File size: 139,049 Bytes
fdd8012
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "f7gr5j8iDEva"
      },
      "outputs": [],
      "source": [
        "!pip install sentence_transformers openai unstructured\n",
        "!pip install plotly\n",
        "!pip install langchain\n",
        "!pip install tiktoken\n",
        "!pip install matplotlib\n",
        "%pip install -Uqqq rich openai tiktoken wandb langchain unstructured tabulate pdf2image chromadb gradio faiss-gpu\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eCDxucPBGp5g",
        "outputId": "b7c353ee-c376-4fb7-f700-961e3dd56781"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: linkify-it-py in /usr/local/lib/python3.10/dist-packages (2.0.2)\n",
            "Requirement already satisfied: uc-micro-py in /usr/local/lib/python3.10/dist-packages (from linkify-it-py) (1.0.2)\n"
          ]
        }
      ],
      "source": [
        "!pip install linkify-it-py"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "u4zmuvSp7WVI"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import os, random\n",
        "from pathlib import Path\n",
        "import tiktoken\n",
        "from getpass import getpass\n",
        "from rich.markdown import Markdown\n",
        "from langchain.chains.qa_with_sources import load_qa_with_sources_chain"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EVxVIwh87WVJ",
        "outputId": "51abb5db-4535-4362-c94e-ef93e9c1485a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "131072"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ],
      "source": [
        "import torch\n",
        "import sys\n",
        "import csv\n",
        "\n",
        "csv.field_size_limit(sys.maxsize)\n",
        "\n",
        "# torch.has_mps"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_1AdkW1kFB20"
      },
      "outputs": [],
      "source": [
        "# os.environ[\"OPENAI_API_KEY\"] = getpass(\"\")\n",
        "#os.environ[\"OPENAI_API_KEY\"] = “API KEY”"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5SuSwwcx7WVK",
        "outputId": "7a126930-89db-4f6e-ee91-7f7300e79457"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "··········\n",
            "OpenAI API key configured\n"
          ]
        }
      ],
      "source": [
        "\n",
        "if os.getenv(\"OPENAI_API_KEY\") is None:\n",
        "  if any(['VSCODE' in x for x in os.environ.keys()]):\n",
        "    print('Please enter password in the VS Code prompt at the top of your VS Code window!')\n",
        "  os.environ[\"OPENAI_API_KEY\"] = getpass(\"\")\n",
        "\n",
        "assert os.getenv(\"OPENAI_API_KEY\", \"\").startswith(\"sk-\"), \"This doesn't look like a valid OpenAI API key\"\n",
        "print(\"OpenAI API key configured\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6fLvsQcLEW0K"
      },
      "source": [
        "#Chunking and RAG"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Qgn9BzjX7WVK",
        "outputId": "3d37d4b9-846f-4539-b46e-56837544dd7d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n"
          ]
        }
      ],
      "source": [
        "from langchain.embeddings.openai import OpenAIEmbeddings\n",
        "import nltk\n",
        "nltk.download('punkt')\n",
        "from langchain.text_splitter import CharacterTextSplitter\n",
        "from langchain.text_splitter import RecursiveCharacterTextSplitter\n",
        "from langchain.vectorstores import Chroma\n",
        "from langchain.document_loaders import TextLoader\n",
        "from langchain.chat_models import ChatOpenAI\n",
        "from langchain import PromptTemplate\n",
        "from langchain.chains import LLMChain\n",
        "from langchain.chains.qa_with_sources import load_qa_with_sources_chain\n",
        "from langchain.llms import OpenAI\n",
        "from langchain.vectorstores import FAISS"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "cgh0f__69Qse"
      },
      "outputs": [],
      "source": [
        "MODEL_NAME = \"text-embedding-3-small\"\n",
        "# MODEL_NAME = \"gpt-4\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "28-76KUp9TdA"
      },
      "outputs": [],
      "source": [
        "# from langchain.document_loaders import DirectoryLoader\n",
        "\n",
        "# def find_md_files(directory):\n",
        "#     \"Find all markdown files in a directory and return a LangChain Document\"\n",
        "#     dl = DirectoryLoader(directory, \"**/*.txt\")\n",
        "#     return dl.load()\n",
        "\n",
        "# documents = find_md_files('data')\n",
        "# len(documents)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "MLiFbXx1-hAv"
      },
      "outputs": [],
      "source": [
        "# documents"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FznW3HMpD4f_"
      },
      "source": [
        "#upload files"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wtENHSIWEf9K"
      },
      "outputs": [],
      "source": [
        "import pandas as pd"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "c3Edx6229xs3"
      },
      "outputs": [],
      "source": [
        "from langchain.document_loaders.csv_loader import CSVLoader\n",
        "# from langchain.docstore.document import Document\n",
        "from langchain.text_splitter import RecursiveCharacterTextSplitter\n",
        "from langchain.document_loaders import DirectoryLoader"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "url='https://drive.google.com/file/d/1gl7WAkJr6Nyke7YckzXxdL-iM4UjhLGX/view?usp=sharing'\n",
        "url='https://drive.google.com/uc?id=' + url.split('/')[-2]\n",
        "df = pd.read_csv(url)"
      ],
      "metadata": {
        "id": "logMHfgiSen2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 556
        },
        "id": "1jr0ZaWqvSPQ",
        "outputId": "b9c06fdc-5d0b-44ce-f1c0-3a8bee0028f2"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                        _id  \\\n",
              "0  63927168098523fa9d5ef047   \n",
              "1  652b21ec754471a9795c291a   \n",
              "2  6529cfaff59ed61396db838c   \n",
              "3  6529d030f59ed61396db8398   \n",
              "4  6529d072f59ed61396db839e   \n",
              "\n",
              "                                               title           author  \\\n",
              "0  Are The Democrats Screwed In The Senate After ...      Nate Silver   \n",
              "1                               Beyond the Narrative     Sam Freedman   \n",
              "2                         Ballot Measures: A Preview     Walter Olson   \n",
              "3           Trump’s Only Real Worldview Is Pettiness  David A. Graham   \n",
              "4  The COVID Bailout of State and Local Governmen...       Eric Boehm   \n",
              "\n",
              "                                         description  publish_date  \\\n",
              "0  \\nAre The Democrats Screwed In The Senate Afte...  1.670516e+12   \n",
              "1                                                NaN  1.697272e+12   \n",
              "2                                                NaN  1.697227e+12   \n",
              "3                                                NaN  1.697214e+12   \n",
              "4                                                NaN  1.697212e+12   \n",
              "\n",
              "                                           full_text  \\\n",
              "0  \\n\\n\\n\\n2028 Election\\nAre The Democrats Screw...   \n",
              "1  We are told before every party conference that...   \n",
              "2  Walter Olson  Voters will go to the polls soon...   \n",
              "3  Let no one say that Donald Trump has lost his ...   \n",
              "4         Two years after Congress authorized a h...   \n",
              "\n",
              "                                                 url categories  \n",
              "0  https://fivethirtyeight.com/features/democrats...   ['news']  \n",
              "1   https://samf.substack.com/p/beyond-the-narrative   ['news']  \n",
              "2  https://www.cato.org/blog/ballot-measures-preview   ['news']  \n",
              "3                 tag:theatlantic.com,2023:50-675637   ['news']  \n",
              "4  https://reason.com/2023/10/13/the-covid-bailou...   ['news']  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7017ff2e-f2a2-4b31-b909-64ce1b83cc12\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>_id</th>\n",
              "      <th>title</th>\n",
              "      <th>author</th>\n",
              "      <th>description</th>\n",
              "      <th>publish_date</th>\n",
              "      <th>full_text</th>\n",
              "      <th>url</th>\n",
              "      <th>categories</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>63927168098523fa9d5ef047</td>\n",
              "      <td>Are The Democrats Screwed In The Senate After ...</td>\n",
              "      <td>Nate Silver</td>\n",
              "      <td>\\nAre The Democrats Screwed In The Senate Afte...</td>\n",
              "      <td>1.670516e+12</td>\n",
              "      <td>\\n\\n\\n\\n2028 Election\\nAre The Democrats Screw...</td>\n",
              "      <td>https://fivethirtyeight.com/features/democrats...</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>652b21ec754471a9795c291a</td>\n",
              "      <td>Beyond the Narrative</td>\n",
              "      <td>Sam Freedman</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697272e+12</td>\n",
              "      <td>We are told before every party conference that...</td>\n",
              "      <td>https://samf.substack.com/p/beyond-the-narrative</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>6529cfaff59ed61396db838c</td>\n",
              "      <td>Ballot Measures: A Preview</td>\n",
              "      <td>Walter Olson</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697227e+12</td>\n",
