model upload
Browse files- added_tokens.json +1 -0
- config.json +36 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +325 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"[S]": 50265, "[E]": 50266}
|
config.json
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/hdd1/mujeen/retrieval_prf/roberta-large-nq",
|
3 |
+
"architectures": [
|
4 |
+
"RobertaForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 1024,
|
13 |
+
"id2label": {
|
14 |
+
"0": 0,
|
15 |
+
"1": 1
|
16 |
+
},
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 4096,
|
19 |
+
"label2id": {
|
20 |
+
"0": 0,
|
21 |
+
"1": 1
|
22 |
+
},
|
23 |
+
"layer_norm_eps": 1e-05,
|
24 |
+
"max_position_embeddings": 514,
|
25 |
+
"model_type": "roberta",
|
26 |
+
"num_attention_heads": 16,
|
27 |
+
"num_hidden_layers": 24,
|
28 |
+
"pad_token_id": 1,
|
29 |
+
"position_embedding_type": "absolute",
|
30 |
+
"problem_type": "single_label_classification",
|
31 |
+
"torch_dtype": "float32",
|
32 |
+
"transformers_version": "4.13.0",
|
33 |
+
"type_vocab_size": 1,
|
34 |
+
"use_cache": true,
|
35 |
+
"vocab_size": 50267
|
36 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:57fc94a142fbe37a386693778c41eebcf07326cd4099a8deeb0ed498f48fa55e
|
3 |
+
size 1421619501
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "do_lower_case": false, "model_max_length": 512, "special_tokens_map_file": "/hdd1/mujeen/retrieval_prf/roberta-large-nq/special_tokens_map.json", "name_or_path": "/hdd1/mujeen/retrieval_prf/roberta-large-nq", "tokenizer_class": "RobertaTokenizer"}
|
trainer_state.json
ADDED
@@ -0,0 +1,325 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8369131635471003,
|
3 |
+
"best_model_checkpoint": "/hdd1/mujeen/retrieval_prf/output/labeler_nq.train.v7.0_nq.dev.v7.0_rlnq_title/checkpoint-76000",
|
4 |
+
"epoch": 0.9999910574558462,
|
5 |
+
"global_step": 83868,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.05,
|
12 |
+
"learning_rate": 1.904755091334001e-05,
|
13 |
+
"loss": 0.5261,
|
14 |
+
"step": 4000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"eval_accuracy": 0.7910401963792575,
|
19 |
+
"eval_loss": 0.4569447934627533,
|
20 |
+
"eval_runtime": 61.7181,
|
21 |
+
"eval_samples_per_second": 211.219,
|
22 |
+
"eval_steps_per_second": 26.41,
|
23 |
+
"step": 4000
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.1,
|
27 |
+
"learning_rate": 1.8093671006820245e-05,
|
28 |
+
"loss": 0.4674,
|
29 |
+
"step": 8000
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.1,
|
33 |
+
"eval_accuracy": 0.7984811291807303,
|
34 |
+
"eval_loss": 0.42789292335510254,
|
35 |
+
"eval_runtime": 62.0897,
|
36 |
+
"eval_samples_per_second": 209.954,
|
37 |
+
"eval_steps_per_second": 26.252,
|
38 |
+
"step": 8000
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.14,
|
42 |
+
"learning_rate": 1.7140744980206995e-05,
|
43 |
+
"loss": 0.4453,
|
44 |
+
"step": 12000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.14,
|
48 |
+
"eval_accuracy": 0.804311138386008,
|
49 |
+
"eval_loss": 0.4111127257347107,
|
50 |
+
"eval_runtime": 73.971,
|
51 |
+
"eval_samples_per_second": 176.231,
|
52 |
+
"eval_steps_per_second": 22.036,
|
53 |
+
"step": 12000
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 0.19,
|
57 |
+
"learning_rate": 1.6186865073687225e-05,
|
58 |
+
"loss": 0.4247,
|
59 |
+
"step": 16000
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.19,
|
63 |
+
"eval_accuracy": 0.8105247008284749,
|
64 |
+
"eval_loss": 0.4133109748363495,
|
65 |
+
"eval_runtime": 67.9092,
|
66 |
+
"eval_samples_per_second": 191.962,
|
67 |
+
"eval_steps_per_second": 24.003,
|
68 |
+
"step": 16000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.24,
|
72 |
+
"learning_rate": 1.5233462107120716e-05,
|
73 |
+
"loss": 0.4097,
|
74 |
+
"step": 20000
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.24,
|
78 |
+
