matt-tries-dl commited on
Commit
8965eb9
1 Parent(s): ce91638

trained v3

Browse files
res3.txt ADDED
@@ -0,0 +1,592 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /home/matt/hf/sqllama-V0/.venv/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: /opt/conda did not contain libcudart.so as expected! Searching further paths...
2
+ warn(msg)
3
+ The tokenizer class you load from this checkpoint is not the same type as the class this function is called from. It may result in unexpected tokenization.
4
+ The tokenizer class you load from this checkpoint is 'LLaMATokenizer'.
5
+ The class this function is called from is 'LlamaTokenizer'.
6
+
7
+ ===================================BUG REPORT===================================
8
+ Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
9
+ ================================================================================
10
+ CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so
11
+ CUDA SETUP: Highest compute capability among GPUs detected: 7.5
12
+ CUDA SETUP: Detected CUDA version 113
13
+ CUDA SETUP: Loading binary /home/matt/hf/sqllama-V0/.venv/lib/python3.7/site-packages/bitsandbytes/libbitsandbytes_cuda113.so...
14
+ Output exceeds the size limit. Open the full output data in a text editor
15
+
16
+ table: 2-16050349-13
17
+ columns: Rank,Name,Team,Games,Points
18
+ Q: What is Games, when Points is less than 340, and when Rank is greater than 3?
19
+ A: SELECT Games FROM 2-16050349-13 WHERE Points < 340 AND Rank > 3
20
+ END
21
+
22
+
23
+ table: 1-28962227-1
24
+ columns: Series,Premiere,Finale,Runners-up,Winner
25
+ Q: What is the date of the finale where Holly Bell was runner-up?
26
+ A: SELECT Finale FROM 1-28962227-1 WHERE Runners-up = 'Holly Bell'
27
+ END
28
+
29
+
30
+ table: 2-10652530-2
31
+ columns: Week,Date,Opponent,Result,Stadium,Record,Attendance
32
+ Q: What was the Browns record after they played the game at the Paul Brown stadium?
33
+ A: SELECT Record FROM 2-10652530-2 WHERE Stadium = 'paul brown stadium'
34
+ END
35
+
36
+
37
+ table: 2-18379129-4
38
+ columns: play,author,company,base,country
39
+ Q: Who is the author of the Play Electra?
40
+ ...
41
+ Q: What is 02-03, when School Year is % Learning In Latvian?
42
+ A: SELECT 02-03 FROM 2-16158579-1 WHERE School year = '% learning in latvian'
43
+ END
44
+
45
+ True
46
+ 92
47
+ 0
48
+ count 56355.000000
49
+ mean 101.219519
50
+ std 21.740325
51
+ min 63.000000
52
+ 25% 87.500000
53
+ 50% 97.000000
54
+ 75% 109.000000
55
+ max 461.000000
56
+ 32084
57
+ [500/500 7:38:36, Epoch 1/2]
58
+ Step Training Loss
59
+ 1 2.748800
60
+ 2 2.723800
61
+ 3 2.737600
62
+ 4 2.707100
63
+ 5 2.692800
64
+ 6 2.720700
65
+ 7 2.681400
66
+ 8 2.736400
67
+ 9 2.701800
68
+ 10 2.711700
69
+ 11 2.685800
70
+ 12 2.684300
71
+ 13 2.686300
72
+ 14 2.698800
73
+ 15 2.659300
74
+ 16 2.688900
75
+ 17 2.661800
76
+ 18 2.677700
77
+ 19 2.647100
78
+ 20 2.679800
79
+ 21 2.652000
80
+ 22 2.628900
81
+ 23 2.656100
82
+ 24 2.669100
83
+ 25 2.667800
84
+ 26 2.636300
85
+ 27 2.616800
86
+ 28 2.630600
87
+ 29 2.621000
88
+ 30 2.602000
89
+ 31 2.607900
90
+ 32 2.635800
91
+ 33 2.594600
92
+ 34 2.604400
93
+ 35 2.618900
94
+ 36 2.563400
95
+ 37 2.589200
96
+ 38 2.552100
97
+ 39 2.583600
98
+ 40 2.554500
99
+ 41 2.557400
100
+ 42 2.536700
101
+ 43 2.535000
102
+ 44 2.557900
103
+ 45 2.530100
104
+ 46 2.527900
105
+ 47 2.510100
106
+ 48 2.539100
107
+ 49 2.500100
108
+ 50 2.536200
109
+ 51 2.487100
110
+ 52 2.521700
111
+ 53 2.532600
112
+ 54 2.494500
113
+ 55 2.468900
114
+ 56 2.468700
115
+ 57 2.474300
116
+ 58 2.480900
117
+ 59 2.442800
118
+ 60 2.472800
119
+ 61 2.452900
120
+ 62 2.452000
121
+ 63 2.443100
122
+ 64 2.446700
123
+ 65 2.415100
124
+ 66 2.376300
125
+ 67 2.411500
126
+ 68 2.403900
127
+ 69 2.383800
128
+ 70 2.427800
129
+ 71 2.419400
130
+ 72 2.371900
131
+ 73 2.364400
132
+ 74 2.360000
133
+ 75 2.337600
134
+ 76 2.332800
135
+ 77 2.315700
136
+ 78 2.344200
137
+ 79 2.331700
138
+ 80 2.303100
139
+ 81 2.324700
140
+ 82 2.285900
141
+ 83 2.268000
142
+ 84 2.260600
143
+ 85 2.286100
144
+ 86 2.233600
145
+ 87 2.266200
146
+ 88 2.217000
147
+ 89 2.249300
148
+ 90 2.239000
149
+ 91 2.221900
150
+ 92 2.223300
151
+ 93 2.179500
152
+ 94 2.204400
153
+ 95 2.193200
154
+ 96 2.163800
155
+ 97 2.158200
156
+ 98 2.127700
157
+ 99 2.141400
158
+ 100 2.121400
159
+ 101 2.115500
160
+ 102 2.125200
161
+ 103 2.140100
162
+ 104 2.118400
163
+ 105 2.110400
164
+ 106 2.097300
165
+ 107 2.071400
166
+ 108 2.083400
167
+ 109 2.090200
168
+ 110 2.078200
169
+ 111 2.061100
170
+ 112 2.047500
171
+ 113 2.006100
172
+ 114 2.023800
173
+ 115 2.014000
174
+ 116 2.008800
175
+ 117 1.988800
176
+ 118 1.984900
177
+ 119 1.971000
178
+ 120 1.924100
179
+ 121 1.953100
180
+ 122 1.957800
181
+ 123 1.952500
182
+ 124 1.890400
183
+ 125 1.915900
184
+ 126 1.901100
185
+ 127 1.879900
186
+ 128 1.834100
187
+ 129 1.855900
188
+ 130 1.853800
189
+ 131 1.869200
190
+ 132 1.821400
191
+ 133 1.835100
192
+ 134 1.817700
193
+ 135 1.785800
194
+ 136 1.764000
195
+ 137 1.796800
196
+ 138 1.751100
197
+ 139 1.756500
198
+ 140 1.789900
199
+ 141 1.773100
200
+ 142 1.729200
201
+ 143 1.700200
202
+ 144 1.721200
203
+ 145 1.690600
204
+ 146 1.687700
205
+ 147 1.743500
206
+ 148 1.690000
207
+ 149 1.687200
208
+ 150 1.663000
209
+ 151 1.648600
210
+ 152 1.667100
211
+ 153 1.665600
212
+ 154 1.647000
213
+ 155 1.629500
214
+ 156 1.620800
215
+ 157 1.616400
216
+ 158 1.658500
217
+ 159 1.593900
218
+ 160 1.604300
219
+ 161 1.621200
220
+ 162 1.607900
221
+ 163 1.591100
222
+ 164 1.598100
223
+ 165 1.579700
224
+ 166 1.545500
225
+ 167 1.582100
226
+ 168 1.568300
227
+ 169 1.557900
228
+ 170 1.561300
229
+ 171 1.521800
230
+ 172 1.542500
231
+ 173 1.502300
232
+ 174 1.513900
233
+ 175 1.501500
234
+ 176 1.551200
235
+ 177 1.495600
236
+ 178 1.504000
237
+ 179 1.512500
238
+ 180 1.488200
239
+ 181 1.492200
240
+ 182 1.494300
241
+ 183 1.494800
242
+ 184 1.446100
243
+ 185 1.514700
244
+ 186 1.450900
245
+ 187 1.476900
246
+ 188 1.447100
247
+ 189 1.490800
248
+ 190 1.433200
249
+ 191 1.438100
250
+ 192 1.410500
251
+ 193 1.422600
252
+ 194 1.405500
253
+ 195 1.439400
254
+ 196 1.448100
255
+ 197 1.410200
256
+ 198 1.403800
257
+ 199 1.464400
258
+ 200 1.417700
259
+ 201 1.419500
260
+ 202 1.419400
261
+ 203 1.387700
262
+ 204 1.400400
263
+ 205 1.404700
264
+ 206 1.398400
265
+ 207 1.358000
266
+ 208 1.359600
267
+ 209 1.367700
268
+ 210 1.358600
269
+ 211 1.369200
270
+ 212 1.373700
271
+ 213 1.395100