              "      <td>Walter Olson  Voters will go to the polls soon...</td>\n",
              "      <td>https://www.cato.org/blog/ballot-measures-preview</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6529d030f59ed61396db8398</td>\n",
              "      <td>Trump’s Only Real Worldview Is Pettiness</td>\n",
              "      <td>David A. Graham</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697214e+12</td>\n",
              "      <td>Let no one say that Donald Trump has lost his ...</td>\n",
              "      <td>tag:theatlantic.com,2023:50-675637</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6529d072f59ed61396db839e</td>\n",
              "      <td>The COVID Bailout of State and Local Governmen...</td>\n",
              "      <td>Eric Boehm</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697212e+12</td>\n",
              "      <td>Two years after Congress authorized a h...</td>\n",
              "      <td>https://reason.com/2023/10/13/the-covid-bailou...</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7017ff2e-f2a2-4b31-b909-64ce1b83cc12')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-7017ff2e-f2a2-4b31-b909-64ce1b83cc12 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-7017ff2e-f2a2-4b31-b909-64ce1b83cc12');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-34e447ed-e678-4955-984c-06d42221f5ad\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-34e447ed-e678-4955-984c-06d42221f5ad')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-34e447ed-e678-4955-984c-06d42221f5ad button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l-Rjki2Yvq01",
        "outputId": "68972d9a-1c04-483c-93d1-ac65e3d065ec"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3327, 8)"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ],
      "source": [
        "df.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lA4AoU3f4dJH"
      },
      "outputs": [],
      "source": [
        "!mkdir data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nsvUBcPJCV_m"
      },
      "outputs": [],
      "source": [
        "df[:100].to_csv('data/df_embed.csv',index=False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Kwy9kXLzYEu7"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. Step 1: Read the data"
      ],
      "metadata": {
        "id": "5WYH_AC18qes"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 157
        },
        "id": "T6LiGVc4LREH",
        "outputId": "819bda32-1735-4f6b-9ab5-d6696d641082"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'We are told before every party conference that the leader will be making perhaps their “most important speech yet”. In reality, unless they are going into the conference with the word “beleaguered” regularly appearing in front of their name, they’ll almost certainly be fine. For struggling leaders a bad speech might sink them, but if they are 20 points ahead in the polls, and coming off a huge by-election win, then they are unlikely to be derailed.It’s even less likely to be a problem if your opponents are noisily falling apart. I wasn’t at the Tory conference but all of the several dozen journalists, politicians and business people I spoke to who had, were unanimous that it was even more shambolic than it appeared on TV. Even hacks from right-wing papers were shaking their heads at the sheer incoherence of it all. It’s hard to be a client journalist when your political masters can’t figure out what they want you to say.You can see how the Tories got into such a mess. There is no record to run on and Sunak doesn’t want to take the blame for the failures of his predecessors. It is, though, always going to be extremely hard to run a “change” message as the governing party. Boris Johnson managed it in 2019 by promising to spend a lot of money and attaching his personal brand to Brexit. But Sunak doesn’t want to spend money and any hope associated with leaving the EU has faded. He has nothing else. A rag-bag of random policy announcements about A-levels and smoking certainly won’t cut it.Even worse than the confusion of the official platform was the circus around it. There was a general consensus, amongst attendees, that it felt more like the conference a party has after it’s lost an election rather than the one before, with all the focus on leadership jockeying and indulging in ideological fantasies.So the lobby arrived at the Labour conference – which I did go to – primed to write positively about the contrast. As long as Starmer and team could look vaguely professional and serious, then they were going to get good coverage. The narrative is in their favour and that makes politics a lot easier. Announcements that would get ripped to pieces if the lobby were moving in for a kill get left unscrutinised. Dull speeches are lauded as triumphs. You’ve got to enjoy it while it lasts. \\xa0\\xa0I don’t want to be churlish about this. Seeming professional and serious might not seem like a high bar but it’s one plenty of Labour conferences have failed to meet in the past. We shouldn’t begrudge weary Labour folk their raptures at a leadership that can tie their own shoelaces. Starmer’s team are competent, well organised, extremely thorough in their voter research and message design, and ruthless when it comes to keeping the hard left out. As a Jew I was highly conscious of how different the conference would have felt last week with the old guard in charge, and don’t underestimate what it took to achieve that transformation.But Labour still haven’t begun to resolve, or even acknowledge, the fundamental contradiction in their current position.Let’s be clear as to what the problem is. It’s not that Starmer “doesn’t have a vision” or “doesn’t stand for anything”. Labour’s “missions” might be a poor campaigning tool (as predicted) but if you bother to read them they’re entirely clear about the country they want to build. And that country is Denmark. Literally in the case of the “opportunity mission” which cites Danish social mobility as the (implausible) goal. Starmer and his team are soft left Labour and want a high-growth, social democratic country with better worker rights, more state direction, and less inequality. They want us to be Scandinavian. Which as visions go seems reasonable enough.Nor is the problem that “they don’t have any policies”. They have far too many fiddly policies. No one who doesn’t spend their time reading policy documents and going to conferences will have heard of them. Did you know Labour is planning a £2.4k retention bonus for teachers completing the early years framework? I doubt it. Their bigger pledges on housebuilding, planning, and electricity infrastructure are substantive and important, albeit lacking in detail. But that will always be true of opposition parties, who have limited resources. The civil service is much maligned but you do need them to do the work on these complex policies.No, the problem is very specific. They have no substantive policies that would involve having to spend any taxpayer money. Starmer’s holding position that he wishes to run “a reforming state, not a cheque-book state” is transparent nonsense, an example of what Duncan Robinson at the Economist calls “Reform Fairy” thinking:“Both main parties agree that although the British state requires a total overhaul, it does not need much more cash. In its bid to move away from the Magic Money Tree, British politics has fallen under the spell of another mythical being: the Reform Fairy. The Magic Money Tree could generate cash at will; the Reform Fairy can apparently improve public services without spending political or financial capital.”I am always surprised at politicians’ capacity to believe things that happen to be convenient for them at any given time, but I just don’t buy they really believe in the Reform Fairy. For a start when it comes to health, welfare and education – the three departments that account for 60% of all spending – they have no proposals for significant reform. I watched a succession of junior shadow ministers struggle through fringe events on each of these issues, acknowledging the terrible state of things right now, promising a decade of transformation, but unable to fill in any of the steps that might get us there. \\xa0 \\xa0The reasons for this are obvious. They are absolutely determined to stop the Tories running a “same old Labour” campaign, or at least not one that seems credible. From an electoral point of view it will work because neither the Tories nor the right-wing press are going to attack them for not pledging to spend enough money. Plus the power of the narrative will protect them from scrutiny. On top of that, it’s now so clear they’re going to win that every media outlet, business, trade union and sector lobbying group wants to be on good terms.It is, though, going to make those first few years in government much harder because they can’t do any serious thinking about reform that goes beyond the very conceptual stage, without getting into costings.That doesn’t mean I left the conference without hope. The first big positive was that Rachel Reeves’ team are doing an extremely canny job at shouting loudly about fiscal discipline while ensuring they keep their room for manoeuvre open post-election. There were hints that investment spending could be excluded from Labour’s headline rule on debt reduction (as it was in Boris Johnson’s first two years as Prime Minister). Reeves may have an “iron-clad commitment” to sticking to fiscal rules but there is still definitional space about what exactly they would be. She also managed to get through conference without ruling out any other tax cuts. My biggest worry for Labour is that at some point over the next year their poll lead dips, the narrative shifts a bit (as journalists get bored) and they get panicked into reducing