Commit
•
5bfb80e
1
Parent(s):
96eeacb
End of training
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Nov28_09-28-45_43fe4c10e15a/events.out.tfevents.1701171037.43fe4c10e15a.986.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +286 -0
README.md
CHANGED
@@ -16,6 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
# debug_seq2seq_squad
|
17 |
|
18 |
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the squad_v2 dataset.
|
|
|
|
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
16 |
# debug_seq2seq_squad
|
17 |
|
18 |
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the squad_v2 dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 1.9764
|
21 |
|
22 |
## Model description
|
23 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_loss": 1.976380467414856,
|
4 |
+
"eval_runtime": 106.9772,
|
5 |
+
"eval_samples": 12267,
|
6 |
+
"eval_samples_per_second": 114.669,
|
7 |
+
"eval_steps_per_second": 14.34,
|
8 |
+
"train_loss": 1.9922608349204722,
|
9 |
+
"train_runtime": 7012.8751,
|
10 |
+
"train_samples": 130319,
|
11 |
+
"train_samples_per_second": 37.166,
|
12 |
+
"train_steps_per_second": 3.097
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_loss": 1.976380467414856,
|
4 |
+
"eval_runtime": 106.9772,
|
5 |
+
"eval_samples": 12267,
|
6 |
+
"eval_samples_per_second": 114.669,
|
7 |
+
"eval_steps_per_second": 14.34
|
8 |
+
}
|
runs/Nov28_09-28-45_43fe4c10e15a/events.out.tfevents.1701171037.43fe4c10e15a.986.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1097f27eaa6bbdf4dbc23871c78e38ff3ea4911c2026e93a0419e300e55a5e22
|
3 |
+
size 364
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"train_loss": 1.9922608349204722,
|
4 |
+
"train_runtime": 7012.8751,
|
5 |
+
"train_samples": 130319,
|
6 |
+
"train_samples_per_second": 37.166,
|
7 |
+
"train_steps_per_second": 3.097
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,286 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 21720,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.05,
|
13 |
+
"learning_rate": 2.930939226519337e-05,
|
14 |
+
"loss": 2.0443,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.09,
|
19 |
+
"learning_rate": 2.861878453038674e-05,
|
20 |
+
"loss": 2.0083,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.14,
|
25 |
+
"learning_rate": 2.7928176795580113e-05,
|
26 |
+
"loss": 2.0213,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.18,
|
31 |
+
"learning_rate": 2.7237569060773482e-05,
|
32 |
+
"loss": 2.0025,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.23,
|
37 |
+
"learning_rate": 2.6546961325966852e-05,
|
38 |
+
"loss": 1.976,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.28,
|
43 |
+
"learning_rate": 2.585635359116022e-05,
|
44 |
+
"loss": 2.0151,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.32,
|
49 |
+
"learning_rate": 2.516574585635359e-05,
|
50 |
+
"loss": 1.9824,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.37,
|
55 |
+
"learning_rate": 2.4475138121546964e-05,
|
56 |
+
"loss": 1.9815,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.41,
|
61 |
+
"learning_rate": 2.3784530386740334e-05,
|
62 |
+
"loss": 1.988,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.46,
|
67 |
+
"learning_rate": 2.3093922651933703e-05,
|
68 |
+
"loss": 1.9769,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.51,
|
73 |
+
"learning_rate": 2.2403314917127073e-05,
|
74 |
+
"loss": 1.9668,
|
75 |
+
"step": 5500
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.55,
|
79 |
+
"learning_rate": 2.1712707182320442e-05,
|
80 |
+
"loss": 1.9765,
|
81 |
+
"step": 6000
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.6,
|
85 |
+
"learning_rate": 2.1022099447513815e-05,
|
86 |
+
"loss": 2.0408,
|
87 |
+
"step": 6500
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.64,
|
91 |
+
"learning_rate": 2.0331491712707185e-05,
|
92 |
+
"loss": 2.052,
|
93 |
+
"step": 7000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.69,
|
97 |
+
"learning_rate": 1.9640883977900554e-05,
|
98 |
+
"loss": 2.0649,
|
99 |
+
"step": 7500
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.74,
|
103 |
+
"learning_rate": 1.8950276243093924e-05,
|
104 |
+
"loss": 2.0353,
|
105 |
+
"step": 8000
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.78,
|
109 |
+
"learning_rate": 1.825966850828729e-05,
|
110 |
+
"loss": 2.0307,
|
111 |
+
"step": 8500
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.83,
|
115 |
+
"learning_rate": 1.7569060773480663e-05,
|
116 |
+
"loss": 2.0445,
|
117 |
+
"step": 9000