272
+ 214 1.360800
273
+ 215 1.343900
274
+ 216 1.330300
275
+ 217 1.328800
276
+ 218 1.369900
277
+ 219 1.346300
278
+ 220 1.379700
279
+ 221 1.326000
280
+ 222 1.334600
281
+ 223 1.339100
282
+ 224 1.349200
283
+ 225 1.324800
284
+ 226 1.303600
285
+ 227 1.299900
286
+ 228 1.338800
287
+ 229 1.331800
288
+ 230 1.351400
289
+ 231 1.314200
290
+ 232 1.293600
291
+ 233 1.322100
292
+ 234 1.295800
293
+ 235 1.302500
294
+ 236 1.338900
295
+ 237 1.308900
296
+ 238 1.290100
297
+ 239 1.323300
298
+ 240 1.270500
299
+ 241 1.246300
300
+ 242 1.303900
301
+ 243 1.324800
302
+ 244 1.216000
303
+ 245 1.303500
304
+ 246 1.304900
305
+ 247 1.273300
306
+ 248 1.278300
307
+ 249 1.252000
308
+ 250 1.283400
309
+ 251 1.271600
310
+ 252 1.300300
311
+ 253 1.265800
312
+ 254 1.249200
313
+ 255 1.252600
314
+ 256 1.265500
315
+ 257 1.228600
316
+ 258 1.257300
317
+ 259 1.288900
318
+ 260 1.257200
319
+ 261 1.243700
320
+ 262 1.272100
321
+ 263 1.252000
322
+ 264 1.264900
323
+ 265 1.268800
324
+ 266 1.256000
325
+ 267 1.230200
326
+ 268 1.231700
327
+ 269 1.243400
328
+ 270 1.285200
329
+ 271 1.225500
330
+ 272 1.217900
331
+ 273 1.209200
332
+ 274 1.224200
333
+ 275 1.226400
334
+ 276 1.261500
335
+ 277 1.223900
336
+ 278 1.244000
337
+ 279 1.226600
338
+ 280 1.235000
339
+ 281 1.213400
340
+ 282 1.177600
341
+ 283 1.218100
342
+ 284 1.231900
343
+ 285 1.200900
344
+ 286 1.223400
345
+ 287 1.235100
346
+ 288 1.232500
347
+ 289 1.230100
348
+ 290 1.225900
349
+ 291 1.182700
350
+ 292 1.237100
351
+ 293 1.201000
352
+ 294 1.213000
353
+ 295 1.205500
354
+ 296 1.181900
355
+ 297 1.198300
356
+ 298 1.195200
357
+ 299 1.215000
358
+ 300 1.195500
359
+ 301 1.186100
360
+ 302 1.174900
361
+ 303 1.184400
362
+ 304 1.207100
363
+ 305 1.181100
364
+ 306 1.195300
365
+ 307 1.189000
366
+ 308 1.180200
367
+ 309 1.167200
368
+ 310 1.206700
369
+ 311 1.203600
370
+ 312 1.186600
371
+ 313 1.224100
372
+ 314 1.180000
373
+ 315 1.186600
374
+ 316 1.150700
375
+ 317 1.165700
376
+ 318 1.178100
377
+ 319 1.148300
378
+ 320 1.153600
379
+ 321 1.189200
380
+ 322 1.182100
381
+ 323 1.183800
382
+ 324 1.202900
383
+ 325 1.196600
384
+ 326 1.200800
385
+ 327 1.153100
386
+ 328 1.212400
387
+ 329 1.167300
388
+ 330 1.188300
389
+ 331 1.179300
390
+ 332 1.211400
391
+ 333 1.169900
392
+ 334 1.179300
393
+ 335 1.153300
394
+ 336 1.188900
395
+ 337 1.179200
396
+ 338 1.217300
397
+ 339 1.169700
398
+ 340 1.177700
399
+ 341 1.197300
400
+ 342 1.177800
401
+ 343 1.169700
402
+ 344 1.186800
403
+ 345 1.180000
404
+ 346 1.193400
405
+ 347 1.171900
406
+ 348 1.190000
407
+ 349 1.160900
408
+ 350 1.170800
409
+ 351 1.166900
410
+ 352 1.183200
411
+ 353 1.118200
412
+ 354 1.185900
413
+ 355 1.157800
414
+ 356 1.160200
415
+ 357 1.184200
416
+ 358 1.172100
417
+ 359 1.143800
418
+ 360 1.178000
419
+ 361 1.157900
420
+ 362 1.151700
421
+ 363 1.196600
422
+ 364 1.181800
423
+ 365 1.195600
424
+ 366 1.165000
425
+ 367 1.157300
426
+ 368 1.165200
427
+ 369 1.167700
428
+ 370 1.184900
429
+ 371 1.168400
430
+ 372 1.150500
431
+ 373 1.152900
432
+ 374 1.158900
433
+ 375 1.143900
434
+ 376 1.157200
435
+ 377 1.146800
436
+ 378 1.142600
437
+ 379 1.140600
438
+ 380 1.142400
439
+ 381 1.114100
440
+ 382 1.169700
441
+ 383 1.142500
442
+ 384 1.176000
443
+ 385 1.160600
444
+ 386 1.164700
445
+ 387 1.124000
446
+ 388 1.134500
447
+ 389 1.185500
448
+ 390 1.154300
449
+ 391 1.125500
450
+ 392 1.174400
451
+ 393 1.132800
452
+ 394 1.145200
453
+ 395 1.129800
454
+ 396 1.140600
455
+ 397 1.126000
456
+ 398 1.182800
457
+ 399 1.127800
458
+ 400 1.155000
459
+ 401 1.134600
460
+ 402 1.155900
461
+ 403 1.150400
462
+ 404 1.141700
463
+ 405 1.131500
464
+ 406 1.169600
465
+ 407 1.170500
466
+ 408 1.129100
467
+ 409 1.151700
468
+ 410 1.168200
469
+ 411 1.109100
470
+ 412 1.129700
471
+ 413 1.143900
472
+ 414 1.157300
473
+ 415 1.128900
474
+ 416 1.171500
475
+ 417 1.141600
476
+ 418 1.157700
477
+ 419 1.137000
478
+ 420 1.154000
479
+ 421 1.167300
480
+ 422 1.137400
481
+ 423 1.121500
482
+ 424 1.128500
483
+ 425 1.130300
484
+ 426 1.162100
485
+ 427 1.155100
486
+ 428 1.145300
487
+ 429 1.121000
488
+ 430 1.182200
489
+ 431 1.157000
490
+ 432 1.162300
491
+ 433 1.135200
492
+ 434 1.141300
493
+ 435 1.151700
494
+ 436 1.148000
495
+ 437 1.132500
496
+ 438 1.163000
497
+ 439 1.116300
498
+ 440 1.142000
499
+ 441 1.091700
500
+ 442 1.141500
501
+ 443 1.154900
502
+ 444 1.120400
503
+ 445 1.173700
504
+ 446 1.138300
505
+ 447 1.135600
506
+ 448 1.138800
507
+ 449 1.126800
508
+ 450 1.129400
509
+ 451 1.146300
510
+ 452 1.104200
511
+ 453 1.163500
512
+ 454 1.169300
513
+ 455 1.147100
514
+ 456 1.157100
515
+ 457 1.122100
516
+ 458 1.121900
517
+ 459 1.150500
518
+ 460 1.115700
519
+ 461 1.121100
520
+ 462 1.123400
521
+ 463 1.097500
522
+ 464 1.103800
523
+ 465 1.167700
524
+ 466 1.130000
525
+ 467 1.164500
526
+ 468 1.127200
527
+ 469 1.133800
528
+ 470 1.132700
529
+ 471 1.122800
530
+ 472 1.159500
531
+ 473 1.122900
532
+ 474 1.105000
533
+ 475 1.145700
534
+ 476 1.086400
535
+ 477 1.112600
536
+ 478 1.139300
537
+ 479 1.135000
538
+ 480 1.135200
539
+ 481 1.117500
540
+ 482 1.102300
541
+ 483 1.147700
542
+ 484 1.119200
543
+ 485 1.125800
544
+ 486 1.135400
545
+ 487 1.149500
546
+ 488 1.099400
547
+ 489 1.153900
548
+ 490 1.122700
549
+ 491 1.089400
550
+ 492 1.167200
551
+ 493 1.151300
552
+ 494 1.131400
553
+ 495 1.131400
554
+ 496 1.145200
555
+ 497 1.125700
556
+ 498 1.119300
557
+ 499 1.128600
558
+ 500 1.121000
559
+ /home/matt/hf/sqllama-V0/.venv/lib/python3.7/site-packages/transformers/generation/utils.py:1220: UserWarning: You have modified the pretrained model configuration to control generation. This is a deprecated strategy to control generation and will be removed soon, in a future version. Please use a generation configuration file (see https://huggingface.co/docs/transformers/main_classes/text_generation)
560
+ "You have modified the pretrained model configuration to control generation. This is a"
561
+ /home/matt/hf/sqllama-V0/.venv/lib/python3.7/site-packages/torch/utils/checkpoint.py:31: UserWarning: None of the inputs have requires_grad=True. Gradients will be None
562
+ warnings.warn("None of the inputs have requires_grad=True. Gradients will be None")
563
+ Output exceeds the size limit. Open the full output data in a text editor
564
+ from model
565
+ <unk>table: 2-11561331-17
566
+ columns: Name,Actual version,System,Platform,License
567
+ Q: Which System's Name is Steem, and has a Freeware License?