their fiscal space to zero.My second positive is that Labour adjacent wonks, who have plausible deniability, are starting to think through some ways through these problems. The centre-left think-tank IPPR published a report that was subtitled, in a direct response to Duncan Robinson’s challenge, “Between the Reform Fairy and the Magic Money Tree”. It acknowledges that spending will be needed, and the short-termist scorecard approach taken by the Treasury is a major impediment to a well-functioning state. Also that the standard “new public management” model of public service reform – with its focus entirely on targets and extrinsic incentives – has run out of road with no obvious alternative offering itself.I don’t agree with all their solutions. For instance, they propose creating a new category of “prevention investment expenditure” to ensure money designed for long-term preventative interventions on, say public health or reducing reoffending,\\xa0is better protected. I get the objective but the fuzziness of the line over what counts as prevention worries me. I do, though, like their idea of treating priority social policy objectives in the same way as fiscal ones, with independent measurement as to whether, for example, the government is reducing waiting lists or building more houses. Tying the Treasury into those metrics could help balance the obsession with the spending scorecard. It is also good to see Labour engaging with these ideas. So there are positive signs, but not yet quite the sense of urgency I’d like to see. While, of course, Labour are aware the country is in a bad way, I’m not sure they yet appreciate quite how much of a challenge it’s going to be. Just in the last few weeks we’ve heard that dozens of councils are approaching bankruptcy; that prisons are now full with those convicted of serious crimes unable to be incarcerated; and that hospitals are already declaring critical incidents before we get to winter. There is still a sense they are over-estimating how far being nicer and more competent that Tory ministers will get them. Its good to know, for instance, \\xa0that they won’t go ahead with the Rwanda deportations, as this will save time and money. But it’s not going to be nearly enough.The biggest panel event I spoke at was an online one organised by Citizens’ Advice on the cost of living crisis. Their stats, on foodbank use, homelessness and so on, are all dire. I was most struck, though, by the story of Sanna. A young woman living in a council flat who was already behind on her bills and then lost just under a third of her Universal Credit for a month, because she missed a job centre appointment and was sanctioned. This will push her further into debt. She is already struggling with anxiety and depression, which will now be exacerbated, making it harder to find work. I wouldn’t be surprised if she ended up joining the ballooning list of people classified as unable to work due to mental health problems. It was such a clear example of how our current approach to our biggest spending policy area is utterly broken – ruining lives and costing taxpayer money – with no party close to offering a solution.\\xa0We need that sense of urgency for Sanna’s sake.Comment is Freed is a reader supported publication. A monthly subscription is £3.50 and an annual one £35. It includes at least four subscriber only posts a month.This post is public so please share if you found it helpful.Share'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 18
        }
      ],
      "source": [
        "df.full_text\t[1]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 556
        },
        "id": "QV58YlrRYKIS",
        "outputId": "8dcd3fa1-420b-415e-ae7a-94714771cf72"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                        _id  \\\n",
              "0  63927168098523fa9d5ef047   \n",
              "1  652b21ec754471a9795c291a   \n",
              "2  6529cfaff59ed61396db838c   \n",
              "3  6529d030f59ed61396db8398   \n",
              "4  6529d072f59ed61396db839e   \n",
              "\n",
              "                                               title           author  \\\n",
              "0  Are The Democrats Screwed In The Senate After ...      Nate Silver   \n",
              "1                               Beyond the Narrative     Sam Freedman   \n",
              "2                         Ballot Measures: A Preview     Walter Olson   \n",
              "3           Trump’s Only Real Worldview Is Pettiness  David A. Graham   \n",
              "4  The COVID Bailout of State and Local Governmen...       Eric Boehm   \n",
              "\n",
              "                                         description  publish_date  \\\n",
              "0  \\nAre The Democrats Screwed In The Senate Afte...  1.670516e+12   \n",
              "1                                                NaN  1.697272e+12   \n",
              "2                                                NaN  1.697227e+12   \n",
              "3                                                NaN  1.697214e+12   \n",
              "4                                                NaN  1.697212e+12   \n",
              "\n",
              "                                           full_text  \\\n",
              "0  \\n\\n\\n\\n2028 Election\\nAre The Democrats Screw...   \n",
              "1  We are told before every party conference that...   \n",
              "2  Walter Olson  Voters will go to the polls soon...   \n",
              "3  Let no one say that Donald Trump has lost his ...   \n",
              "4         Two years after Congress authorized a h...   \n",
              "\n",
              "                                                 url categories  \n",
              "0  https://fivethirtyeight.com/features/democrats...   ['news']  \n",
              "1   https://samf.substack.com/p/beyond-the-narrative   ['news']  \n",
              "2  https://www.cato.org/blog/ballot-measures-preview   ['news']  \n",
              "3                 tag:theatlantic.com,2023:50-675637   ['news']  \n",
              "4  https://reason.com/2023/10/13/the-covid-bailou...   ['news']  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e0435ba1-01e3-4008-9219-1accd86c5d0a\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>_id</th>\n",
              "      <th>title</th>\n",
              "      <th>author</th>\n",
              "      <th>description</th>\n",
              "      <th>publish_date</th>\n",
              "      <th>full_text</th>\n",
              "      <th>url</th>\n",
              "      <th>categories</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>63927168098523fa9d5ef047</td>\n",
              "      <td>Are The Democrats Screwed In The Senate After ...</td>\n",
              "      <td>Nate Silver</td>\n",
              "      <td>\\nAre The Democrats Screwed In The Senate Afte...</td>\n",
              "      <td>1.670516e+12</td>\n",
              "      <td>\\n\\n\\n\\n2028 Election\\nAre The Democrats Screw...</td>\n",
              "      <td>https://fivethirtyeight.com/features/democrats...</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>652b21ec754471a9795c291a</td>\n",
              "      <td>Beyond the Narrative</td>\n",
              "      <td>Sam Freedman</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697272e+12</td>\n",
              "      <td>We are told before every party conference that...</td>\n",
              "      <td>https://samf.substack.com/p/beyond-the-narrative</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>6529cfaff59ed61396db838c</td>\n",
              "      <td>Ballot Measures: A Preview</td>\n",
              "      <td>Walter Olson</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697227e+12</td>\n",
              "      <td>Walter Olson  Voters will go to the polls soon...</td>\n",
              "      <td>https://www.cato.org/blog/ballot-measures-preview</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6529d030f59ed61396db8398</td>\n",
              "      <td>Trump’s Only Real Worldview Is Pettiness</td>\n",
              "      <td>David A. Graham</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697214e+12</td>\n",
              "      <td>Let no one say that Donald Trump has lost his ...</td>\n",
              "      <td>tag:theatlantic.com,2023:50-675637</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>6529d072f59ed61396db839e</td>\n",
              "      <td>The COVID Bailout of State and Local Governmen...</td>\n",
              "      <td>Eric Boehm</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.697212e+12</td>\n",
              "      <td>Two years after Congress authorized a h...</td>\n",
              "      <td>https://reason.com/2023/10/13/the-covid-bailou...</td>\n",
              "      <td>['news']</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e0435ba1-01e3-4008-9219-1accd86c5d0a')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-e0435ba1-01e3-4008-9219-1accd86c5d0a button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-e0435ba1-01e3-4008-9219-1accd86c5d0a');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-67a834e8-3806-4d57-822a-370719726ac5\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-67a834e8-3806-4d57-822a-370719726ac5')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-67a834e8-3806-4d57-822a-370719726ac5 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QSOiVXVxekH_"
      },
      "outputs": [],
      "source": [
        "path= 'data/df_embed.csv'\n",
        "loader = CSVLoader(file_path=path,source_column=\"title\")\n",
        "\n",
        "data = loader.load()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q2dWBLgyEy_n",
