batterydata
commited on
Commit
·
e2eab22
1
Parent(s):
080d91f
update
Browse files- all_results.json +12 -12
- config.json +1 -1
- eval_results.json +8 -8
- pytorch_model.bin +1 -1
- tokenizer_config.json +1 -1
- train_results.json +4 -4
- trainer_state.json +148 -148
- training_args.bin +1 -1
all_results.json
CHANGED
@@ -1,17 +1,17 @@
|
|
1 |
{
|
2 |
"epoch": 15.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_f1": 0.
|
5 |
-
"eval_loss": 0.
|
6 |
-
"eval_precision": 0.
|
7 |
-
"eval_recall": 0.
|
8 |
-
"eval_runtime":
|
9 |
"eval_samples": 1451,
|
10 |
-
"eval_samples_per_second":
|
11 |
-
"eval_steps_per_second":
|
12 |
-
"train_loss": 0.
|
13 |
-
"train_runtime":
|
14 |
"train_samples": 13054,
|
15 |
-
"train_samples_per_second":
|
16 |
-
"train_steps_per_second":
|
17 |
}
|
|
|
1 |
{
|
2 |
"epoch": 15.0,
|
3 |
+
"eval_accuracy": 0.978401158962202,
|
4 |
+
"eval_f1": 0.96785434549785,
|
5 |
+
"eval_loss": 0.15796583890914917,
|
6 |
+
"eval_precision": 0.9678388746803069,
|
7 |
+
"eval_recall": 0.9678698168100003,
|
8 |
+
"eval_runtime": 3.7378,
|
9 |
"eval_samples": 1451,
|
10 |
+
"eval_samples_per_second": 388.2,
|
11 |
+
"eval_steps_per_second": 24.346,
|
12 |
+
"train_loss": 0.03262665262993644,
|
13 |
+
"train_runtime": 785.1159,
|
14 |
"train_samples": 13054,
|
15 |
+
"train_samples_per_second": 249.403,
|
16 |
+
"train_steps_per_second": 15.59
|
17 |
}
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "/lus/theta-fs0/projects/SolarWindowsADSP/shu/models/
|
3 |
"architectures": [
|
4 |
"BertForTokenClassification"
|
5 |
],
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "/lus/theta-fs0/projects/SolarWindowsADSP/shu/models/best_models/bert-base-cased/",
|
3 |
"architectures": [
|
4 |
"BertForTokenClassification"
|
5 |
],
|
eval_results.json
CHANGED
@@ -1,12 +1,12 @@
|
|
1 |
{
|
2 |
"epoch": 15.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_f1": 0.
|
5 |
-
"eval_loss": 0.
|
6 |
-
"eval_precision": 0.
|
7 |
-
"eval_recall": 0.
|
8 |
-
"eval_runtime":
|
9 |
"eval_samples": 1451,
|
10 |
-
"eval_samples_per_second":
|
11 |
-
"eval_steps_per_second":
|
12 |
}
|
|
|
1 |
{
|
2 |
"epoch": 15.0,
|
3 |
+
"eval_accuracy": 0.978401158962202,
|
4 |
+
"eval_f1": 0.96785434549785,
|
5 |
+
"eval_loss": 0.15796583890914917,
|
6 |
+
"eval_precision": 0.9678388746803069,
|
7 |
+
"eval_recall": 0.9678698168100003,
|
8 |
+
"eval_runtime": 3.7378,
|
9 |
"eval_samples": 1451,
|
10 |
+
"eval_samples_per_second": 388.2,
|
11 |
+
"eval_steps_per_second": 24.346
|
12 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 431110641
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a2501b95940948e0e7914b12893a4db2e51e91522e86608321b56de7825d840c
|
3 |
size 431110641
|
tokenizer_config.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "
|
|
|
1 |
+
{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "/lus/theta-fs0/projects/SolarWindowsADSP/shu/models/best_models/bert-base-cased/", "tokenizer_class": "BertTokenizer"}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 15.0,
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
"train_samples": 13054,
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 15.0,
|
3 |
+
"train_loss": 0.03262665262993644,
|
4 |
+
"train_runtime": 785.1159,
|
5 |
"train_samples": 13054,
|
6 |
+
"train_samples_per_second": 249.403,
|
7 |
+
"train_steps_per_second": 15.59
|
8 |
}
|
trainer_state.json
CHANGED
@@ -10,335 +10,335 @@
|
|
10 |
{
|
11 |
"epoch": 0.61,
|
12 |
"learning_rate": 1.9183006535947716e-05,
|
13 |
-
"loss": 0.