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.87,
|
121 |
+
"learning_rate": 1.6878453038674033e-05,
|
122 |
+
"loss": 2.0231,
|
123 |
+
"step": 9500
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.92,
|
127 |
+
"learning_rate": 1.6187845303867402e-05,
|
128 |
+
"loss": 2.0082,
|
129 |
+
"step": 10000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.97,
|
133 |
+
"learning_rate": 1.5497237569060772e-05,
|
134 |
+
"loss": 2.0162,
|
135 |
+
"step": 10500
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 1.01,
|
139 |
+
"learning_rate": 1.4806629834254145e-05,
|
140 |
+
"loss": 2.0191,
|
141 |
+
"step": 11000
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 1.06,
|
145 |
+
"learning_rate": 1.4116022099447514e-05,
|
146 |
+
"loss": 1.9899,
|
147 |
+
"step": 11500
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 1.1,
|
151 |
+
"learning_rate": 1.3425414364640886e-05,
|
152 |
+
"loss": 2.0006,
|
153 |
+
"step": 12000
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 1.15,
|
157 |
+
"learning_rate": 1.2734806629834255e-05,
|
158 |
+
"loss": 1.9936,
|
159 |
+
"step": 12500
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 1.2,
|
163 |
+
"learning_rate": 1.2044198895027623e-05,
|
164 |
+
"loss": 1.9787,
|
165 |
+
"step": 13000
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 1.24,
|
169 |
+
"learning_rate": 1.1353591160220994e-05,
|
170 |
+
"loss": 1.9711,
|
171 |
+
"step": 13500
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 1.29,
|
175 |
+
"learning_rate": 1.0662983425414364e-05,
|
176 |
+
"loss": 1.953,
|
177 |
+
"step": 14000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 1.34,
|
181 |
+
"learning_rate": 9.972375690607735e-06,
|
182 |
+
"loss": 1.9803,
|
183 |
+
"step": 14500
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 1.38,
|
187 |
+
"learning_rate": 9.281767955801105e-06,
|
188 |
+
"loss": 1.9653,
|
189 |
+
"step": 15000
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 1.43,
|
193 |
+
"learning_rate": 8.591160220994474e-06,
|
194 |
+
"loss": 1.9643,
|
195 |
+
"step": 15500
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 1.47,
|
199 |
+
"learning_rate": 7.900552486187846e-06,
|
200 |
+
"loss": 1.9663,
|
201 |
+
"step": 16000
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 1.52,
|
205 |
+
"learning_rate": 7.209944751381215e-06,
|
206 |
+
"loss": 1.9701,
|
207 |
+
"step": 16500
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 1.57,
|
211 |
+
"learning_rate": 6.519337016574586e-06,
|
212 |
+
"loss": 1.9657,
|
213 |
+
"step": 17000
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 1.61,
|
217 |
+
"learning_rate": 5.828729281767956e-06,
|
218 |
+
"loss": 1.9622,
|
219 |
+
"step": 17500
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.66,
|
223 |
+
"learning_rate": 5.1381215469613265e-06,
|
224 |
+
"loss": 1.9683,
|
225 |
+
"step": 18000
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 1.7,
|
229 |
+
"learning_rate": 4.447513812154697e-06,
|
230 |
+
"loss": 1.9654,
|
231 |
+
"step": 18500
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.75,
|
235 |
+
"learning_rate": 3.756906077348067e-06,
|
236 |
+
"loss": 1.9609,
|
237 |
+
"step": 19000
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.8,
|
241 |
+
"learning_rate": 3.0662983425414365e-06,
|
242 |
+
"loss": 1.9639,
|
243 |
+
"step": 19500
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.84,
|
247 |
+
"learning_rate": 2.375690607734807e-06,
|
248 |
+
"loss": 1.9848,
|
249 |
+
"step": 20000
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.89,
|
253 |
+
"learning_rate": 1.6850828729281769e-06,
|
254 |
+
"loss": 1.975,
|
255 |
+
"step": 20500
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.93,
|
259 |
+
"learning_rate": 9.944751381215469e-07,
|
260 |
+
"loss": 1.9509,
|
261 |
+
"step": 21000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 1.98,
|
265 |
+
"learning_rate": 3.0386740331491715e-07,
|
266 |
+
"loss": 1.9651,
|
267 |
+
"step": 21500
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 2.0,
|
271 |
+
"step": 21720,
|
272 |
+
"total_flos": 2.645641233904435e+16,
|
273 |
+
"train_loss": 1.9922608349204722,
|
274 |
+
"train_runtime": 7012.8751,
|
275 |
+
"train_samples_per_second": 37.166,
|
276 |
+
"train_steps_per_second": 3.097
|
277 |
+
}
|
278 |
+
],
|
279 |
+
"logging_steps": 500,
|
280 |
+
"max_steps": 21720,
|
281 |
+
"num_train_epochs": 2,
|
282 |
+
"save_steps": 500,
|
283 |
+
"total_flos": 2.645641233904435e+16,
|
284 |
+
"trial_name": null,
|
285 |
+
"trial_params": null
|
286 |
+
}
|