568
+ A: SELECT Name FROM 2-11561331-17 WHERE License = 'Freeware' AND System = 'Steem'
569
+ END
570
+ \end{code}
571
+
572
+
573
+
574
+ expected answer
575
+ SELECT System FROM 2-11561331-17 WHERE License = 'freeware' AND Name = 'steem'
576
+ END
577
+
578
+ from model
579
+ <unk>table: 1-18847736-2
580
+ columns: Game,Date,Opponent,Result,Dolphins points,Opponents,Record,Attendance
581
+ Q: What is the date when the opponent is the New England Patriots?
582
+ A: SELECT Date FROM 1-18847736-2 WHERE Opponent = 'New England Patriots'
583
+ END
584
+ \end
585
+
586
+ expected answer
587
+ SELECT Date FROM 1-18847736-2 WHERE Opponent = 'New England Patriots'
588
+ END
589
+ ...
590
+ expected answer
591
+ SELECT Manufacturer FROM 1-17801022-1 WHERE Date = 'November 2'
592
+ END
sqllama-out3/adapter_config.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "decapoda-research/llama-7b-hf",
3
+ "bias": "none",
4
+ "enable_lora": null,
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "lora_alpha": 16,
8
+ "lora_dropout": 0.1,
9
+ "merge_weights": false,
10
+ "modules_to_save": null,
11
+ "peft_type": "LORA",
12
+ "r": 4,
13
+ "target_modules": [
14
+ "q_proj",
15
+ "v_proj"
16
+ ],
17
+ "task_type": "CASUAL_LM"
18
+ }
sqllama-out3/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d12d7102022c5ae89b006db222885aa4acde5b853063bf288c24ebc430f26bb7
3
+ size 8434381
wikisql.ipynb CHANGED
@@ -55,7 +55,7 @@
55
  {
56
  "data": {
57
  "application/vnd.jupyter.widget-view+json": {
58
- "model_id": "0b970b9989854c33aa4d19fa9457d7a2",
59
  "version_major": 2,
60
  "version_minor": 0
61
  },
@@ -96,38 +96,38 @@
96
  "output_type": "stream",
97
  "text": [
98
  "\n",
99
- "table: 2-1137692-1\n",
100
- "columns: Entrant,Constructor,Chassis,Engine †,Tyre,Driver,Rounds\n",
101
- "Q: What were the rounds on the Engine † of the Ferrari 048?\n",
102
- "A: SELECT Rounds FROM 2-1137692-1 WHERE Engine = 'ferrari 048'\n",
103
  "END\n",
104
  "\n",
105
  "\n",
106
- "table: 1-21530474-1\n",
107
- "columns: Chassis code,Model no.,Production years,Drivetrain,Transmission,Engine type,Engine code,Region(s)\n",
108
- "Q: Name the drivetrain for 1ur-fse for usf41\n",
109
- "A: SELECT Drivetrain FROM 1-21530474-1 WHERE Engine code = '1UR-FSE' AND Chassis code = 'USF41'\n",
110
  "END\n",
111
  "\n",
112
  "\n",
113
- "table: 2-14155087-1\n",
114
- "columns: Callsign,Area served,Frequency,Band,On-air ID,Purpose\n",
115
- "Q: What is the Callsign with an Area of tamworth and frequency of 0 88.9?\n",
116
- "A: SELECT Callsign FROM 2-14155087-1 WHERE Area served = 'tamworth' AND Frequency = '0 88.9'\n",
117
  "END\n",
118
  "\n",
119
  "\n",
120
- "table: 2-17580726-2\n",
121
- "columns: Date,Opponent,Venue,Score,Attendance,Scorers\n",
122
- "Q: What is the number of people in attendance when Tonbridge Angels is the opponent?\n",
123
- "A: SELECT Attendance FROM 2-17580726-2 WHERE Opponent = 'tonbridge angels'\n",
124
  "END\n",
125
  "\n",
126
  "\n",
127
- "table: 1-27986200-3\n",
128
- "columns: Proceed to Quarter-final,Match points,Aggregate score,Points margin,Eliminated from competition\n",
129
- "Q: What were the match points when Bordeaux-Bègles was eliminated from competition? \n",
130
- "A: SELECT Match points FROM 1-27986200-3 WHERE Eliminated from competition = 'Bordeaux-Bègles'\n",
131
  "END\n",
132
  "\n"
133
  ]
@@ -278,7 +278,7 @@
278
  {
279
  "data": {
280
  "application/vnd.jupyter.widget-view+json": {
281
- "model_id": "01debeeb68bb40a7b83031e88c5ace1e",
282
  "version_major": 2,
283
  "version_minor": 0
284
  },
@@ -320,12 +320,11 @@
320
  "LORA_R = 4\n",
321
  "LORA_ALPHA = 16\n",
322
  "LORA_DROPOUT = .1\n",
323
- "CUTOFF_LEN = 256\n",
324
  "BATCH = 128\n",
325
  "MICRO_BATCH = 4\n",
326
  "N_GAS = BATCH//MICRO_BATCH\n",
327
- "EPOCHS = 1\n",
328
- "LR = 1e-4\n",
329
  "\n",
330
  "lora_cfg = LoraConfig(\n",
331
  " r = LORA_R,\n",
@@ -345,7 +344,7 @@
345
  " learning_rate=LR,\n",
346
  " fp16=True,\n",
347
  " logging_steps=1,\n",
348
- " output_dir='sqllama-out2',\n",
349
  " save_total_limit=3,\n",
350
  " remove_unused_columns=False\n",
351
  ")\n"
@@ -362,8 +361,8 @@
362
  "\n",
363
  " <div>\n",
364
  " \n",
365
- " <progress value='250' max='250' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
366
- " [250/250 3:49:26, Epoch 0/1]\n",
367
  " </div>\n",
368
  " <table border=\"1\" class=\"dataframe\">\n",
369
  " <thead>\n",
@@ -379,999 +378,7 @@
379
  " </tr>\n",
380
  " <tr>\n",
381
  " <td>2</td>\n",
382
- " <td>2.699100</td>\n",
383
- " </tr>\n",
384
- " <tr>\n",
385
- " <td>3</td>\n",
386
- " <td>2.670200</td>\n",
387
- " </tr>\n",
388
- " <tr>\n",
389
- " <td>4</td>\n",
390
- " <td>2.600500</td>\n",
391
- " </tr>\n",
392
- " <tr>\n",
393
- " <td>5</td>\n",
394
- " <td>2.560100</td>\n",
395
- " </tr>\n",
396
- " <tr>\n",
397
- " <td>6</td>\n",
398
- " <td>2.556800</td>\n",
399
- " </tr>\n",
400
- " <tr>\n",
401
- " <td>7</td>\n",
402
- " <td>2.498100</td>\n",
403
- " </tr>\n",
404
- " <tr>\n",
405
- " <td>8</td>\n",
406
- " <td>2.515400</td>\n",
407
- " </tr>\n",
408
- " <tr>\n",
409
- " <td>9</td>\n",
410
- " <td>2.436100</td>\n",
411
- " </tr>\n",
412
- " <tr>\n",
413
- " <td>10</td>\n",
414
- " <td>2.411700</td>\n",
415
- " </tr>\n",
416
- " <tr>\n",
417
- " <td>11</td>\n",
418
- " <td>2.346400</td>\n",
419
- " </tr>\n",
420
- " <tr>\n",
421
- " <td>12</td>\n",
422
- " <td>2.276300</td>\n",
423
- " </tr>\n",
424
- " <tr>\n",
425
- " <td>13</td>\n",
426
- " <td>2.238000</td>\n",
427
- " </tr>\n",
428
- " <tr>\n",
429
- " <td>14</td>\n",
430
- " <td>2.189100</td>\n",
431
- " </tr>\n",
432
- " <tr>\n",
433
- " <td>15</td>\n",
434
- " <td>2.109200</td>\n",
435
- " </tr>\n",
436
- " <tr>\n",
437
- " <td>16</td>\n",
438
- " <td>2.058000</td>\n",
439
- " </tr>\n",
440
- " <tr>\n",
441
- " <td>17</td>\n",
442
- " <td>1.983900</td>\n",
443
- " </tr>\n",
444
- " <tr>\n",
445
- " <td>18</td>\n",
446
- " <td>1.928600</td>\n",
447
- " </tr>\n",