        "outputId": "3c5b0017-13ac-4420-b787-89971f592925"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "page_content=\"_id: 63927168098523fa9d5ef047\\ntitle: Are The Democrats Screwed In The Senate After 2024?\\nauthor: Nate Silver\\ndescription: Are The Democrats Screwed In The Senate After 2024?\\nNo, but the party faces an uphill battlee\\n\\n\\nI’m not that impressed by long-term projections of Democratic doom in the Senate. Mostly that’s because I’m not that impressed by long-term political projections in general. Political coalitions change; not many people would have had it on their bingo card that Democrats would obtain a Senate majority in 2020 by winning two runoffs in Georgia, and then would actually expand on that majority in 2022 following another runoff in Georgia despite their party controlling the White House.\\nWith that said, while I don’t think we can say very much about what politics will look like 10 or 20 years from now, a medium-term look-ahead is possible. Do Democrats have a reasonable hope of regaining a trifecta — control of the House, Senate and presidency — at any point in the near future (2024, 2026 or 2028)? And how long will they hold onto control of the Senate and the presidency, allowing them to appoint Supreme Court justices without Republican help?\\nMy colleague Geoffrey Skelley already took an initial look at the 2024 Senate map, which is bad for Democrats, but I wanted to dig slightly deeper and also consider 2026 and 2028. Like Geoffrey, I’ll be focused on the Senate, because that’s the weakest link in Democrats’ effort to win back the trifecta. Despite some disadvantages in the Electoral College, Democrats obviously have no trouble winning the presidency. And they nearly held onto the House this year despite losing the popular vote for the House. The current House map doesn’t have much — if any — of a Republican skew, and Democrats should be able to win back the House if the national political environment is reasonably good for them.\\nSo let’s look at the 2024, 2026 and 2028 Senate maps (yes, 2028 features the same map as the midterm we just finished; I just can’t let go, I guess). We’ll classify races into bronze, silver, gold and platinum “tiers” in terms of how feasible they are as pickup opportunities. I’m deliberately using this goofy metallurgical theme as opposed to the categories FiveThirtyEight typically uses (e.g., “toss-up,” “lean Republican”) so you’ll recognize these as being back-of-the-envelope suppositions rather than any sort of official race ratings. \\nIn general, these ratings lean heavily on looking at how close a race was the last time it was contested, with some subjective adjustments for candidate quality. I’m also considering how the electoral environment has shifted within the state — Florida has gotten redder since Rick Scott was elected in 2018, for instance — although it is definitely not always safe to assume a state will continue trending in the same direction.\\n2024\\nRepublican pickup opportunities\\nPlatinum tier: West Virginia (Manchin), Montana (Tester), Ohio (Brown), \\nGold tier: Arizona (Sinema), Nevada (Rosen)\\nSilver tier: Wisconsin (Baldwin), Michigan (Stabenow), Pennsylvania (Casey)\\nBronze tier: New Jersey (Menendez), Virginia (Kaine)\\nAs Geoffrey noted, the platinum-tier states listed here are the Democrats’ main problem. They have three senators up for reelection in states that Donald Trump carried by 8 (Ohio), 16 (Montana) and 39 (West Virginia!) percentage points in 2020. Each of those Democratic senators won reelection in 2018, of course, but that was a strong Democratic year, and there’s been further movement of white voters without college degrees — a key part of the electorate in these states — toward Republicans since then.\\xa0\\nI don’t want to focus too much on any one individual race, but it’s probably safe to assume that Joe Manchin has the toughest reelection campaign of all. Meanwhile, maybe Sherrod Brown can feel a bit more comfortable following Tim Ryan’s relatively good performance in Ohio this year, but Ryan did lose, and Brown may face a tougher opponent than Ryan did in J.D. Vance. Democrats also run the risk of retirements here; neither Manchin nor Jon Tester has officially yet announced they will seek reelection.\\nBeyond the platinum tier, Democrats face some other risks. Kyrsten Sinema is likely to face a serious primary challenge. Democrats keep winning close races for Congress in Nevada but it’s a very purple state. Tammy Baldwin, Debbie Stabenow and Bob Casey won by fairly comfortable margins in 2018 and will likely be fine in the event of a decent-to-good Democratic year, but if 2024 turns into a good Republican year, they could be in trouble.\\nDemocratic pickup opportunities\\nPlatinum tier: None\\nGold tier: None\\nSilver tier: Texas (Cruz), Florida (Scott)\\nBronze tier: None\\nConversely, it’s very slim pickings for Democrats. On paper, you’d think that Texas might be close — Ted Cruz won by less than 3 percentage points there in 2018 and Texas is more competitive than it once was — but the erosion of the Democratic vote in South Texas has prevented Texas from turning truly purple. And although Scott won by only 0.12 percentage points in 2018, Republicans did so well in Florida in 2020 and 2022 that I was tempted to demote his race all the way to the bronze tier. Plus, there’s a good chance that the GOP’s 2024 nominee will be either Florida Gov. Ron DeSantis or Florida resident Donald Trump, which could help drive Republican turnout.\\nBeyond that, any other races look like an extreme stretch for Democrats. Maybe Josh Hawley in Missouri could be vulnerable in a 2018-type environment, but that’s less likely in a presidential year and Hawley’s approval rating isn’t that bad. Put him in the “honorable mention” tier, I guess.\\nGiven all this, I’d imagine the modal outcome from 2024 is something like Republicans picking up about three Senate seats. There’s uncertainty around that; it’s surely within the realm of possibility that Democrats hold the Senate, but it’s also possible that Republicans could wind up with 54 or 55 seats. And of course, all of these races are correlated. If the GOP is doing well enough to win the presidency in 2024, they should have no trouble also winning the Senate.9\\n2026\\nRepublican pickup opportunities\\xa0\\nPlatinum tier: None\\nGold tier: Georgia (Ossoff), Michigan (Peters)\\nSilver tier: New Hampshire (Shaheen)\\nBronze tier: Minnesota (Smith), Virginia (Warner)\\nThe 2026 map is more balanced. Democrats undoubtedly feel pretty good about how they’ve performed in Georgia and Michigan in the past two cycles, but Jon Ossoff and Gary Peters won by extremely narrow margins in 2020 and it would be a mistake to assume they’re safe. It’s not a particularly deep set of pickup opportunities for Republicans, though if Democrats retain the presidency in 2024, you’d expect Republicans to have good midterms in 2026, and New Hampshire and perhaps even Minnesota and Virginia could potentially be in play.\\xa0\\nDemocratic pickup opportunities\\xa0\\nPlatinum tier: None\\nGold tier: Maine (Collins), North Carolina (Tillis)\\nSilver tier: Alaska (Sullivan)\\nBronze tier: Texas (Cornyn), Iowa (Ernst)\\nMeanwhile, Democrats will try again in the two states that were most disappointing to them in 2020: Maine and North Carolina. Given all the focus on Georgia, I wonder if North Carolina hasn’t become underrated as a place where Democrats could gain ground in the future; the states are fairly similar demographically and it wouldn’t take that much of a shift for Democrats to go from narrowly losing races in North Carolina to narrowly winning them. Susan Collins will be 74 years old in 2026, meanwhile, an age at which there’s typically some chance of retirement.\\nDemocrats also have a below-the-radar pickup opportunity in Alaska, where Mary Peltola was elected to a full term in the U.S. House10 last month by a 10-point margin after ranked-choice votes were tabulated. GOP incumbent Dan Sullivan has a mediocre approval rating and a Peltola-Sullivan race would potentially be competitive.\\nSuppose, though, that Democrats win the presidency but lose the Senate in 2024. It’s really not as crazy as it sounds. If every state votes identically to how it did in the 2020 presidential election — and every Senate race follows the presidential vote — then Democrats would come out of 2024 with the presidency but only 48 Senate seats after losing West Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from the GOP in the 2026 midterm while controlling the presidency? Unlikely — but then again, Democrats gaining Senate seats this year seemed unlikely and they did it.\\n2028\\nRepublican pickup opportunities\\nPlatinum tier: None\\nGold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), Georgia (Warnock) \\nSilver tier: New Hampshire (Hassan)\\nBronze tier: Oregon (Wyden)\\nAll right, now we’re literally coming full circle to consider the races that were just contested in last month’s midterm. So we’ll keep it pretty brief. Yes, Democrats have to feel pretty good about their wins in Nevada, Pennsylvania, Arizona and Georgia — especially given that the overall political environment wasn’t that great for Democrats this year. But they were narrow wins, and (with the possible exception of Nevada) they came against a mediocre set of GOP opponents. Maybe some of these races are toward the silver end of the gold tier, but more likely than not they’ll be competitive again.