111111112e-05,
|
590 |
+
"loss": 0.008,
|
591 |
+
"step": 850
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 9.56,
|
595 |
+
"learning_rate": 8.88888888888889e-06,
|
596 |
+
"loss": 0.0154,
|
597 |
+
"step": 860
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 9.67,
|
601 |
+
"learning_rate": 6.666666666666667e-06,
|
602 |
+
"loss": 0.0096,
|
603 |
+
"step": 870
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 9.78,
|
607 |
+
"learning_rate": 4.444444444444445e-06,
|
608 |
+
"loss": 0.0078,
|
609 |
+
"step": 880
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 9.89,
|
613 |
+
"learning_rate": 2.2222222222222225e-06,
|
614 |
+
"loss": 0.0081,
|
615 |
+
"step": 890
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 10.0,
|
619 |
+
"learning_rate": 0.0,
|
620 |
+
"loss": 0.0112,
|
621 |
+
"step": 900
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 10.0,
|
625 |
+
"eval_accuracy": 0.9175925925925926,
|
626 |
+
"eval_loss": 0.3201213479042053,
|
627 |
+
"eval_runtime": 12.9343,
|
628 |
+
"eval_samples_per_second": 83.499,
|
629 |
+
"eval_steps_per_second": 10.437,
|
630 |
+
"step": 900
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 10.0,
|
634 |
+
"step": 900,
|
635 |
+
"total_flos": 3.347863979094835e+18,
|
636 |
+
"train_loss": 0.24984459381964472,
|
637 |
+
"train_runtime": 897.5432,
|
638 |
+
"train_samples_per_second": 48.131,
|
639 |
+
"train_steps_per_second": 1.003
|
640 |
}
|
641 |
],
|
642 |
"logging_steps": 10,
|
643 |
+
"max_steps": 900,
|
644 |
+
"num_train_epochs": 10,
|
645 |
"save_steps": 1000,
|
646 |
+
"total_flos": 3.347863979094835e+18,
|
647 |
"trial_name": null,
|
648 |
"trial_params": null
|
649 |
}
|