448
- " <tr>\n",
449
- " <td>19</td>\n",
450
- " <td>1.824100</td>\n",
451
- " </tr>\n",
452
- " <tr>\n",
453
- " <td>20</td>\n",
454
- " <td>1.794700</td>\n",
455
- " </tr>\n",
456
- " <tr>\n",
457
- " <td>21</td>\n",
458
- " <td>1.681200</td>\n",
459
- " </tr>\n",
460
- " <tr>\n",
461
- " <td>22</td>\n",
462
- " <td>1.598900</td>\n",
463
- " </tr>\n",
464
- " <tr>\n",
465
- " <td>23</td>\n",
466
- " <td>1.562000</td>\n",
467
- " </tr>\n",
468
- " <tr>\n",
469
- " <td>24</td>\n",
470
- " <td>1.527200</td>\n",
471
- " </tr>\n",
472
- " <tr>\n",
473
- " <td>25</td>\n",
474
- " <td>1.518700</td>\n",
475
- " </tr>\n",
476
- " <tr>\n",
477
- " <td>26</td>\n",
478
- " <td>1.493100</td>\n",
479
- " </tr>\n",
480
- " <tr>\n",
481
- " <td>27</td>\n",
482
- " <td>1.500500</td>\n",
483
- " </tr>\n",
484
- " <tr>\n",
485
- " <td>28</td>\n",
486
- " <td>1.464000</td>\n",
487
- " </tr>\n",
488
- " <tr>\n",
489
- " <td>29</td>\n",
490
- " <td>1.386900</td>\n",
491
- " </tr>\n",
492
- " <tr>\n",
493
- " <td>30</td>\n",
494
- " <td>1.373400</td>\n",
495
- " </tr>\n",
496
- " <tr>\n",
497
- " <td>31</td>\n",
498
- " <td>1.362200</td>\n",
499
- " </tr>\n",
500
- " <tr>\n",
501
- " <td>32</td>\n",
502
- " <td>1.360800</td>\n",
503
- " </tr>\n",
504
- " <tr>\n",
505
- " <td>33</td>\n",
506
- " <td>1.321000</td>\n",
507
- " </tr>\n",
508
- " <tr>\n",
509
- " <td>34</td>\n",
510
- " <td>1.310500</td>\n",
511
- " </tr>\n",
512
- " <tr>\n",
513
- " <td>35</td>\n",
514
- " <td>1.302600</td>\n",
515
- " </tr>\n",
516
- " <tr>\n",
517
- " <td>36</td>\n",
518
- " <td>1.256100</td>\n",
519
- " </tr>\n",
520
- " <tr>\n",
521
- " <td>37</td>\n",
522
- " <td>1.252500</td>\n",
523
- " </tr>\n",
524
- " <tr>\n",
525
- " <td>38</td>\n",
526
- " <td>1.202300</td>\n",
527
- " </tr>\n",
528
- " <tr>\n",
529
- " <td>39</td>\n",
530
- " <td>1.249100</td>\n",
531
- " </tr>\n",
532
- " <tr>\n",
533
- " <td>40</td>\n",
534
- " <td>1.188600</td>\n",
535
- " </tr>\n",
536
- " <tr>\n",
537
- " <td>41</td>\n",
538
- " <td>1.203200</td>\n",
539
- " </tr>\n",
540
- " <tr>\n",
541
- " <td>42</td>\n",
542
- " <td>1.150000</td>\n",
543
- " </tr>\n",
544
- " <tr>\n",
545
- " <td>43</td>\n",
546
- " <td>1.182000</td>\n",
547
- " </tr>\n",
548
- " <tr>\n",
549
- " <td>44</td>\n",
550
- " <td>1.192300</td>\n",
551
- " </tr>\n",
552
- " <tr>\n",
553
- " <td>45</td>\n",
554
- " <td>1.133100</td>\n",
555
- " </tr>\n",
556
- " <tr>\n",
557
- " <td>46</td>\n",
558
- " <td>1.119600</td>\n",
559
- " </tr>\n",
560
- " <tr>\n",
561
- " <td>47</td>\n",
562
- " <td>1.097000</td>\n",
563
- " </tr>\n",
564
- " <tr>\n",
565
- " <td>48</td>\n",
566
- " <td>1.142100</td>\n",
567
- " </tr>\n",
568
- " <tr>\n",
569
- " <td>49</td>\n",
570
- " <td>1.117200</td>\n",
571
- " </tr>\n",
572
- " <tr>\n",
573
- " <td>50</td>\n",
574
- " <td>1.129200</td>\n",
575
- " </tr>\n",
576
- " <tr>\n",
577
- " <td>51</td>\n",
578
- " <td>1.087300</td>\n",
579
- " </tr>\n",
580
- " <tr>\n",
581
- " <td>52</td>\n",
582
- " <td>1.098700</td>\n",
583
- " </tr>\n",
584
- " <tr>\n",
585
- " <td>53</td>\n",
586
- " <td>1.135400</td>\n",
587
- " </tr>\n",
588
- " <tr>\n",
589
- " <td>54</td>\n",
590
- " <td>1.071700</td>\n",
591
- " </tr>\n",
592
- " <tr>\n",
593
- " <td>55</td>\n",
594
- " <td>1.087300</td>\n",
595
- " </tr>\n",
596
- " <tr>\n",
597
- " <td>56</td>\n",
598
- " <td>1.051400</td>\n",
599
- " </tr>\n",
600
- " <tr>\n",
601
- " <td>57</td>\n",
602
- " <td>1.068300</td>\n",
603
- " </tr>\n",
604
- " <tr>\n",
605
- " <td>58</td>\n",
606
- " <td>1.092500</td>\n",
607
- " </tr>\n",
608
- " <tr>\n",
609
- " <td>59</td>\n",
610
- " <td>1.068600</td>\n",
611
- " </tr>\n",
612
- " <tr>\n",
613
- " <td>60</td>\n",
614
- " <td>1.072800</td>\n",
615
- " </tr>\n",
616
- " <tr>\n",
617
- " <td>61</td>\n",
618
- " <td>1.074000</td>\n",
619
- " </tr>\n",
620
- " <tr>\n",
621
- " <td>62</td>\n",
622
- " <td>1.060400</td>\n",
623
- " </tr>\n",
624
- " <tr>\n",
625
- " <td>63</td>\n",
626
- " <td>1.065800</td>\n",
627
- " </tr>\n",
628
- " <tr>\n",
629
- " <td>64</td>\n",
630
- " <td>1.075900</td>\n",
631
- " </tr>\n",
632
- " <tr>\n",
633
- " <td>65</td>\n",
634
- " <td>1.059500</td>\n",
635
- " </tr>\n",
636
- " <tr>\n",
637
- " <td>66</td>\n",
638
- " <td>1.039600</td>\n",
639
- " </tr>\n",
640
- " <tr>\n",
641
- " <td>67</td>\n",
642
- " <td>1.051400</td>\n",
643
- " </tr>\n",
644
- " <tr>\n",
645
- " <td>68</td>\n",
646
- " <td>1.049500</td>\n",
647
- " </tr>\n",
648
- " <tr>\n",
649
- " <td>69</td>\n",
650
- " <td>1.023800</td>\n",
651
- " </tr>\n",
652
- " <tr>\n",
653
- " <td>70</td>\n",
654
- " <td>1.071900</td>\n",
655
- " </tr>\n",
656
- " <tr>\n",
657
- " <td>71</td>\n",
658
- " <td>1.051000</td>\n",
659
- " </tr>\n",
660
- " <tr>\n",
661
- " <td>72</td>\n",
662
- " <td>1.034700</td>\n",
663
- " </tr>\n",
664
- " <tr>\n",
665
- " <td>73</td>\n",
666
- " <td>1.041600</td>\n",
667
- " </tr>\n",
668
- " <tr>\n",
669
- " <td>74</td>\n",
670
- " <td>1.030900</td>\n",
671
- " </tr>\n",
672
- " <tr>\n",
673
- " <td>75</td>\n",
674
- " <td>1.010800</td>\n",
675
- " </tr>\n",
676
- " <tr>\n",
677
- " <td>76</td>\n",
678
- " <td>1.019800</td>\n",
679
- " </tr>\n",
680
- " <tr>\n",
681
- " <td>77</td>\n",
682
- " <td>1.005000</td>\n",
683
- " </tr>\n",
684
- " <tr>\n",
685
- " <td>78</td>\n",
686
- " <td>1.043800</td>\n",
687
- " </tr>\n",
688
- " <tr>\n",
689
- " <td>79</td>\n",
690
- " <td>1.009200</td>\n",
691
- " </tr>\n",
692
- " <tr>\n",
693
- " <td>80</td>\n",
694
- " <td>1.017100</td>\n",
695
- " </tr>\n",
696
- " <tr>\n",
697
- " <td>81</td>\n",
698
- " <td>1.044600</td>\n",
699
- " </tr>\n",
700
- " <tr>\n",
701
- " <td>82</td>\n",
702
- " <td>1.022600</td>\n",
703
- " </tr>\n",
704
- " <tr>\n",
705
- " <td>83</td>\n",
706
- " <td>1.011400</td>\n",
707
- " </tr>\n",
708
- " <tr>\n",
709
- " <td>84</td>\n",
710
- " <td>0.996600</td>\n",
711
- " </tr>\n",
712
- " <tr>\n",
713
- " <td>85</td>\n",
714
- " <td>1.029900</td>\n",
715
- " </tr>\n",