\\nAnd since we’re projecting a full six years out, I’m going to be a little bit more creative about considering long-shot opportunities in the bronze tier. Although the Senate races in Washington and Colorado received attention this year (Republicans spent a lot of money on both, only to lose by double digits), they wound up coming just as close in Oregon with a QAnon candidate as part of a comparatively strong year for the GOP in the Beaver State. Could a more mainstream GOP nominee have made it a race? Oregon is quite white, it has fewer people with college degrees than the “Portlandia” stereotype might make you assume, and Democratic incumbent Ron Wyden will be 79 years old in 2028 …\\nDemocratic pickup opportunities\\nPlatinum tier: None\\nGold tier: Wisconsin (Johnson), North Carolina (Budd)\\nSilver tier: Vance (Ohio)\\nBronze tier: Alaska (Murkowski), Florida (Rubio), Utah (Lee), Iowa (Grassley)\\nAgain, we’re mostly just rehashing this year’s map. Ron Johnson, who originally hadn’t planned on running in 2020, is another retirement threat in 2026 — and he barely won reelection anyway. In Ohio, Sherrod Brown could seek to return to the Senate if he loses in 2024 by challenging J.D. Vance. And 2028 is far enough from now that I wouldn’t want to rule out Democrats being competitive in states as far-flung as Utah, where independent Evan McMullin ran a fairly competitive race against Mike Lee this year.\\n\\nEven with an additional senator going into 2023, the 2024 map is still so bad for Democrats that keeping the Senate for years to come will be a fairly tough order. The party’s prospects might rest more upon limiting the damage in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a bad 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.\\nThat bleak picture may shape the next few years of political maneuvering. When Vox’s Dylan Matthews suggested on Twitter that liberal Justices Sonia Sotomayor (age 68) and Elena Kagan (age 62) should retire while Democrats have their Senate majority and be replaced by younger justices, it didn’t go over well. But it’s a perfectly rational suggestion if Democrats don’t feel like gambling with their judicial future. (Consider how consequential Ruth Bader Ginsburg’s decision not to retire has been for liberals.) Democrats have a narrow path to Senate control after 2024, but it’s narrow indeed, and one that might require the GOP continuing to nominate bad candidates — and a fair share of luck.\\npublish_date: 1670515535000.0\\nfull_text: 2028 Election\\nAre The Democrats Screwed In The Senate After 2024?\\nNo, but the party faces an uphill battle.\\n\\n\\n\\nBy Nate Silver\\n\\n\\nDec. 8, 2022, at 11:05 AM\\t\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\nSamuel Corum / Getty Images\\n\\n\\n\\n\\nI’m not that impressed by long-term projections of Democratic doom in the Senate. Mostly that’s because I’m not that impressed by long-term political projections in general. Political coalitions change; not many people would have had it on their bingo card that Democrats would obtain a Senate majority in 2020 by winning two runoffs in Georgia, and then would actually expand on that majority in 2022 following another runoff in Georgia despite their party controlling the White House.\\nWith that said, while I don’t think we can say very much about what politics will look like 10 or 20 years from now, a medium-term look-ahead is possible. Do Democrats have a reasonable hope of regaining a trifecta — control of the House, Senate and presidency — at any point in the near future (2024, 2026 or 2028)? And how long will they hold onto control of the Senate and the presidency, allowing them to appoint Supreme Court justices without Republican help?\\nMy colleague Geoffrey Skelley already took an initial look at the 2024 Senate map, which is bad for Democrats, but I wanted to dig slightly deeper and also consider 2026 and 2028. Like Geoffrey, I’ll be focused on the Senate, because that’s the weakest link in Democrats’ effort to win back the trifecta. Despite some disadvantages in the Electoral College, Democrats obviously have no trouble winning the presidency. And they nearly held onto the House this year despite losing the popular vote for the House. The current House map doesn’t have much — if any — of a Republican skew, and Democrats should be able to win back the House if the national political environment is reasonably good for them.\\nSo let’s look at the 2024, 2026 and 2028 Senate maps (yes, 2028 features the same map as the midterm we just finished; I just can’t let go, I guess). We’ll classify races into bronze, silver, gold and platinum “tiers” in terms of how feasible they are as pickup opportunities. I’m deliberately using this goofy metallurgical theme as opposed to the categories FiveThirtyEight typically uses (e.g., “toss-up,” “lean Republican”) so you’ll recognize these as being back-of-the-envelope suppositions rather than any sort of official race ratings. \\nIn general, these ratings lean heavily on looking at how close a race was the last time it was contested, with some subjective adjustments for candidate quality. I’m also considering how the electoral environment has shifted within the state — Florida has gotten redder since Rick Scott was elected in 2018, for instance — although it is definitely not always safe to assume a state will continue trending in the same direction.\\n2024\\nRepublican pickup opportunities\\nPlatinum tier: West Virginia (Manchin), Montana (Tester), Ohio (Brown), \\nGold tier: Arizona (Sinema), Nevada (Rosen)\\nSilver tier: Wisconsin (Baldwin), Michigan (Stabenow), Pennsylvania (Casey)\\nBronze tier: New Jersey (Menendez), Virginia (Kaine)\\nAs Geoffrey noted, the platinum-tier states listed here are the Democrats’ main problem. They have three senators up for reelection in states that Donald Trump carried by 8 (Ohio), 16 (Montana) and 39 (West Virginia!) percentage points in 2020. Each of those Democratic senators won reelection in 2018, of course, but that was a strong Democratic year, and there’s been further movement of white voters without college degrees — a key part of the electorate in these states — toward Republicans since then.\\xa0\\nI don’t want to focus too much on any one individual race, but it’s probably safe to assume that Joe Manchin has the toughest reelection campaign of all. Meanwhile, maybe Sherrod Brown can feel a bit more comfortable following Tim Ryan’s relatively good performance in Ohio this year, but Ryan did lose, and Brown may face a tougher opponent than Ryan did in J.D. Vance. Democrats also run the risk of retirements here; neither Manchin nor Jon Tester has officially yet announced they will seek reelection.\\nBeyond the platinum tier, Democrats face some other risks. Kyrsten Sinema is likely to face a serious primary challenge. Democrats keep winning close races for Congress in Nevada but it’s a very purple state. Tammy Baldwin, Debbie Stabenow and Bob Casey won by fairly comfortable margins in 2018 and will likely be fine in the event of a decent-to-good Democratic year, but if 2024 turns into a good Republican year, they could be in trouble.\\nDemocratic pickup opportunities\\nPlatinum tier: None\\nGold tier: None\\nSilver tier: Texas (Cruz), Florida (Scott)\\nBronze tier: None\\nConversely, it’s very slim pickings for Democrats. On paper, you’d think that Texas might be close — Ted Cruz won by less than 3 percentage points there in 2018 and Texas is more competitive than it once was — but the erosion of the Democratic vote in South Texas has prevented Texas from turning truly purple. And although Scott won by only 0.12 percentage points in 2018, Republicans did so well in Florida in 2020 and 2022 that I was tempted to demote his race all the way to the bronze tier. Plus, there’s a good chance that the GOP’s 2024 nominee will be either Florida Gov. Ron DeSantis or Florida resident Donald Trump, which could help drive Republican turnout.\\nBeyond that, any other races look like an extreme stretch for Democrats. Maybe Josh Hawley in Missouri could be vulnerable in a 2018-type environment, but that’s less likely in a presidential year and Hawley’s approval rating isn’t that bad. Put him in the “honorable mention” tier, I guess.\\nGiven all this, I’d imagine the modal outcome from 2024 is something like Republicans picking up about three Senate seats. There’s uncertainty around that; it’s surely within the realm of possibility that Democrats hold the Senate, but it’s also possible that Republicans could wind up with 54 or 55 seats. And of course, all of these races are correlated. If the GOP is doing well enough to win the presidency in 2024, they should have no trouble also winning the Senate.9\\n2026\\nRepublican pickup opportunities\\xa0\\nPlatinum tier: None\\nGold tier: Georgia (Ossoff), Michigan (Peters)\\nSilver tier: New Hampshire (Shaheen)\\nBronze tier: Minnesota (Smith), Virginia (Warner)\\nThe 2026 map is more balanced. Democrats undoubtedly feel pretty good about how they’ve performed in Georgia and Michigan in the past two cycles, but Jon Ossoff and Gary Peters won by extremely narrow margins in 2020 and it would be a mistake to assume they’re safe. It’s not a particularly deep set of pickup opportunities for Republicans, though if Democrats retain the presidency in 2024, you’d expect Republicans to have good midterms in 2026, and New Hampshire and perhaps even Minnesota and Virginia could potentially be in play.\\xa0\\nDemocratic pickup opportunities\\xa0\\nPlatinum tier: None\\nGold tier: Maine (Collins), North Carolina (Tillis)\\nSilver tier: Alaska (Sullivan)\\nBronze tier: Texas (Cornyn), Iowa (Ernst)\\nMeanwhile, Democrats will try again in the two states that were most disappointing to them in 2020: Maine and North Carolina. Given all the focus on Georgia, I wonder if North Carolina hasn’t become underrated as a place where Democrats could gain ground in the future; the states are fairly similar demographically and it wouldn’t take that much of a shift for Democrats to go from narrowly losing races in North Carolina to narrowly winning them. Susan Collins will be 74 years old in 2026, meanwhile, an age at which there’s typically some chance of retirement.