"eval_accuracy": 0.8149739183798711,
|
79 |
+
"eval_loss": 0.411438524723053,
|
80 |
+
"eval_runtime": 76.7711,
|
81 |
+
"eval_samples_per_second": 169.803,
|
82 |
+
"eval_steps_per_second": 21.232,
|
83 |
+
"step": 20000
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.29,
|
87 |
+
"learning_rate": 1.4280059140554206e-05,
|
88 |
+
"loss": 0.3915,
|
89 |
+
"step": 24000
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.29,
|
93 |
+
"eval_accuracy": 0.8158944461491255,
|
94 |
+
"eval_loss": 0.4367925524711609,
|
95 |
+
"eval_runtime": 73.8165,
|
96 |
+
"eval_samples_per_second": 176.6,
|
97 |
+
"eval_steps_per_second": 22.082,
|
98 |
+
"step": 24000
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.33,
|
102 |
+
"learning_rate": 1.3326179234034435e-05,
|
103 |
+
"loss": 0.3783,
|
104 |
+
"step": 28000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.33,
|
108 |
+
"eval_accuracy": 0.8168916845658177,
|
109 |
+
"eval_loss": 0.43372228741645813,
|
110 |
+
"eval_runtime": 74.7748,
|
111 |
+
"eval_samples_per_second": 174.337,
|
112 |
+
"eval_steps_per_second": 21.799,
|
113 |
+
"step": 28000
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.38,
|
117 |
+
"learning_rate": 1.2373014737444558e-05,
|
118 |
+
"loss": 0.365,
|
119 |
+
"step": 32000
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.38,
|
123 |
+
"eval_accuracy": 0.8242559067198527,
|
124 |
+
"eval_loss": 0.40461644530296326,
|
125 |
+
"eval_runtime": 77.9179,
|
126 |
+
"eval_samples_per_second": 167.304,
|
127 |
+
"eval_steps_per_second": 20.919,
|
128 |
+
"step": 32000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.43,
|
132 |
+
"learning_rate": 1.1419373300901418e-05,
|
133 |
+
"loss": 0.3477,
|
134 |
+
"step": 36000
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.43,
|
138 |
+
"eval_accuracy": 0.828628413623811,
|
139 |
+
"eval_loss": 0.4262824058532715,
|
140 |
+
"eval_runtime": 74.7614,
|
141 |
+
"eval_samples_per_second": 174.368,
|
142 |
+
"eval_steps_per_second": 21.803,
|
143 |
+
"step": 36000
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 0.48,
|
147 |
+
"learning_rate": 1.0465731864358278e-05,
|
148 |
+
"loss": 0.3341,
|
149 |
+
"step": 40000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.48,
|
153 |
+
"eval_accuracy": 0.8258668303160479,
|
154 |
+
"eval_loss": 0.40725135803222656,
|
155 |
+
"eval_runtime": 77.1145,
|
156 |
+
"eval_samples_per_second": 169.047,
|
157 |
+
"eval_steps_per_second": 21.137,
|
158 |
+
"step": 40000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.52,
|
162 |
+
"learning_rate": 9.51209042781514e-06,
|
163 |
+
"loss": 0.3214,
|
164 |
+
"step": 44000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.52,
|
168 |
+
"eval_accuracy": 0.8284749923289353,
|
169 |
+
"eval_loss": 0.4358045756816864,
|
170 |
+
"eval_runtime": 79.2128,
|
171 |
+
"eval_samples_per_second": 164.569,
|
172 |
+
"eval_steps_per_second": 20.577,
|
173 |
+
"step": 44000
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 0.57,
|
177 |
+
"learning_rate": 8.55868746124863e-06,
|
178 |
+
"loss": 0.3112,
|
179 |
+
"step": 48000
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.57,
|
183 |
+
"eval_accuracy": 0.8285517029763732,
|
184 |
+
"eval_loss": 0.4215088188648224,
|
185 |
+
"eval_runtime": 76.3765,
|
186 |
+
"eval_samples_per_second": 170.681,
|
187 |
+
"eval_steps_per_second": 21.342,
|
188 |
+
"step": 48000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.62,
|
192 |
+
"learning_rate": 7.6050460247054905e-06,
|
193 |
+
"loss": 0.2996,
|
194 |
+
"step": 52000
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.62,
|
198 |
+
"eval_accuracy": 0.8277845965019944,
|
199 |
+
"eval_loss": 0.4198075234889984,
|
200 |
+
"eval_runtime": 75.7616,
|
201 |
+
"eval_samples_per_second": 172.066,
|
202 |
+
"eval_steps_per_second": 21.515,
|
203 |
+
"step": 52000