716
- " <tr>\n",
717
- " <td>86</td>\n",
718
- " <td>0.988200</td>\n",
719
- " </tr>\n",
720
- " <tr>\n",
721
- " <td>87</td>\n",
722
- " <td>1.005600</td>\n",
723
- " </tr>\n",
724
- " <tr>\n",
725
- " <td>88</td>\n",
726
- " <td>0.986600</td>\n",
727
- " </tr>\n",
728
- " <tr>\n",
729
- " <td>89</td>\n",
730
- " <td>1.025300</td>\n",
731
- " </tr>\n",
732
- " <tr>\n",
733
- " <td>90</td>\n",
734
- " <td>1.012500</td>\n",
735
- " </tr>\n",
736
- " <tr>\n",
737
- " <td>91</td>\n",
738
- " <td>0.988100</td>\n",
739
- " </tr>\n",
740
- " <tr>\n",
741
- " <td>92</td>\n",
742
- " <td>1.001800</td>\n",
743
- " </tr>\n",
744
- " <tr>\n",
745
- " <td>93</td>\n",
746
- " <td>0.987100</td>\n",
747
- " </tr>\n",
748
- " <tr>\n",
749
- " <td>94</td>\n",
750
- " <td>1.017600</td>\n",
751
- " </tr>\n",
752
- " <tr>\n",
753
- " <td>95</td>\n",
754
- " <td>0.998500</td>\n",
755
- " </tr>\n",
756
- " <tr>\n",
757
- " <td>96</td>\n",
758
- " <td>0.966600</td>\n",
759
- " </tr>\n",
760
- " <tr>\n",
761
- " <td>97</td>\n",
762
- " <td>0.983700</td>\n",
763
- " </tr>\n",
764
- " <tr>\n",
765
- " <td>98</td>\n",
766
- " <td>0.961800</td>\n",
767
- " </tr>\n",
768
- " <tr>\n",
769
- " <td>99</td>\n",
770
- " <td>0.969000</td>\n",
771
- " </tr>\n",
772
- " <tr>\n",
773
- " <td>100</td>\n",
774
- " <td>0.989200</td>\n",
775
- " </tr>\n",
776
- " <tr>\n",
777
- " <td>101</td>\n",
778
- " <td>0.956400</td>\n",
779
- " </tr>\n",
780
- " <tr>\n",
781
- " <td>102</td>\n",
782
- " <td>0.976000</td>\n",
783
- " </tr>\n",
784
- " <tr>\n",
785
- " <td>103</td>\n",
786
- " <td>1.000100</td>\n",
787
- " </tr>\n",
788
- " <tr>\n",
789
- " <td>104</td>\n",
790
- " <td>1.001500</td>\n",
791
- " </tr>\n",
792
- " <tr>\n",
793
- " <td>105</td>\n",
794
- " <td>0.995900</td>\n",
795
- " </tr>\n",
796
- " <tr>\n",
797
- " <td>106</td>\n",
798
- " <td>0.989700</td>\n",
799
- " </tr>\n",
800
- " <tr>\n",
801
- " <td>107</td>\n",
802
- " <td>0.965700</td>\n",
803
- " </tr>\n",
804
- " <tr>\n",
805
- " <td>108</td>\n",
806
- " <td>0.968400</td>\n",
807
- " </tr>\n",
808
- " <tr>\n",
809
- " <td>109</td>\n",
810
- " <td>1.019600</td>\n",
811
- " </tr>\n",
812
- " <tr>\n",
813
- " <td>110</td>\n",
814
- " <td>1.000100</td>\n",
815
- " </tr>\n",
816
- " <tr>\n",
817
- " <td>111</td>\n",
818
- " <td>0.978500</td>\n",
819
- " </tr>\n",
820
- " <tr>\n",
821
- " <td>112</td>\n",
822
- " <td>0.978900</td>\n",
823
- " </tr>\n",
824
- " <tr>\n",
825
- " <td>113</td>\n",
826
- " <td>0.952600</td>\n",
827
- " </tr>\n",
828
- " <tr>\n",
829
- " <td>114</td>\n",
830
- " <td>0.975400</td>\n",
831
- " </tr>\n",
832
- " <tr>\n",
833
- " <td>115</td>\n",
834
- " <td>0.989400</td>\n",
835
- " </tr>\n",
836
- " <tr>\n",
837
- " <td>116</td>\n",
838
- " <td>0.968500</td>\n",
839
- " </tr>\n",
840
- " <tr>\n",
841
- " <td>117</td>\n",
842
- " <td>0.960100</td>\n",
843
- " </tr>\n",
844
- " <tr>\n",
845
- " <td>118</td>\n",
846
- " <td>0.979100</td>\n",
847
- " </tr>\n",
848
- " <tr>\n",
849
- " <td>119</td>\n",
850
- " <td>0.955100</td>\n",
851
- " </tr>\n",
852
- " <tr>\n",
853
- " <td>120</td>\n",
854
- " <td>0.934800</td>\n",
855
- " </tr>\n",
856
- " <tr>\n",
857
- " <td>121</td>\n",
858
- " <td>0.943600</td>\n",
859
- " </tr>\n",
860
- " <tr>\n",
861
- " <td>122</td>\n",
862
- " <td>0.976700</td>\n",
863
- " </tr>\n",
864
- " <tr>\n",
865
- " <td>123</td>\n",
866
- " <td>0.998700</td>\n",
867
- " </tr>\n",
868
- " <tr>\n",
869
- " <td>124</td>\n",
870
- " <td>0.930500</td>\n",
871
- " </tr>\n",
872
- " <tr>\n",
873
- " <td>125</td>\n",
874
- " <td>0.953500</td>\n",
875
- " </tr>\n",
876
- " <tr>\n",
877
- " <td>126</td>\n",
878
- " <td>0.978000</td>\n",
879
- " </tr>\n",
880
- " <tr>\n",
881
- " <td>127</td>\n",
882
- " <td>0.967300</td>\n",
883
- " </tr>\n",
884
- " <tr>\n",
885
- " <td>128</td>\n",
886
- " <td>0.929400</td>\n",
887
- " </tr>\n",
888
- " <tr>\n",
889
- " <td>129</td>\n",
890
- " <td>0.963100</td>\n",
891
- " </tr>\n",
892
- " <tr>\n",
893
- " <td>130</td>\n",
894
- " <td>0.961500</td>\n",
895
- " </tr>\n",
896
- " <tr>\n",
897
- " <td>131</td>\n",
898
- " <td>0.978500</td>\n",
899
- " </tr>\n",
900
- " <tr>\n",
901
- " <td>132</td>\n",
902
- " <td>0.937200</td>\n",
903
- " </tr>\n",
904
- " <tr>\n",
905
- " <td>133</td>\n",
906
- " <td>0.953400</td>\n",
907
- " </tr>\n",
908
- " <tr>\n",
909
- " <td>134</td>\n",
910
- " <td>0.962000</td>\n",
911
- " </tr>\n",
912
- " <tr>\n",
913
- " <td>135</td>\n",
914
- " <td>0.950700</td>\n",
915
- " </tr>\n",
916
- " <tr>\n",
917
- " <td>136</td>\n",
918
- " <td>0.925100</td>\n",
919
- " </tr>\n",
920
- " <tr>\n",
921
- " <td>137</td>\n",
922
- " <td>0.958800</td>\n",
923
- " </tr>\n",
924
- " <tr>\n",
925
- " <td>138</td>\n",
926
- " <td>0.926200</td>\n",
927
- " </tr>\n",
928
- " <tr>\n",
929
- " <td>139</td>\n",
930
- " <td>0.930600</td>\n",
931
- " </tr>\n",
932
- " <tr>\n",
933
- " <td>140</td>\n",
934
- " <td>0.968900</td>\n",
935
- " </tr>\n",
936
- " <tr>\n",
937
- " <td>141</td>\n",
938
- " <td>0.970400</td>\n",
939
- " </tr>\n",
940
- " <tr>\n",
941
- " <td>142</td>\n",
942
- " <td>0.927100</td>\n",
943
- " </tr>\n",
944
- " <tr>\n",
945
- " <td>143</td>\n",
946
- " <td>0.911800</td>\n",
947
- " </tr>\n",
948
- " <tr>\n",
949
- " <td>144</td>\n",
950
- " <td>0.953200</td>\n",
951
- " </tr>\n",
952
- " <tr>\n",
953
- " <td>145</td>\n",
954
- " <td>0.907100</td>\n",
955
- " </tr>\n",
956
- " <tr>\n",
957
- " <td>146</td>\n",
958
- " <td>0.935900</td>\n",
959
- " </tr>\n",
960
- " <tr>\n",
961
- " <td>147</td>\n",
962
- " <td>0.970600</td>\n",
963
- " </tr>\n",
964
- " <tr>\n",
965
- " <td>148</td>\n",
966
- " <td>0.920400</td>\n",
967
- " </tr>\n",
968
- " <tr>\n",
969
- " <td>149</td>\n",
970
- " <td>0.930200</td>\n",
971
- " </tr>\n",
972
- " <tr>\n",
973
- " <td>150</td>\n",
974
- " <td>0.926700</td>\n",
975
- " </tr>\n",
976
- " <tr>\n",
977
- " <td>151</td>\n",
978
- " <td>0.913400</td>\n",
979
- " </tr>\n",
980
- " <tr>\n",
981