\\nDemocrats also have a below-the-radar pickup opportunity in Alaska, where Mary Peltola was elected to a full term in the U.S. House10 last month by a 10-point margin after ranked-choice votes were tabulated. GOP incumbent Dan Sullivan has a mediocre approval rating and a Peltola-Sullivan race would potentially be competitive.\\nSuppose, though, that Democrats win the presidency but lose the Senate in 2024. It’s really not as crazy as it sounds. If every state votes identically to how it did in the 2020 presidential election — and every Senate race follows the presidential vote — then Democrats would come out of 2024 with the presidency but only 48 Senate seats after losing West Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from the GOP in the 2026 midterm while controlling the presidency? Unlikely — but then again, Democrats gaining Senate seats this year seemed unlikely and they did it.\\n2028\\nRepublican pickup opportunities\\nPlatinum tier: None\\nGold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), Georgia (Warnock) \\nSilver tier: New Hampshire (Hassan)\\nBronze tier: Oregon (Wyden)\\nAll right, now we’re literally coming full circle to consider the races that were just contested in last month’s midterm. So we’ll keep it pretty brief. Yes, Democrats have to feel pretty good about their wins in Nevada, Pennsylvania, Arizona and Georgia — especially given that the overall political environment wasn’t that great for Democrats this year. But they were narrow wins, and (with the possible exception of Nevada) they came against a mediocre set of GOP opponents. Maybe some of these races are toward the silver end of the gold tier, but more likely than not they’ll be competitive again.\\nAnd since we’re projecting a full six years out, I’m going to be a little bit more creative about considering long-shot opportunities in the bronze tier. Although the Senate races in Washington and Colorado received attention this year (Republicans spent a lot of money on both, only to lose by double digits), they wound up coming just as close in Oregon with a QAnon candidate as part of a comparatively strong year for the GOP in the Beaver State. Could a more mainstream GOP nominee have made it a race? Oregon is quite white, it has fewer people with college degrees than the “Portlandia” stereotype might make you assume, and Democratic incumbent Ron Wyden will be 79 years old in 2028 …\\nDemocratic pickup opportunities\\nPlatinum tier: None\\nGold tier: Wisconsin (Johnson), North Carolina (Budd)\\nSilver tier: Vance (Ohio)\\nBronze tier: Alaska (Murkowski), Florida (Rubio), Utah (Lee), Iowa (Grassley)\\nAgain, we’re mostly just rehashing this year’s map. Ron Johnson, who originally hadn’t planned on running in 2020, is another retirement threat in 2026 — and he barely won reelection anyway. In Ohio, Sherrod Brown could seek to return to the Senate if he loses in 2024 by challenging J.D. Vance. And 2028 is far enough from now that I wouldn’t want to rule out Democrats being competitive in states as far-flung as Utah, where independent Evan McMullin ran a fairly competitive race against Mike Lee this year.\\n\\nEven with an additional senator going into 2023, the 2024 map is still so bad for Democrats that keeping the Senate for years to come will be a fairly tough order. The party’s prospects might rest more upon limiting the damage in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a bad 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.\\nThat bleak picture may shape the next few years of political maneuvering. When Vox’s Dylan Matthews suggested on Twitter that liberal Justices Sonia Sotomayor (age 68) and Elena Kagan (age 62) should retire while Democrats have their Senate majority and be replaced by younger justices, it didn’t go over well. But it’s a perfectly rational suggestion if Democrats don’t feel like gambling with their judicial future. (Consider how consequential Ruth Bader Ginsburg’s decision not to retire has been for liberals.) Democrats have a narrow path to Senate control after 2024, but it’s narrow indeed, and one that might require the GOP continuing to nominate bad candidates — and a fair share of luck.\\nurl: https://fivethirtyeight.com/features/democrats-senate-chances-2024-and-beyond/\\ncategories: ['news']\" metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}\n"
          ]
        }
      ],
      "source": [
        "print(data[0])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9OlS2Kx8FlVx",
        "outputId": "8e891267-6921-46ad-961c-ddd0a2d33760"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "You have 100 document(s) in your data\n",
            "There are 23887 words in your document\n"
          ]
        }
      ],
      "source": [
        "print (f'You have {len(data)} document(s) in your data')\n",
        "print (f'There are {len(data[0].page_content)} words in your document')"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "2. Breakdown data - Chunking Process"
      ],
      "metadata": {
        "id": "Vm0lO--f80NZ"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FOUcWZF3Gk8n"
      },
      "outputs": [],
      "source": [
        "text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)\n",
        "texts = text_splitter.split_documents(data)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xdJDpvcBGnp3",
        "outputId": "3c129195-32d6-4df8-b0d2-1e49a74f1609"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1192"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ],
      "source": [
        "len(texts)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jaRsK0qaZ9Fk",
        "outputId": "91bf63db-9e03-4db0-8506-f4303d5fe782"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "You now have 1192 document(s) in your data\n",
            "There are 951 characters in your document\n"
          ]
        }
      ],
      "source": [
        "print (f'You now have {len(texts)} document(s) in your data')\n",
        "print (f'There are {len(texts[1].page_content)} characters in your document')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qaFGDzkjMRJG",
        "outputId": "9249d2ba-1070-4de0-df11-0dfc1a7c8160"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Document(page_content='Suppose, though, that Democrats win the presidency but lose the Senate in 2024. It’s really not as crazy as it sounds. If every state votes identically to how it did in the 2020 presidential election — and every Senate race follows the presidential vote — then Democrats would come out of 2024 with the presidency but only 48 Senate seats after losing West Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from the GOP in the 2026 midterm while controlling the presidency? Unlikely — but then again, Democrats gaining Senate seats this year seemed unlikely and they did it.\\n2028\\nRepublican pickup opportunities\\nPlatinum tier: None\\nGold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), Georgia (Warnock) \\nSilver tier: New Hampshire (Hassan)\\nBronze tier: Oregon (Wyden)', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0})"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "texts[10]"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -U langchain-openai\n",
        "from langchain_openai import OpenAIEmbeddings\n",
        "from langchain_openai import ChatOpenAI"
      ],
      "metadata": {
        "id": "O-acByeSuScf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d6deb120-c4a2-4b21-b4cb-536c783c774f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting langchain-openai\n",
            "  Downloading langchain_openai-0.0.5-py3-none-any.whl (29 kB)\n",
            "Requirement already satisfied: langchain-core<0.2,>=0.1.16 in /usr/local/lib/python3.10/dist-packages (from langchain-openai) (0.1.18)\n",
            "Requirement already satisfied: numpy<2,>=1 in /usr/local/lib/python3.10/dist-packages (from langchain-openai) (1.23.5)\n",
            "Requirement already satisfied: openai<2.0.0,>=1.10.0 in /usr/local/lib/python3.10/dist-packages (from langchain-openai) (1.11.0)\n",
            "Requirement already satisfied: tiktoken<0.6.0,>=0.5.2 in /usr/local/lib/python3.10/dist-packages (from langchain-openai) (0.5.2)\n",
            "Requirement already satisfied: PyYAML>=5.3 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (6.0.1)\n",
            "Requirement already satisfied: anyio<5,>=3 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (3.7.1)\n",
            "Requirement already satisfied: jsonpatch<2.0,>=1.33 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (1.33)\n",
            "Requirement already satisfied: langsmith<0.1,>=0.0.83 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (0.0.86)\n",
            "Requirement already satisfied: packaging<24.0,>=23.2 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (23.2)\n",
            "Requirement already satisfied: pydantic<3,>=1 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (2.6.0)\n",
            "Requirement already satisfied: requests<3,>=2 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (2.31.0)\n",
            "Requirement already satisfied: tenacity<9.0.0,>=8.1.0 in /usr/local/lib/python3.10/dist-packages (from langchain-core<0.2,>=0.1.16->langchain-openai) (8.2.3)\n",
            "Requirement already satisfied: distro<2,>=1.7.0 in /usr/lib/python3/dist-packages (from openai<2.0.0,>=1.10.0->langchain-openai) (1.7.0)\n",
            "Requirement already satisfied: httpx<1,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from openai<2.0.0,>=1.10.0->langchain-openai) (0.26.0)\n",
            "Requirement already satisfied: sniffio in /usr/local/lib/python3.10/dist-packages (from openai<2.0.0,>=1.10.0->langchain-openai) (1.3.0)\n",
            "Requirement already satisfied: tqdm>4 in /usr/local/lib/python3.10/dist-packages (from openai<2.0.0,>=1.10.0->langchain-openai) (4.66.1)\n",