|
14 |
"step": 500
|
15 |
},
|
16 |
{
|
17 |
"epoch": 1.0,
|
18 |
-
"eval_accuracy": 0.
|
19 |
-
"eval_f1": 0.
|
20 |
-
"eval_loss": 0.
|
21 |
-
"eval_precision": 0.
|
22 |
-
"eval_recall": 0.
|
23 |
-
"eval_runtime":
|
24 |
-
"eval_samples_per_second":
|
25 |
-
"eval_steps_per_second":
|
26 |
"step": 816
|
27 |
},
|
28 |
{
|
29 |
"epoch": 1.23,
|
30 |
"learning_rate": 1.8366013071895427e-05,
|
31 |
-
"loss": 0.
|
32 |
"step": 1000
|
33 |
},
|
34 |
{
|
35 |
"epoch": 1.84,
|
36 |
"learning_rate": 1.7549019607843138e-05,
|
37 |
-
"loss": 0.
|
38 |
"step": 1500
|
39 |
},
|
40 |
{
|
41 |
"epoch": 2.0,
|
42 |
-
"eval_accuracy": 0.
|
43 |
-
"eval_f1": 0.
|
44 |
-
"eval_loss": 0.
|
45 |
-
"eval_precision": 0.
|
46 |
-
"eval_recall": 0.
|
47 |
-
"eval_runtime":
|
48 |
-
"eval_samples_per_second":
|
49 |
-
"eval_steps_per_second":
|
50 |
"step": 1632
|
51 |
},
|
52 |
{
|
53 |
"epoch": 2.45,
|
54 |
"learning_rate": 1.6732026143790852e-05,
|
55 |
-
"loss": 0.
|
56 |
"step": 2000
|
57 |
},
|
58 |
{
|
59 |
"epoch": 3.0,
|
60 |
-
"eval_accuracy": 0.
|
61 |
-
"eval_f1": 0.
|
62 |
-
"eval_loss": 0.
|
63 |
-
"eval_precision": 0.
|
64 |
-
"eval_recall": 0.
|
65 |
-
"eval_runtime":
|
66 |
-
"eval_samples_per_second":
|
67 |
-
"eval_steps_per_second":
|
68 |
"step": 2448
|
69 |
},
|
70 |
{
|
71 |
"epoch": 3.06,
|
72 |
"learning_rate": 1.5915032679738563e-05,
|
73 |
-
"loss": 0.
|
74 |
"step": 2500
|
75 |
},
|
76 |
{
|
77 |
"epoch": 3.68,
|
78 |
"learning_rate": 1.5098039215686276e-05,
|
79 |
-
"loss": 0.
|
80 |
"step": 3000
|
81 |
},
|
82 |
{
|
83 |
"epoch": 4.0,
|
84 |
-
"eval_accuracy": 0.
|
85 |
-
"eval_f1": 0.
|
86 |
-
"eval_loss": 0.
|
87 |
-
"eval_precision": 0.
|
88 |
-
"eval_recall": 0.
|
89 |
-
"eval_runtime":
|
90 |
-
"eval_samples_per_second":
|
91 |
-
"eval_steps_per_second":
|
92 |
"step": 3264
|
93 |
},
|
94 |
{
|
95 |
"epoch": 4.29,
|
96 |
"learning_rate": 1.4281045751633989e-05,
|
97 |
-
"loss": 0.
|
98 |
"step": 3500
|
99 |
},
|
100 |
{
|
101 |
"epoch": 4.9,
|
102 |
"learning_rate": 1.3464052287581701e-05,
|
103 |
-
"loss": 0.
|
104 |
"step": 4000
|
105 |
},
|
106 |
{
|
107 |
"epoch": 5.0,
|
108 |
-
"eval_accuracy": 0.
|
109 |
-
"eval_f1": 0.