- " <td>152</td>\n",
982
- " <td>0.926800</td>\n",
983
- " </tr>\n",
984
- " <tr>\n",
985
- " <td>153</td>\n",
986
- " <td>0.967200</td>\n",
987
- " </tr>\n",
988
- " <tr>\n",
989
- " <td>154</td>\n",
990
- " <td>0.939500</td>\n",
991
- " </tr>\n",
992
- " <tr>\n",
993
- " <td>155</td>\n",
994
- " <td>0.910600</td>\n",
995
- " </tr>\n",
996
- " <tr>\n",
997
- " <td>156</td>\n",
998
- " <td>0.926400</td>\n",
999
- " </tr>\n",
1000
- " <tr>\n",
1001
- " <td>157</td>\n",
1002
- " <td>0.935400</td>\n",
1003
- " </tr>\n",
1004
- " <tr>\n",
1005
- " <td>158</td>\n",
1006
- " <td>0.967700</td>\n",
1007
- " </tr>\n",
1008
- " <tr>\n",
1009
- " <td>159</td>\n",
1010
- " <td>0.899000</td>\n",
1011
- " </tr>\n",
1012
- " <tr>\n",
1013
- " <td>160</td>\n",
1014
- " <td>0.916600</td>\n",
1015
- " </tr>\n",
1016
- " <tr>\n",
1017
- " <td>161</td>\n",
1018
- " <td>0.961600</td>\n",
1019
- " </tr>\n",
1020
- " <tr>\n",
1021
- " <td>162</td>\n",
1022
- " <td>0.898200</td>\n",
1023
- " </tr>\n",
1024
- " <tr>\n",
1025
- " <td>163</td>\n",
1026
- " <td>0.944600</td>\n",
1027
- " </tr>\n",
1028
- " <tr>\n",
1029
- " <td>164</td>\n",
1030
- " <td>0.935700</td>\n",
1031
- " </tr>\n",
1032
- " <tr>\n",
1033
- " <td>165</td>\n",
1034
- " <td>0.922500</td>\n",
1035
- " </tr>\n",
1036
- " <tr>\n",
1037
- " <td>166</td>\n",
1038
- " <td>0.897600</td>\n",
1039
- " </tr>\n",
1040
- " <tr>\n",
1041
- " <td>167</td>\n",
1042
- " <td>0.968600</td>\n",
1043
- " </tr>\n",
1044
- " <tr>\n",
1045
- " <td>168</td>\n",
1046
- " <td>0.927400</td>\n",
1047
- " </tr>\n",
1048
- " <tr>\n",
1049
- " <td>169</td>\n",
1050
- " <td>0.910900</td>\n",
1051
- " </tr>\n",
1052
- " <tr>\n",
1053
- " <td>170</td>\n",
1054
- " <td>0.904700</td>\n",
1055
- " </tr>\n",
1056
- " <tr>\n",
1057
- " <td>171</td>\n",
1058
- " <td>0.899800</td>\n",
1059
- " </tr>\n",
1060
- " <tr>\n",
1061
- " <td>172</td>\n",
1062
- " <td>0.896400</td>\n",
1063
- " </tr>\n",
1064
- " <tr>\n",
1065
- " <td>173</td>\n",
1066
- " <td>0.862100</td>\n",
1067
- " </tr>\n",
1068
- " <tr>\n",
1069
- " <td>174</td>\n",
1070
- " <td>0.909100</td>\n",
1071
- " </tr>\n",
1072
- " <tr>\n",
1073
- " <td>175</td>\n",
1074
- " <td>0.903200</td>\n",
1075
- " </tr>\n",
1076
- " <tr>\n",
1077
- " <td>176</td>\n",
1078
- " <td>0.958600</td>\n",
1079
- " </tr>\n",
1080
- " <tr>\n",
1081
- " <td>177</td>\n",
1082
- " <td>0.902500</td>\n",
1083
- " </tr>\n",
1084
- " <tr>\n",
1085
- " <td>178</td>\n",
1086
- " <td>0.894900</td>\n",
1087
- " </tr>\n",
1088
- " <tr>\n",
1089
- " <td>179</td>\n",
1090
- " <td>0.937900</td>\n",
1091
- " </tr>\n",
1092
- " <tr>\n",
1093
- " <td>180</td>\n",
1094
- " <td>0.900700</td>\n",
1095
- " </tr>\n",
1096
- " <tr>\n",
1097
- " <td>181</td>\n",
1098
- " <td>0.922300</td>\n",
1099
- " </tr>\n",
1100
- " <tr>\n",
1101
- " <td>182</td>\n",
1102
- " <td>0.939300</td>\n",
1103
- " </tr>\n",
1104
- " <tr>\n",
1105
- " <td>183</td>\n",
1106
- " <td>0.932600</td>\n",
1107
- " </tr>\n",
1108
- " <tr>\n",
1109
- " <td>184</td>\n",
1110
- " <td>0.913300</td>\n",
1111
- " </tr>\n",
1112
- " <tr>\n",
1113
- " <td>185</td>\n",
1114
- " <td>0.941700</td>\n",
1115
- " </tr>\n",
1116
- " <tr>\n",
1117
- " <td>186</td>\n",
1118
- " <td>0.886300</td>\n",
1119
- " </tr>\n",
1120
- " <tr>\n",
1121
- " <td>187</td>\n",
1122
- " <td>0.918000</td>\n",
1123
- " </tr>\n",
1124
- " <tr>\n",
1125
- " <td>188</td>\n",
1126
- " <td>0.884000</td>\n",
1127
- " </tr>\n",
1128
- " <tr>\n",
1129
- " <td>189</td>\n",
1130
- " <td>0.947400</td>\n",
1131
- " </tr>\n",
1132
- " <tr>\n",
1133
- " <td>190</td>\n",
1134
- " <td>0.894500</td>\n",
1135
- " </tr>\n",
1136
- " <tr>\n",
1137
- " <td>191</td>\n",
1138
- " <td>0.929300</td>\n",
1139
- " </tr>\n",
1140
- " <tr>\n",
1141
- " <td>192</td>\n",
1142
- " <td>0.877300</td>\n",
1143
- " </tr>\n",
1144
- " <tr>\n",
1145
- " <td>193</td>\n",
1146
- " <td>0.894300</td>\n",
1147
- " </tr>\n",
1148
- " <tr>\n",
1149
- " <td>194</td>\n",
1150
- " <td>0.867800</td>\n",
1151
- " </tr>\n",
1152
- " <tr>\n",
1153
- " <td>195</td>\n",
1154
- " <td>0.913500</td>\n",
1155
- " </tr>\n",
1156
- " <tr>\n",
1157
- " <td>196</td>\n",
1158
- " <td>0.908100</td>\n",
1159
- " </tr>\n",
1160
- " <tr>\n",
1161
- " <td>197</td>\n",
1162
- " <td>0.931200</td>\n",
1163
- " </tr>\n",
1164
- " <tr>\n",
1165
- " <td>198</td>\n",
1166
- " <td>0.911000</td>\n",
1167
- " </tr>\n",
1168
- " <tr>\n",
1169
- " <td>199</td>\n",
1170
- " <td>0.941800</td>\n",
1171
- " </tr>\n",
1172
- " <tr>\n",
1173
- " <td>200</td>\n",
1174
- " <td>0.913000</td>\n",
1175
- " </tr>\n",
1176
- " <tr>\n",
1177
- " <td>201</td>\n",
1178
- " <td>0.921800</td>\n",
1179
- " </tr>\n",
1180
- " <tr>\n",
1181
- " <td>202</td>\n",
1182
- " <td>0.921700</td>\n",
1183
- " </tr>\n",
1184
- " <tr>\n",
1185
- " <td>203</td>\n",
1186
- " <td>0.914500</td>\n",
1187
- " </tr>\n",
1188
- " <tr>\n",
1189
- " <td>204</td>\n",
1190
- " <td>0.910500</td>\n",
1191
- " </tr>\n",
1192
- " <tr>\n",
1193
- " <td>205</td>\n",
1194
- " <td>0.906600</td>\n",
1195
- " </tr>\n",
1196
- " <tr>\n",
1197
- " <td>206</td>\n",
1198
- " <td>0.915100</td>\n",
1199
- " </tr>\n",
1200
- " <tr>\n",
1201
- " <td>207</td>\n",
1202
- " <td>0.881600</td>\n",
1203
- " </tr>\n",
1204
- " <tr>\n",
1205
- " <td>208</td>\n",
1206
- " <td>0.884700</td>\n",
1207
- " </tr>\n",
1208
- " <tr>\n",
1209
- " <td>209</td>\n",
1210
- " <td>0.902900</td>\n",
1211
- " </tr>\n",
1212
- " <tr>\n",
1213
- " <td>210</td>\n",
1214
- " <td>0.882600</td>\n",
1215
- " </tr>\n",
1216
- " <tr>\n",
1217
- " <td>211</td>\n",
1218
- " <td>0.891000</td>\n",
1219
- " </tr>\n",
1220
- " <tr>\n",
1221
- " <td>212</td>\n",
1222
- " <td>0.914400</td>\n",
1223
- " </tr>\n",
1224
- " <tr>\n",
1225
- " <td>213</td>\n",
1226
- " <td>0.930400</td>\n",
1227
- " </tr>\n",
1228
- " <tr>\n",
1229
- " <td>214</td>\n",
1230
- " <td>0.891100</td>\n",
1231
- " </tr>\n",
1232
- " <tr>\n",
1233
- " <td>215</td>\n",
1234
- " <td>0.859300</td>\n",
1235