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 0.67,
|
207 |
+
"learning_rate": 6.651643058138981e-06,
|
208 |
+
"loss": 0.2863,
|
209 |
+
"step": 56000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.67,
|
213 |
+
"eval_accuracy": 0.8310831543418227,
|
214 |
+
"eval_loss": 0.4417212903499603,
|
215 |
+
"eval_runtime": 80.0618,
|
216 |
+
"eval_samples_per_second": 162.824,
|
217 |
+
"eval_steps_per_second": 20.359,
|
218 |
+
"step": 56000
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.72,
|
222 |
+
"learning_rate": 5.698240091572471e-06,
|
223 |
+
"loss": 0.2739,
|
224 |
+
"step": 60000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.72,
|
228 |
+
"eval_accuracy": 0.8222614298864682,
|
229 |
+
"eval_loss": 0.5145458579063416,
|
230 |
+
"eval_runtime": 78.9172,
|
231 |
+
"eval_samples_per_second": 165.186,
|
232 |
+
"eval_steps_per_second": 20.655,
|
233 |
+
"step": 60000
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.76,
|
237 |
+
"learning_rate": 4.744598655029332e-06,
|
238 |
+
"loss": 0.2656,
|
239 |
+
"step": 64000
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.76,
|
243 |
+
"eval_accuracy": 0.8265572261429887,
|
244 |
+
"eval_loss": 0.5052666068077087,
|
245 |
+
"eval_runtime": 77.2292,
|
246 |
+
"eval_samples_per_second": 168.796,
|
247 |
+
"eval_steps_per_second": 21.106,
|
248 |
+
"step": 64000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.81,
|
252 |
+
"learning_rate": 3.7911956884628224e-06,
|
253 |
+
"loss": 0.2558,
|
254 |
+
"step": 68000
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.81,
|
258 |
+
"eval_accuracy": 0.8353022399509051,
|
259 |
+
"eval_loss": 0.47119611501693726,
|
260 |
+
"eval_runtime": 79.1339,
|
261 |
+
"eval_samples_per_second": 164.733,
|
262 |
+
"eval_steps_per_second": 20.598,
|
263 |
+
"step": 68000
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.86,
|
267 |
+
"learning_rate": 2.8377927218963137e-06,
|
268 |
+
"loss": 0.2467,
|
269 |
+
"step": 72000
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.86,
|
273 |
+
"eval_accuracy": 0.8312365756366984,
|
274 |
+
"eval_loss": 0.5034319162368774,
|
275 |
+
"eval_runtime": 82.6755,
|
276 |
+
"eval_samples_per_second": 157.677,
|
277 |
+
"eval_steps_per_second": 19.716,
|
278 |
+
"step": 72000
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.91,
|
282 |
+
"learning_rate": 1.8839128153765443e-06,
|
283 |
+
"loss": 0.2342,
|
284 |
+
"step": 76000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.91,
|
288 |
+
"eval_accuracy": 0.8369131635471003,
|
289 |
+
"eval_loss": 0.4584212303161621,
|
290 |
+
"eval_runtime": 81.5904,
|
291 |
+
"eval_samples_per_second": 159.774,
|
292 |
+
"eval_steps_per_second": 19.978,
|
293 |
+
"step": 76000
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 0.95,
|
297 |
+
"learning_rate": 9.305098488100349e-07,
|
298 |
+
"loss": 0.23,
|
299 |
+
"step": 80000
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 0.95,
|
303 |
+
"eval_accuracy": 0.8341515802393372,
|
304 |
+
"eval_loss": 0.49471235275268555,
|
305 |
+
"eval_runtime": 87.9548,
|
306 |
+
"eval_samples_per_second": 148.213,
|
307 |
+
"eval_steps_per_second": 18.532,
|
308 |
+
"step": 80000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 1.0,
|
312 |
+
"step": 83868,
|
313 |
+
"total_flos": 7.653345366712497e+17,
|
314 |
+
"train_loss": 0.3355050985067308,
|
315 |
+
"train_runtime": 56077.3175,
|
316 |
+
"train_samples_per_second": 47.859,
|
317 |
+
"train_steps_per_second": 1.496
|
318 |
+
}
|
319 |
+
],
|
320 |
+
"max_steps": 83868,
|
321 |
+
"num_train_epochs": 1,
|
322 |
+
"total_flos": 7.653345366712497e+17,
|
323 |
+
"trial_name": null,
|
324 |
+
"trial_params": null
|
325 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6b3651d5c4c3cc55e28b9dfa67c2a58c42d0d7720a97083fbfccf4d8feed9884
|
3 |
+
size 3055
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|