|
110 |
-
"eval_loss": 0.
|
111 |
-
"eval_precision": 0.
|
112 |
-
"eval_recall": 0.
|
113 |
-
"eval_runtime":
|
114 |
-
"eval_samples_per_second":
|
115 |
-
"eval_steps_per_second":
|
116 |
"step": 4080
|
117 |
},
|
118 |
{
|
119 |
"epoch": 5.51,
|
120 |
"learning_rate": 1.2647058823529412e-05,
|
121 |
-
"loss": 0.
|
122 |
"step": 4500
|
123 |
},
|
124 |
{
|
125 |
"epoch": 6.0,
|
126 |
-
"eval_accuracy": 0.
|
127 |
-
"eval_f1": 0.
|
128 |
-
"eval_loss": 0.
|
129 |
-
"eval_precision": 0.
|
130 |
-
"eval_recall": 0.
|
131 |
-
"eval_runtime": 3.
|
132 |
-
"eval_samples_per_second": 379.
|
133 |
-
"eval_steps_per_second": 23.
|
134 |
"step": 4896
|
135 |
},
|
136 |
{
|
137 |
"epoch": 6.13,
|
138 |
"learning_rate": 1.1830065359477125e-05,
|
139 |
-
"loss": 0.
|
140 |
"step": 5000
|
141 |
},
|
142 |
{
|
143 |
"epoch": 6.74,
|
144 |
"learning_rate": 1.1013071895424838e-05,
|
145 |
-
"loss": 0.
|
146 |
"step": 5500
|
147 |
},
|
148 |
{
|
149 |
"epoch": 7.0,
|
150 |
-
"eval_accuracy": 0.
|
151 |
-
"eval_f1": 0.
|
152 |
-
"eval_loss": 0.
|
153 |
-
"eval_precision": 0.
|
154 |
-
"eval_recall": 0.
|
155 |
-
"eval_runtime":
|
156 |
-
"eval_samples_per_second":
|
157 |
-
"eval_steps_per_second":
|
158 |
"step": 5712
|
159 |
},
|
160 |
{
|
161 |
"epoch": 7.35,
|
162 |
"learning_rate": 1.0196078431372549e-05,
|
163 |
-
"loss": 0.
|
164 |
"step": 6000
|
165 |
},
|
166 |
{
|
167 |
"epoch": 7.97,
|
168 |
"learning_rate": 9.379084967320261e-06,
|
169 |
-
"loss": 0.
|
170 |
"step": 6500
|
171 |
},
|
172 |
{
|
173 |
"epoch": 8.0,
|
174 |
-
"eval_accuracy": 0.
|
175 |
-
"eval_f1": 0.
|
176 |
-
"eval_loss": 0.
|
177 |
-
"eval_precision": 0.
|
178 |
-
"eval_recall": 0.
|
179 |
-
"eval_runtime":
|
180 |
-
"eval_samples_per_second":
|
181 |
-
"eval_steps_per_second":
|
182 |
"step": 6528
|
183 |
},
|
184 |
{
|
185 |
"epoch": 8.58,
|
186 |
"learning_rate": 8.562091503267974e-06,
|
187 |
-
"loss": 0.
|
188 |
"step": 7000
|
189 |
},
|
190 |
{
|
191 |
"epoch": 9.0,
|
192 |
-
"eval_accuracy": 0.
|
193 |
-
"eval_f1": 0.
|
194 |
-
"eval_loss": 0.
|
195 |
-
"eval_precision": 0.
|
196 |
-
"eval_recall": 0.
|
197 |
-
"eval_runtime":
|
198 |
-
"eval_samples_per_second":
|
199 |
-
"eval_steps_per_second":
|
200 |
"step": 7344
|
201 |
},
|
202 |
{
|
203 |
"epoch": 9.19,
|
204 |
"learning_rate": 7.745098039215687e-06,
|
205 |
-
"loss": 0.
|
206 |
"step": 7500
|
207 |
},
|
208 |
{
|
209 |
"epoch": 9.8,
|
210 |
"learning_rate": 6.928104575163399e-06,
|
211 |
-
"loss": 0.