- " </tr>\n",
1236
- " <tr>\n",
1237
- " <td>216</td>\n",
1238
- " <td>0.891800</td>\n",
1239
- " </tr>\n",
1240
- " <tr>\n",
1241
- " <td>217</td>\n",
1242
- " <td>0.873000</td>\n",
1243
- " </tr>\n",
1244
- " <tr>\n",
1245
- " <td>218</td>\n",
1246
- " <td>0.925900</td>\n",
1247
- " </tr>\n",
1248
- " <tr>\n",
1249
- " <td>219</td>\n",
1250
- " <td>0.905700</td>\n",
1251
- " </tr>\n",
1252
- " <tr>\n",
1253
- " <td>220</td>\n",
1254
- " <td>0.921200</td>\n",
1255
- " </tr>\n",
1256
- " <tr>\n",
1257
- " <td>221</td>\n",
1258
- " <td>0.890200</td>\n",
1259
- " </tr>\n",
1260
- " <tr>\n",
1261
- " <td>222</td>\n",
1262
- " <td>0.915800</td>\n",
1263
- " </tr>\n",
1264
- " <tr>\n",
1265
- " <td>223</td>\n",
1266
- " <td>0.887300</td>\n",
1267
- " </tr>\n",
1268
- " <tr>\n",
1269
- " <td>224</td>\n",
1270
- " <td>0.898300</td>\n",
1271
- " </tr>\n",
1272
- " <tr>\n",
1273
- " <td>225</td>\n",
1274
- " <td>0.865600</td>\n",
1275
- " </tr>\n",
1276
- " <tr>\n",
1277
- " <td>226</td>\n",
1278
- " <td>0.873900</td>\n",
1279
- " </tr>\n",
1280
- " <tr>\n",
1281
- " <td>227</td>\n",
1282
- " <td>0.904800</td>\n",
1283
- " </tr>\n",
1284
- " <tr>\n",
1285
- " <td>228</td>\n",
1286
- " <td>0.917900</td>\n",
1287
- " </tr>\n",
1288
- " <tr>\n",
1289
- " <td>229</td>\n",
1290
- " <td>0.923400</td>\n",
1291
- " </tr>\n",
1292
- " <tr>\n",
1293
- " <td>230</td>\n",
1294
- " <td>0.939700</td>\n",
1295
- " </tr>\n",
1296
- " <tr>\n",
1297
- " <td>231</td>\n",
1298
- " <td>0.913400</td>\n",
1299
- " </tr>\n",
1300
- " <tr>\n",
1301
- " <td>232</td>\n",
1302
- " <td>0.873100</td>\n",
1303
- " </tr>\n",
1304
- " <tr>\n",
1305
- " <td>233</td>\n",
1306
- " <td>0.896700</td>\n",
1307
- " </tr>\n",
1308
- " <tr>\n",
1309
- " <td>234</td>\n",
1310
- " <td>0.892100</td>\n",
1311
- " </tr>\n",
1312
- " <tr>\n",
1313
- " <td>235</td>\n",
1314
- " <td>0.902100</td>\n",
1315
- " </tr>\n",
1316
- " <tr>\n",
1317
- " <td>236</td>\n",
1318
- " <td>0.927200</td>\n",
1319
- " </tr>\n",
1320
- " <tr>\n",
1321
- " <td>237</td>\n",
1322
- " <td>0.912900</td>\n",
1323
- " </tr>\n",
1324
- " <tr>\n",
1325
- " <td>238</td>\n",
1326
- " <td>0.872900</td>\n",
1327
- " </tr>\n",
1328
- " <tr>\n",
1329
- " <td>239</td>\n",
1330
- " <td>0.904700</td>\n",
1331
- " </tr>\n",
1332
- " <tr>\n",
1333
- " <td>240</td>\n",
1334
- " <td>0.879600</td>\n",
1335
- " </tr>\n",
1336
- " <tr>\n",
1337
- " <td>241</td>\n",
1338
- " <td>0.879800</td>\n",
1339
- " </tr>\n",
1340
- " <tr>\n",
1341
- " <td>242</td>\n",
1342
- " <td>0.908800</td>\n",
1343
- " </tr>\n",
1344
- " <tr>\n",
1345
- " <td>243</td>\n",
1346
- " <td>0.909800</td>\n",
1347
- " </tr>\n",
1348
- " <tr>\n",
1349
- " <td>244</td>\n",
1350
- " <td>0.838400</td>\n",
1351
- " </tr>\n",
1352
- " <tr>\n",
1353
- " <td>245</td>\n",
1354
- " <td>0.889200</td>\n",
1355
- " </tr>\n",
1356
- " <tr>\n",
1357
- " <td>246</td>\n",
1358
- " <td>0.912900</td>\n",
1359
- " </tr>\n",
1360
- " <tr>\n",
1361
- " <td>247</td>\n",
1362
- " <td>0.879700</td>\n",
1363
- " </tr>\n",
1364
- " <tr>\n",
1365
- " <td>248</td>\n",
1366
- " <td>0.910700</td>\n",
1367
- " </tr>\n",
1368
- " <tr>\n",
1369
- " <td>249</td>\n",
1370
- " <td>0.845400</td>\n",
1371
- " </tr>\n",
1372
- " <tr>\n",
1373
- " <td>250</td>\n",
1374
- " <td>0.882200</td>\n",
1375
  " </tr>\n",
1376
  " </tbody>\n",
1377
  "</table><p>"
@@ -1382,6 +389,22 @@
1382
  },
1383
  "metadata": {},
1384
  "output_type": "display_data"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1385
  }
1386
  ],
1387
  "source": [
@@ -1392,12 +415,12 @@
1392
  " data_collator=transformers.DataCollatorForLanguageModeling(tokenizer, mlm=False)\n",
1393
  ")\n",
1394
  "trainer.train(resume_from_checkpoint=False)\n",
1395
- "model.save_pretrained('sqllama-out2')"
1396
  ]
1397
  },
1398
  {
1399
  "cell_type": "code",
1400
- "execution_count": 11,
1401
  "metadata": {},
1402
  "outputs": [
1403
  {
 
55
  {
56
  "data": {
57
  "application/vnd.jupyter.widget-view+json": {
58
+ "model_id": "f8ad2d1a5de842bcb6b7e3c6972d9074",
59
  "version_major": 2,
60
  "version_minor": 0
61
  },
 
96
  "output_type": "stream",
97
  "text": [
98
  "\n",
99
+ "table: 2-17672470-19\n",
100
+ "columns: Stage,Winner,General Classification,Mountains Classification,Points Classification,Sprints classification,Team Classification\n",
101
+ "Q: What is the stage of Gerolsteiner?\n",
102
+ "A: SELECT Stage FROM 2-17672470-19 WHERE Team Classification = 'gerolsteiner'\n",
103
  "END\n",
104
  "\n",
105
  "\n",
106
+ "table: 2-12518301-2\n",
107
+ "columns: Rider,Matches,Rides,Bonus Pts,Total Points\n",
108
+ "Q: What was the average number of points with bonus pts less than 31 with the rider dennis gavros?\n",
109
+ "A: SELECT AVG Total Points FROM 2-12518301-2 WHERE Rider = 'dennis gavros' AND Bonus Pts < 31\n",
110
  "END\n",
111
  "\n",
112
  "\n",
113
+ "table: 1-27961684-1\n",
114
+ "columns: Institution,City,State,Team Name,Affiliation,Enrollment,Home Conference\n",
115
+ "Q: How many states were there when there was an enrollment of 2789?\n",
116
+ "A: SELECT COUNT State FROM 1-27961684-1 WHERE Enrollment = 2789\n",
117
  "END\n",
118
  "\n",
119
  "\n",
120
+ "table: 2-17441442-2\n",
121
+ "columns: Res.,Record,Opponent,Method,Event,Round,Time,Location\n",
122
+ "Q: What is the round number when the record is 15–7–1?\n",
123
+ "A: SELECT COUNT Round FROM 2-17441442-2 WHERE Record = '15–7–1'\n",
124
  "END\n",
125
  "\n",
126
  "\n",
127
+ "table: 2-17406982-1\n",
128
+ "columns: Round,Pick,Player,Position,School/Club Team\n",
129
+ "Q: What pick in round 5 did the 49ers pick Jim Pilot?\n",
130
+ "A: SELECT SUM Pick FROM 2-17406982-1 WHERE Player = 'jim pilot' AND Round > 5\n",
131
  "END\n",
132
  "\n"
133
  ]
 
278
  {
279
  "data": {
280
  "application/vnd.jupyter.widget-view+json": {
281
+ "model_id": "708e075933754c6c940eeae9e3d3abc9",
282
  "version_major": 2,
283
  "version_minor": 0
284
  },
 
320
  "LORA_R = 4\n",
321
  "LORA_ALPHA = 16\n",