            "Requirement already satisfied: typing-extensions<5,>=4.7 in /usr/local/lib/python3.10/dist-packages (from openai<2.0.0,>=1.10.0->langchain-openai) (4.9.0)\n",
            "Requirement already satisfied: regex>=2022.1.18 in /usr/local/lib/python3.10/dist-packages (from tiktoken<0.6.0,>=0.5.2->langchain-openai) (2023.12.25)\n",
            "Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.10/dist-packages (from anyio<5,>=3->langchain-core<0.2,>=0.1.16->langchain-openai) (3.6)\n",
            "Requirement already satisfied: exceptiongroup in /usr/local/lib/python3.10/dist-packages (from anyio<5,>=3->langchain-core<0.2,>=0.1.16->langchain-openai) (1.2.0)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.10/dist-packages (from httpx<1,>=0.23.0->openai<2.0.0,>=1.10.0->langchain-openai) (2023.11.17)\n",
            "Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.10/dist-packages (from httpx<1,>=0.23.0->openai<2.0.0,>=1.10.0->langchain-openai) (1.0.2)\n",
            "Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.10/dist-packages (from httpcore==1.*->httpx<1,>=0.23.0->openai<2.0.0,>=1.10.0->langchain-openai) (0.14.0)\n",
            "Requirement already satisfied: jsonpointer>=1.9 in /usr/local/lib/python3.10/dist-packages (from jsonpatch<2.0,>=1.33->langchain-core<0.2,>=0.1.16->langchain-openai) (2.4)\n",
            "Requirement already satisfied: annotated-types>=0.4.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1->langchain-core<0.2,>=0.1.16->langchain-openai) (0.6.0)\n",
            "Requirement already satisfied: pydantic-core==2.16.1 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1->langchain-core<0.2,>=0.1.16->langchain-openai) (2.16.1)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2->langchain-core<0.2,>=0.1.16->langchain-openai) (3.3.2)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2->langchain-core<0.2,>=0.1.16->langchain-openai) (2.0.7)\n",
            "Installing collected packages: langchain-openai\n",
            "Successfully installed langchain-openai-0.0.5\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ZV4m9sie7WVV"
      },
      "outputs": [],
      "source": [
        "embeddings = OpenAIEmbeddings()\n",
        "db = FAISS.from_documents(texts, embeddings)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "AMkkNB3U7WVV"
      },
      "outputs": [],
      "source": [
        "query = \"\"\"Why is it so bad for Democrats in Senate in 2024\"\"\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tCaQyBkz7WVV"
      },
      "outputs": [],
      "source": [
        "def get_response_from_query(db, query, k=10):\n",
        "    \"\"\"\n",
        "\n",
        "    \"\"\"\n",
        "\n",
        "    docs = db.similarity_search(query, k=k) # extremely important\n",
        "\n",
        "    docs_page_content = \" \".join([d.page_content for d in docs])\n",
        "\n",
        "    # llm = BardLLM()\n",
        "    llm = ChatOpenAI(model_name=\"gpt-3.5-turbo-16k\",temperature=0)\n",
        "\n",
        "    prompt = PromptTemplate(\n",
        "        input_variables=[\"question\", \"docs\"],\n",
        "        template=\"\"\"\n",
        "        You are a bot that is open to discussions about different cultural, philosophical and political exchanges. I will use do different analysis to the articles provided to me. Stay truthful and if you weren't provided any resources give your oppinion only.\n",
        "        Answer the following question: {question}\n",
        "        By searching the following articles: {docs}\n",
        "\n",
        "        Only use the factual information from the documents. Make sure to mention key phrases from the articles.\n",
        "\n",
        "        If you feel like you don't have enough information to answer the question, say \"I don't know\".\n",
        "\n",
        "        \"\"\",\n",
        "    )\n",
        "\n",
        "    chain = LLMChain(llm=llm, prompt=prompt)\n",
        "    # chain = RetrievalQAWithSourcesChain.from_chain_type(llm=llm, prompt=prompt,\n",
        "    #                                                     chain_type=\"stuff\", retriever=db.as_retriever(), return_source_documents=True)\n",
        "\n",
        "    response = chain.run(question=query, docs=docs_page_content,return_source_documents=True)\n",
        "    r_text = str(response)\n",
        "\n",
        "    ##evaluation part\n",
        "\n",
        "    prompt_eval = PromptTemplate(\n",
        "        input_variables=[\"answer\", \"docs\"],\n",
        "        template=\"\"\"\n",
        "        You job is to evaluate if the response to a given context is faithful.\n",
        "\n",
        "        for the following: {answer}\n",
        "        By searching the following article: {docs}\n",
        "\n",
        "       Give a reason why they are similar or not, start with a Yes or a No.\n",
        "\n",
        "        \"\"\",\n",
        "    )\n",
        "\n",
        "    chain_part_2 = LLMChain(llm=llm, prompt=prompt_eval)\n",
        "\n",
        "    evals = chain_part_2.run(answer=r_text, docs=docs_page_content)\n",
        "\n",
        "\n",
        "    return response,docs,evals"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "twZ_ILhJ7WVV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8256686e-896c-40a0-aa72-e0c5cf4f2b2d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/langchain_core/_api/deprecation.py:117: LangChainDeprecationWarning: The function `run` was deprecated in LangChain 0.1.0 and will be removed in 0.2.0. Use invoke instead.\n",
            "  warn_deprecated(\n"
          ]
        }
      ],
      "source": [
        "answer,sources,evals=get_response_from_query(db,query,15)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Xaa84xr109s5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e0205d3c-a717-4d0f-97cb-b084e167a058"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Document(page_content='Even with an additional senator going into 2023, the 2024 map is still so bad for Democrats that keeping the Senate for years to come will be a fairly tough order. The party’s prospects might rest more upon limiting the damage in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a bad 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='Even with an additional senator going into 2023, the 2024 map is still so bad for Democrats that keeping the Senate for years to come will be a fairly tough order. The party’s prospects might rest more upon limiting the damage in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a bad 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='My colleague Geoffrey Skelley already took an initial look at the 2024 Senate map, which is bad for Democrats, but I wanted to dig slightly deeper and also consider 2026 and 2028. Like Geoffrey, I’ll be focused on the Senate, because that’s the weakest link in Democrats’ effort to win back the trifecta. Despite some disadvantages in the Electoral College, Democrats obviously have no trouble winning the presidency. And they nearly held onto the House this year despite losing the popular vote for the House. The current House map doesn’t have much — if any — of a Republican skew, and Democrats should be able to win back the House if the national political environment is reasonably good for them.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='My colleague Geoffrey Skelley already took an initial look at the 2024 Senate map, which is bad for Democrats, but I wanted to dig slightly deeper and also consider 2026 and 2028. Like Geoffrey, I’ll be focused on the Senate, because that’s the weakest link in Democrats’ effort to win back the trifecta. Despite some disadvantages in the Electoral College, Democrats obviously have no trouble winning the presidency. And they nearly held onto the House this year despite losing the popular vote for the House. The current House map doesn’t have much — if any — of a Republican skew, and Democrats should be able to win back the House if the national political environment is reasonably good for them.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='That bleak picture may shape the next few years of political maneuvering. When Vox’s Dylan Matthews suggested on Twitter that liberal Justices Sonia Sotomayor (age 68) and Elena Kagan (age 62) should retire while Democrats have their Senate majority and be replaced by younger justices, it didn’t go over well. But it’s a perfectly rational suggestion if Democrats don’t feel like gambling with their judicial future. (Consider how consequential Ruth Bader Ginsburg’s decision not to retire has been for liberals.) Democrats have a narrow path to Senate control after 2024, but it’s narrow indeed, and one that might require the GOP continuing to nominate bad candidates — and a fair share of luck.\\npublish_date: 1670515535000.0\\nfull_text: 2028 Election\\nAre The Democrats Screwed In The Senate After 2024?\\nNo, but the party faces an uphill battle.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='Suppose, though, that Democrats win the presidency but lose the Senate in 2024. It’s really not as crazy as it sounds. If every state votes identically to how it did in the 2020 presidential election — and every Senate race follows the presidential vote — then Democrats would come out of 2024 with the presidency but only 48 Senate seats after losing West Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from the GOP in the 2026 midterm while controlling the presidency? Unlikely — but then again, Democrats gaining Senate seats this year seemed unlikely and they did it.