|
212 |
"step": 8000
|
213 |
},
|
214 |
{
|
215 |
"epoch": 10.0,
|
216 |
-
"eval_accuracy": 0.
|
217 |
-
"eval_f1": 0.
|
218 |
-
"eval_loss": 0.
|
219 |
-
"eval_precision": 0.
|
220 |
-
"eval_recall": 0.
|
221 |
-
"eval_runtime":
|
222 |
-
"eval_samples_per_second":
|
223 |
-
"eval_steps_per_second":
|
224 |
"step": 8160
|
225 |
},
|
226 |
{
|
227 |
"epoch": 10.42,
|
228 |
"learning_rate": 6.111111111111112e-06,
|
229 |
-
"loss": 0.
|
230 |
"step": 8500
|
231 |
},
|
232 |
{
|
233 |
"epoch": 11.0,
|
234 |
-
"eval_accuracy": 0.
|
235 |
-
"eval_f1": 0.
|
236 |
-
"eval_loss": 0.
|
237 |
-
"eval_precision": 0.
|
238 |
-
"eval_recall": 0.
|
239 |
-
"eval_runtime": 2.
|
240 |
-
"eval_samples_per_second":
|
241 |
-
"eval_steps_per_second":
|
242 |
"step": 8976
|
243 |
},
|
244 |
{
|
245 |
"epoch": 11.03,
|
246 |
"learning_rate": 5.294117647058824e-06,
|
247 |
-
"loss": 0.
|
248 |
"step": 9000
|
249 |
},
|
250 |
{
|
251 |
"epoch": 11.64,
|
252 |
"learning_rate": 4.477124183006537e-06,
|
253 |
-
"loss": 0.
|
254 |
"step": 9500
|
255 |
},
|
256 |
{
|
257 |
"epoch": 12.0,
|
258 |
-
"eval_accuracy": 0.
|
259 |
-
"eval_f1": 0.
|
260 |
-
"eval_loss": 0.
|
261 |
-
"eval_precision": 0.
|
262 |
-
"eval_recall": 0.
|
263 |
-
"eval_runtime":
|
264 |
-
"eval_samples_per_second":
|
265 |
-
"eval_steps_per_second":
|
266 |
"step": 9792
|
267 |
},
|
268 |
{
|
269 |
"epoch": 12.25,
|
270 |
"learning_rate": 3.6601307189542484e-06,
|
271 |
-
"loss": 0.
|
272 |
"step": 10000
|
273 |
},
|
274 |
{
|
275 |
"epoch": 12.87,
|
276 |
"learning_rate": 2.843137254901961e-06,
|
277 |
-
"loss": 0.
|
278 |
"step": 10500
|
279 |
},
|
280 |
{
|
281 |
"epoch": 13.0,
|
282 |
-
"eval_accuracy": 0.
|
283 |
-
"eval_f1": 0.
|
284 |
-
"eval_loss": 0.
|
285 |
-
"eval_precision": 0.
|
286 |
-
"eval_recall": 0.
|
287 |
-
"eval_runtime": 3.
|
288 |
-
"eval_samples_per_second":
|
289 |
-
"eval_steps_per_second": 23.
|
290 |
"step": 10608
|
291 |
},
|
292 |
{
|
293 |
"epoch": 13.48,
|
294 |
"learning_rate": 2.0261437908496734e-06,
|
295 |
-
"loss": 0.
|
296 |
"step": 11000
|
297 |
},
|
298 |
{
|
299 |
"epoch": 14.0,
|
300 |
-
"eval_accuracy": 0.
|
301 |
-
"eval_f1": 0.
|
302 |
-
"eval_loss": 0.
|
303 |
-
"eval_precision": 0.
|
304 |
-
"eval_recall": 0.
|
305 |
-
"eval_runtime":
|
306 |
-
"eval_samples_per_second":
|
307 |
-
"eval_steps_per_second":
|
308 |
"step": 11424
|
309 |
},
|
310 |
{
|
311 |
"epoch": 14.09,
|
312 |
"learning_rate": 1.2091503267973858e-06,
|
313 |
-
"loss": 0.
|
314 |
"step": 11500
|
315 |
},
|
316 |
{
|
317 |
"epoch": 14.71,
|
318 |
"learning_rate": 3.921568627450981e-07,
|
319 |
-
"loss": 0.