322
  "LORA_DROPOUT = .1\n",
 
323
  "BATCH = 128\n",
324
  "MICRO_BATCH = 4\n",
325
  "N_GAS = BATCH//MICRO_BATCH\n",
326
+ "EPOCHS = 2\n",
327
+ "LR = 1e-5\n",
328
  "\n",
329
  "lora_cfg = LoraConfig(\n",
330
  " r = LORA_R,\n",
 
344
  " learning_rate=LR,\n",
345
  " fp16=True,\n",
346
  " logging_steps=1,\n",
347
+ " output_dir='sqllama-out3',\n",
348
  " save_total_limit=3,\n",
349
  " remove_unused_columns=False\n",
350
  ")\n"
 
361
  "\n",
362
  " <div>\n",
363
  " \n",
364
+ " <progress value='4' max='500' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
365
+ " [ 4/500 01:51 < 7:38:51, 0.02 it/s, Epoch 0.01/2]\n",
366
  " </div>\n",
367
  " <table border=\"1\" class=\"dataframe\">\n",
368
  " <thead>\n",
 
378
  " </tr>\n",
379
  " <tr>\n",
380
  " <td>2</td>\n",
381
+ " <td>2.725100</td>\n",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
382
  " </tr>\n",
383
  " </tbody>\n",
384
  "</table><p>"
 
389
  },
390
  "metadata": {},
391
  "output_type": "display_data"
392
+ },
393
+ {
394
+ "ename": "KeyboardInterrupt",
395
+ "evalue": "",
396
+ "output_type": "error",
397
+ "traceback": [
398
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
399
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
400
+ "\u001b[0;32m/var/tmp/ipykernel_24178/3667964638.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mdata_collator\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtransformers\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDataCollatorForLanguageModeling\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtokenizer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmlm\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 6\u001b[0m )\n\u001b[0;32m----> 7\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresume_from_checkpoint\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 8\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msave_pretrained\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'sqllama-out3'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
401
+ "\u001b[0;32m~/hf/sqllama-V0/.venv/lib/python3.7/site-packages/transformers/trainer.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, resume_from_checkpoint, trial, ignore_keys_for_eval, **kwargs)\u001b[0m\n\u001b[1;32m 1664\u001b[0m \u001b[0mresume_from_checkpoint\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mresume_from_checkpoint\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1665\u001b[0m \u001b[0mtrial\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtrial\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1666\u001b[0;31m \u001b[0mignore_keys_for_eval\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mignore_keys_for_eval\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1667\u001b[0m )\n\u001b[1;32m 1668\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
402
+ "\u001b[0;32m~/hf/sqllama-V0/.venv/lib/python3.7/site-packages/transformers/trainer.py\u001b[0m in \u001b[0;36m_inner_training_loop\u001b[0;34m(self, batch_size, args, resume_from_checkpoint, trial, ignore_keys_for_eval)\u001b[0m\n\u001b[1;32m 1927\u001b[0m \u001b[0mtr_loss_step\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtraining_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1928\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1929\u001b[0;31m \u001b[0mtr_loss_step\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtraining_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1930\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1931\u001b[0m if (\n",
403
+ "\u001b[0;32m~/hf/sqllama-V0/.venv/lib/python3.7/site-packages/transformers/trainer.py\u001b[0m in \u001b[0;36mtraining_step\u001b[0;34m(self, model, inputs)\u001b[0m\n\u001b[1;32m 2707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2708\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdo_grad_scaling\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2709\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscaler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscale\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2710\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muse_apex\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2711\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mamp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscale_loss\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mscaled_loss\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
404
+ "\u001b[0;32m~/hf/sqllama-V0/.venv/lib/python3.7/site-packages/torch/_tensor.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[1;32m 487\u001b[0m )\n\u001b[1;32m 488\u001b[0m torch.autograd.backward(\n\u001b[0;32m--> 489\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgradient\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 490\u001b[0m )\n\u001b[1;32m 491\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
405
+ "\u001b[0;32m~/hf/sqllama-V0/.venv/lib/python3.7/site-packages/torch/autograd/__init__.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[1;32m 197\u001b[0m Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass\n\u001b[1;32m 198\u001b[0m \u001b[0mtensors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgrad_tensors_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 199\u001b[0;31m allow_unreachable=True, accumulate_grad=True) # Calls into the C++ engine to run the backward pass\n\u001b[0m\u001b[1;32m 200\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 201\u001b[0m def grad(\n",
406
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
407
+ ]
408
  }
409
  ],
410
  "source": [
 
415
  " data_collator=transformers.DataCollatorForLanguageModeling(tokenizer, mlm=False)\n",
416
  ")\n",
417
  "trainer.train(resume_from_checkpoint=False)\n",
418
+ "model.save_pretrained('sqllama-out3')"
419
  ]
420
  },
421
  {
422
  "cell_type": "code",
423
+ "execution_count": null,
424
  "metadata": {},
425
  "outputs": [
426
  {