|
320 |
"step": 12000
|
321 |
},
|
322 |
{
|
323 |
"epoch": 15.0,
|
324 |
-
"eval_accuracy": 0.
|
325 |
-
"eval_f1": 0.
|
326 |
-
"eval_loss": 0.
|
327 |
-
"eval_precision": 0.
|
328 |
-
"eval_recall": 0.
|
329 |
-
"eval_runtime": 3.
|
330 |
-
"eval_samples_per_second":
|
331 |
-
"eval_steps_per_second": 24.
|
332 |
"step": 12240
|
333 |
},
|
334 |
{
|
335 |
"epoch": 15.0,
|
336 |
"step": 12240,
|
337 |
"total_flos": 5.119363918698906e+16,
|
338 |
-
"train_loss": 0.
|
339 |
-
"train_runtime":
|
340 |
-
"train_samples_per_second":
|
341 |
-
"train_steps_per_second":
|
342 |
}
|
343 |
],
|
344 |
"max_steps": 12240,
|
|
|
10 |
{
|
11 |
"epoch": 0.61,
|
12 |
"learning_rate": 1.9183006535947716e-05,
|
13 |
+
"loss": 0.4329,
|
14 |
"step": 500
|
15 |
},
|
16 |
{
|
17 |
"epoch": 1.0,
|
18 |
+
"eval_accuracy": 0.9725273278019229,
|
19 |
+
"eval_f1": 0.9629807384654764,
|
20 |
+
"eval_loss": 0.10430943965911865,
|
21 |
+
"eval_precision": 0.9637516411028211,
|
22 |
+
"eval_recall": 0.9622110681287765,
|
23 |
+
"eval_runtime": 2.936,
|
24 |
+
"eval_samples_per_second": 494.214,
|
25 |
+
"eval_steps_per_second": 30.995,
|
26 |
"step": 816
|
27 |
},
|
28 |
{
|
29 |
"epoch": 1.23,
|
30 |
"learning_rate": 1.8366013071895427e-05,
|
31 |
+
"loss": 0.0811,
|
32 |
"step": 1000
|
33 |
},
|
34 |
{
|
35 |
"epoch": 1.84,
|
36 |
"learning_rate": 1.7549019607843138e-05,
|
37 |
+
"loss": 0.0607,
|
38 |
"step": 1500
|
39 |
},
|
40 |
{
|
41 |
"epoch": 2.0,
|
42 |
+
"eval_accuracy": 0.9751349927564862,
|
43 |
+
"eval_f1": 0.9635150914122107,
|
44 |
+
"eval_loss": 0.09400150179862976,
|
45 |
+
"eval_precision": 0.9641165172855314,
|
46 |
+
"eval_recall": 0.9629144154224879,
|
47 |
+
"eval_runtime": 2.7493,
|
48 |
+
"eval_samples_per_second": 527.777,
|
49 |
+
"eval_steps_per_second": 33.1,
|
50 |
"step": 1632
|
51 |
},
|
52 |
{
|
53 |
"epoch": 2.45,
|
54 |
"learning_rate": 1.6732026143790852e-05,
|
55 |
+
"loss": 0.0428,
|
56 |
"step": 2000
|
57 |
},
|
58 |
{
|
59 |
"epoch": 3.0,
|
60 |
+
"eval_accuracy": 0.9755037534571316,
|
61 |
+
"eval_f1": 0.9632957125043962,
|
62 |
+
"eval_loss": 0.09555820375680923,
|
63 |
+
"eval_precision": 0.9633573141486811,
|
64 |
+
"eval_recall": 0.9632341187378113,
|
65 |
+
"eval_runtime": 2.7415,
|
66 |
+
"eval_samples_per_second": 529.269,
|
67 |
+
"eval_steps_per_second": 33.193,
|
68 |
"step": 2448
|
69 |
},
|
70 |
{
|
71 |
"epoch": 3.06,
|
72 |
"learning_rate": 1.5915032679738563e-05,
|
73 |
+
"loss": 0.0363,
|
74 |
"step": 2500
|
75 |
},
|
76 |
{
|
77 |
"epoch": 3.68,
|
78 |
"learning_rate": 1.5098039215686276e-05,
|
79 |
+
"loss": 0.0249,
|
80 |
"step": 3000
|
81 |
},
|
82 |
{
|
83 |
"epoch": 4.0,
|
84 |
+
"eval_accuracy": 0.9765046753588832,
|
85 |
+
"eval_f1": 0.9649540551338394,
|
86 |
+
"eval_loss": 0.0989251509308815,
|
87 |
+
"eval_precision": 0.9646919734151329,
|
88 |
+
"eval_recall": 0.9652162792928163,
|
89 |
+
"eval_runtime": 2.7759,
|
90 |
+
"eval_samples_per_second": 522.708,
|
91 |
+
"eval_steps_per_second": 32.782,
|
92 |
"step": 3264
|
93 |
},
|
94 |
{
|
95 |
"epoch": 4.29,
|
96 |
"learning_rate": 1.4281045751633989e-05,
|
97 |
+
"loss": 0.0221,
|
98 |
"step": 3500
|
99 |
},
|
100 |
{
|
101 |
"epoch": 4.9,
|
102 |
"learning_rate": 1.3464052287581701e-05,
|
103 |
+
"loss": 0.0177,
|
104 |
"step": 4000
|
105 |
},
|
106 |
{
|
107 |
"epoch": 5.0,
|
108 |
+
"eval_accuracy": 0.976451995258791,
|
109 |
+
"eval_f1": 0.9646675296035285,
|
110 |
+
"eval_loss": 0.10913769155740738,
|
111 |
+
"eval_precision": 0.9643747204294204,
|
112 |
+
"eval_recall": 0.9649605166405576,
|
113 |
+
"eval_runtime": 3.7551,
|
114 |
+
"eval_samples_per_second": 386.403,
|
115 |
+
"eval_steps_per_second": 24.233,
|
116 |
"step": 4080
|
117 |
},
|
118 |
{
|
119 |
"epoch": 5.51,
|
120 |
"learning_rate": 1.2647058823529412e-05,
|
121 |
+
"loss": 0.0129,
|
122 |
"step": 4500
|
123 |
},
|
124 |
{
|
125 |
"epoch": 6.0,
|
126 |
+
"eval_accuracy": 0.9780060582115107,
|
127 |
+
"eval_f1": 0.9675017575254042,
|
128 |
+
"eval_loss": 0.11124212294816971,
|
129 |
+
"eval_precision": 0.967038231818327,
|
130 |
+
"eval_recall": 0.9679657278045973,
|
131 |
+
"eval_runtime": 3.8249,
|
132 |
+
"eval_samples_per_second": 379.357,
|
133 |
+
"eval_steps_per_second": 23.792,
|
134 |
"step": 4896
|
135 |
},
|
136 |
{
|
137 |
"epoch": 6.13,
|
138 |
"learning_rate": 1.1830065359477125e-05,
|
139 |
+
"loss": 0.0119,
|
140 |
"step": 5000
|
141 |
},
|
142 |
{
|
143 |
"epoch": 6.74,
|
144 |
"learning_rate": 1.1013071895424838e-05,
|
145 |
+
"loss": 0.0084,
|
146 |
"step": 5500
|
147 |
},
|
148 |
{
|
149 |
"epoch": 7.0,
|
150 |
+
"eval_accuracy": 0.9783221388120638,
|
151 |
+
"eval_f1": 0.9677996994212259,
|
152 |
+
"eval_loss": 0.12139276415109634,
|
153 |
+
"eval_precision": 0.967985415933732,
|
154 |
+
"eval_recall": 0.9676140541577416,
|
155 |
+
"eval_runtime": 3.747,
|
156 |
+
"eval_samples_per_second": 387.246,
|
157 |
+
"eval_steps_per_second": 24.286,
|
158 |
"step": 5712
|
159 |
},
|
160 |
{
|
161 |
"epoch": 7.35,
|
162 |
"learning_rate": 1.0196078431372549e-05,
|
163 |
+
"loss": 0.0076,
|
164 |
"step": 6000
|
165 |
},
|
166 |
{
|
167 |
"epoch": 7.97,
|
168 |
"learning_rate": 9.379084967320261e-06,
|
169 |
+
"loss": 0.0061,
|
170 |
"step": 6500
|
171 |
},
|
172 |
{
|
173 |
"epoch": 8.0,
|
174 |
+
"eval_accuracy": 0.9775319373106809,
|
175 |
+
"eval_f1": 0.9664444551020734,
|
176 |
+
"eval_loss": 0.13343702256679535,
|
177 |
+
"eval_precision": 0.9665217113257019,
|
178 |
+
"eval_recall": 0.9663672112279804,
|
179 |
+
"eval_runtime": 2.7649,
|
180 |
+
"eval_samples_per_second": 524.787,
|
181 |
+
"eval_steps_per_second": 32.912,
|
182 |
"step": 6528
|
183 |
},
|
184 |
{
|
185 |
"epoch": 8.58,
|
186 |
"learning_rate": 8.562091503267974e-06,
|
187 |
+
"loss": 0.0048,
|
188 |
"step": 7000
|
189 |
},
|
190 |
{
|
191 |
"epoch": 9.0,
|
192 |
+
"eval_accuracy": 0.9775055972606348,
|
193 |
+
"eval_f1": 0.9661577811525858,
|
194 |
+
"eval_loss": 0.13892702758312225,
|
195 |
+
"eval_precision": 0.9662041181736795,
|
196 |
+
"eval_recall": 0.9661114485757217,
|
197 |
+
"eval_runtime": 3.8371,
|
198 |
+
"eval_samples_per_second": 378.146,
|
199 |
+
"eval_steps_per_second": 23.716,
|
200 |
"step": 7344
|
201 |
},
|
202 |
{
|
203 |
"epoch": 9.19,
|
204 |
"learning_rate": 7.745098039215687e-06,
|
205 |
+
"loss": 0.0047,
|
206 |
"step": 7500
|
207 |
},
|
208 |
{
|
209 |
"epoch": 9.8,
|
210 |
"learning_rate": 6.928104575163399e-06,
|
211 |
+
"loss": 0.0044,
|
212 |
"step": 8000
|
213 |
},
|
214 |
{
|
215 |
"epoch": 10.0,
|
216 |
+
"eval_accuracy": 0.9781640985117872,
|
217 |
+
"eval_f1": 0.967271564473579,
|
218 |
+
"eval_loss": 0.14489226043224335,
|
219 |
+
"eval_precision": 0.9674726540011515,
|
220 |
+
"eval_recall": 0.9670705585216919,
|
221 |
+
"eval_runtime": 3.7406,
|
222 |
+
"eval_samples_per_second": 387.909,
|
223 |
+
"eval_steps_per_second": 24.328,
|
224 |
"step": 8160
|
225 |
},
|
226 |
{
|
227 |
"epoch": 10.42,
|
228 |
"learning_rate": 6.111111111111112e-06,
|
229 |
+
"loss": 0.0034,
|
230 |
"step": 8500
|
231 |
},
|
232 |
{
|
233 |
"epoch": 11.0,
|
234 |
+
"eval_accuracy": 0.9779797181614646,
|
235 |
+
"eval_f1": 0.9672149490880608,
|
236 |
+
"eval_loss": 0.14611582458019257,
|
237 |
+
"eval_precision": 0.9671994884910486,
|
238 |
+
"eval_recall": 0.9672304101793535,
|
239 |
+
"eval_runtime": 2.8969,
|
240 |
+
"eval_samples_per_second": 500.886,
|
241 |
+
"eval_steps_per_second": 31.413,
|
242 |
"step": 8976
|
243 |
},
|
244 |
{
|
245 |
"epoch": 11.03,
|
246 |
"learning_rate": 5.294117647058824e-06,
|
247 |
+
"loss": 0.0034,
|
248 |
"step": 9000
|
249 |
},
|
250 |
{
|
251 |
"epoch": 11.64,
|
252 |
"learning_rate": 4.477124183006537e-06,
|
253 |
+
"loss": 0.0025,
|
254 |
"step": 9500
|
255 |
},
|
256 |
{
|
257 |
"epoch": 12.0,
|
258 |
+
"eval_accuracy": 0.9784274990122481,
|
259 |
+
"eval_f1": 0.9676501159350764,
|
260 |
+
"eval_loss": 0.14966140687465668,
|
261 |
+
"eval_precision": 0.9680061428205784,
|
262 |
+
"eval_recall": 0.9672943508424182,
|
263 |
+
"eval_runtime": 2.7972,
|
264 |
+
"eval_samples_per_second": 518.731,
|
265 |
+
"eval_steps_per_second": 32.532,
|
266 |
"step": 9792
|
267 |
},
|
268 |
{
|
269 |
"epoch": 12.25,
|
270 |
"learning_rate": 3.6601307189542484e-06,
|
271 |
+
"loss": 0.0022,
|
272 |
"step": 10000
|
273 |
},
|
274 |
{
|
275 |
"epoch": 12.87,
|
276 |
"learning_rate": 2.843137254901961e-06,
|
277 |
+
"loss": 0.0024,
|
278 |
"step": 10500
|
279 |
},
|
280 |
{
|
281 |
"epoch": 13.0,
|
282 |
+
"eval_accuracy": 0.977848017911234,
|
283 |
+
"eval_f1": 0.9670199670679264,
|
284 |
+
"eval_loss": 0.15780866146087646,
|
285 |
+
"eval_precision": 0.9670972692971798,
|
286 |
+
"eval_recall": 0.9669426771955625,
|
287 |
+
"eval_runtime": 3.8111,
|
288 |
+
"eval_samples_per_second": 380.734,
|
289 |
+
"eval_steps_per_second": 23.878,
|
290 |
"step": 10608
|
291 |
},
|
292 |
{
|
293 |
"epoch": 13.48,
|
294 |
"learning_rate": 2.0261437908496734e-06,
|
295 |
+
"loss": 0.0014,
|
296 |
"step": 11000
|
297 |
},
|
298 |
{
|
299 |
"epoch": 14.0,
|
300 |
+
"eval_accuracy": 0.978111418411695,
|
301 |
+
"eval_f1": 0.9674521213671389,
|
302 |
+
"eval_loss": 0.15866732597351074,
|
303 |
+
"eval_precision": 0.9675139888089528,
|
304 |
+
"eval_recall": 0.9673902618370153,
|
305 |
+
"eval_runtime": 2.8382,
|
306 |
+
"eval_samples_per_second": 511.243,
|
307 |
+
"eval_steps_per_second": 32.063,
|
308 |
"step": 11424
|
309 |
},
|
310 |
{
|
311 |
"epoch": 14.09,
|
312 |
"learning_rate": 1.2091503267973858e-06,
|
313 |
+
"loss": 0.0019,
|
314 |
"step": 11500
|
315 |
},
|
316 |
{
|
317 |
"epoch": 14.71,
|
318 |
"learning_rate": 3.921568627450981e-07,
|
319 |
+
"loss": 0.0016,
|
320 |
"step": 12000
|
321 |
},
|
322 |
{
|
323 |
"epoch": 15.0,
|
324 |
+
"eval_accuracy": 0.978401158962202,
|
325 |
+
"eval_f1": 0.96785434549785,
|
326 |
+
"eval_loss": 0.15796583890914917,
|
327 |
+
"eval_precision": 0.9678388746803069,
|
328 |
+
"eval_recall": 0.9678698168100003,
|
329 |
+
"eval_runtime": 3.7576,
|
330 |
+
"eval_samples_per_second": 386.151,
|
331 |
+
"eval_steps_per_second": 24.218,
|
332 |
"step": 12240
|
333 |
},
|
334 |
{
|
335 |
"epoch": 15.0,
|
336 |
"step": 12240,
|
337 |
"total_flos": 5.119363918698906e+16,
|
338 |
+
"train_loss": 0.03262665262993644,
|
339 |
+
"train_runtime": 785.1159,
|
340 |
+
"train_samples_per_second": 249.403,
|
341 |
+
"train_steps_per_second": 15.59
|
342 |
}
|
343 |
],
|
344 |
"max_steps": 12240,
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3119
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2ad2280933fd2dfde8ee13fc62c6b195937dddc4b2505fce56231e549d3c475
|
3 |
size 3119
|