\\n2028\\nRepublican pickup opportunities\\nPlatinum tier: None\\nGold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), Georgia (Warnock) \\nSilver tier: New Hampshire (Hassan)\\nBronze tier: Oregon (Wyden)', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='Suppose, though, that Democrats win the presidency but lose the Senate in 2024. It’s really not as crazy as it sounds. If every state votes identically to how it did in the 2020 presidential election — and every Senate race follows the presidential vote — then Democrats would come out of 2024 with the presidency but only 48 Senate seats after losing West Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from the GOP in the 2026 midterm while controlling the presidency? Unlikely — but then again, Democrats gaining Senate seats this year seemed unlikely and they did it.\\n2028\\nRepublican pickup opportunities\\nPlatinum tier: None\\nGold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), Georgia (Warnock) \\nSilver tier: New Hampshire (Hassan)\\nBronze tier: Oregon (Wyden)', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='Beyond that, any other races look like an extreme stretch for Democrats. Maybe Josh Hawley in Missouri could be vulnerable in a 2018-type environment, but that’s less likely in a presidential year and Hawley’s approval rating isn’t that bad. Put him in the “honorable mention” tier, I guess.\\nGiven all this, I’d imagine the modal outcome from 2024 is something like Republicans picking up about three Senate seats. There’s uncertainty around that; it’s surely within the realm of possibility that Democrats hold the Senate, but it’s also possible that Republicans could wind up with 54 or 55 seats. And of course, all of these races are correlated. If the GOP is doing well enough to win the presidency in 2024, they should have no trouble also winning the Senate.9\\n2026\\nRepublican pickup opportunities\\xa0\\nPlatinum tier: None\\nGold tier: Georgia (Ossoff), Michigan (Peters)\\nSilver tier: New Hampshire (Shaheen)\\nBronze tier: Minnesota (Smith), Virginia (Warner)', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='Beyond that, any other races look like an extreme stretch for Democrats. Maybe Josh Hawley in Missouri could be vulnerable in a 2018-type environment, but that’s less likely in a presidential year and Hawley’s approval rating isn’t that bad. Put him in the “honorable mention” tier, I guess.\\nGiven all this, I’d imagine the modal outcome from 2024 is something like Republicans picking up about three Senate seats. There’s uncertainty around that; it’s surely within the realm of possibility that Democrats hold the Senate, but it’s also possible that Republicans could wind up with 54 or 55 seats. And of course, all of these races are correlated. If the GOP is doing well enough to win the presidency in 2024, they should have no trouble also winning the Senate.9\\n2026\\nRepublican pickup opportunities\\xa0\\nPlatinum tier: None\\nGold tier: Georgia (Ossoff), Michigan (Peters)\\nSilver tier: New Hampshire (Shaheen)\\nBronze tier: Minnesota (Smith), Virginia (Warner)', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='I’m not that impressed by long-term projections of Democratic doom in the Senate. Mostly that’s because I’m not that impressed by long-term political projections in general. Political coalitions change; not many people would have had it on their bingo card that Democrats would obtain a Senate majority in 2020 by winning two runoffs in Georgia, and then would actually expand on that majority in 2022 following another runoff in Georgia despite their party controlling the White House.\\nWith that said, while I don’t think we can say very much about what politics will look like 10 or 20 years from now, a medium-term look-ahead is possible. Do Democrats have a reasonable hope of regaining a trifecta — control of the House, Senate and presidency — at any point in the near future (2024, 2026 or 2028)? And how long will they hold onto control of the Senate and the presidency, allowing them to appoint Supreme Court justices without Republican help?', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='I’m not that impressed by long-term projections of Democratic doom in the Senate. Mostly that’s because I’m not that impressed by long-term political projections in general. Political coalitions change; not many people would have had it on their bingo card that Democrats would obtain a Senate majority in 2020 by winning two runoffs in Georgia, and then would actually expand on that majority in 2022 following another runoff in Georgia despite their party controlling the White House.\\nWith that said, while I don’t think we can say very much about what politics will look like 10 or 20 years from now, a medium-term look-ahead is possible. Do Democrats have a reasonable hope of regaining a trifecta — control of the House, Senate and presidency — at any point in the near future (2024, 2026 or 2028)? And how long will they hold onto control of the Senate and the presidency, allowing them to appoint Supreme Court justices without Republican help?', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content=\"That bleak picture may shape the next few years of political maneuvering. When Vox’s Dylan Matthews suggested on Twitter that liberal Justices Sonia Sotomayor (age 68) and Elena Kagan (age 62) should retire while Democrats have their Senate majority and be replaced by younger justices, it didn’t go over well. But it’s a perfectly rational suggestion if Democrats don’t feel like gambling with their judicial future. (Consider how consequential Ruth Bader Ginsburg’s decision not to retire has been for liberals.) Democrats have a narrow path to Senate control after 2024, but it’s narrow indeed, and one that might require the GOP continuing to nominate bad candidates — and a fair share of luck.\\nurl: https://fivethirtyeight.com/features/democrats-senate-chances-2024-and-beyond/\\ncategories: ['news']\", metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='_id: 63927168098523fa9d5ef047\\ntitle: Are The Democrats Screwed In The Senate After 2024?\\nauthor: Nate Silver\\ndescription: Are The Democrats Screwed In The Senate After 2024?\\nNo, but the party faces an uphill battlee', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='I don’t want to focus too much on any one individual race, but it’s probably safe to assume that Joe Manchin has the toughest reelection campaign of all. Meanwhile, maybe Sherrod Brown can feel a bit more comfortable following Tim Ryan’s relatively good performance in Ohio this year, but Ryan did lose, and Brown may face a tougher opponent than Ryan did in J.D. Vance. Democrats also run the risk of retirements here; neither Manchin nor Jon Tester has officially yet announced they will seek reelection.\\nBeyond the platinum tier, Democrats face some other risks. Kyrsten Sinema is likely to face a serious primary challenge. Democrats keep winning close races for Congress in Nevada but it’s a very purple state. Tammy Baldwin, Debbie Stabenow and Bob Casey won by fairly comfortable margins in 2018 and will likely be fine in the event of a decent-to-good Democratic year, but if 2024 turns into a good Republican year, they could be in trouble.\\nDemocratic pickup opportunities', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}),\n",
              " Document(page_content='I don’t want to focus too much on any one individual race, but it’s probably safe to assume that Joe Manchin has the toughest reelection campaign of all. Meanwhile, maybe Sherrod Brown can feel a bit more comfortable following Tim Ryan’s relatively good performance in Ohio this year, but Ryan did lose, and Brown may face a tougher opponent than Ryan did in J.D. Vance. Democrats also run the risk of retirements here; neither Manchin nor Jon Tester has officially yet announced they will seek reelection.\\nBeyond the platinum tier, Democrats face some other risks. Kyrsten Sinema is likely to face a serious primary challenge. Democrats keep winning close races for Congress in Nevada but it’s a very purple state. Tammy Baldwin, Debbie Stabenow and Bob Casey won by fairly comfortable margins in 2018 and will likely be fine in the event of a decent-to-good Democratic year, but if 2024 turns into a good Republican year, they could be in trouble.\\nDemocratic pickup opportunities', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0})]"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ],
      "source": [
        "sources"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K7P30pL57WVV",
        "outputId": "928ac23c-0111-4520-ea8c-e1cbbff9e803"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "> Question:\n",
            "'Why is it so bad for Democrats in Senate in 2024'\n",
            "\n",
            "> Answer:\n",
            "('Based on the articles provided, it is suggested that the 2024 Senate map is '\n",
            " 'unfavorable for Democrats, making it difficult for them to maintain control '\n",
            " 'of the Senate. The articles mention that Democrats may face challenges in '\n",
            " 'the 2024 Senate races, potentially losing seats in states like West '\n",
            " 'Virginia, Montana, and Ohio. It is also mentioned that Democrats have a '\n",
            " 'narrow path to Senate control after 2024 and that their prospects might rely '\n",
            " 'on limiting the damage in 2024 to have a chance of regaining the Senate in '\n",
            " '2026 or 2028. The articles highlight the importance of the GOP nominating '\n",
            " 'bad candidates and luck for Democrats to have a better chance of regaining '\n",
            " 'the Senate. Overall, the articles suggest that the 2024 Senate map poses '\n",
            " 'challenges for Democrats and that it may be difficult for them to regain '\n",
            " 'control of the Senate before 2030 or 2032.')\n",
            "\n",
            "> Eval:\n",
            "('Yes, the response is faithful to the context. The response accurately '\n",
            " 'reflects the information provided in the articles about the challenges '\n",
            " 'Democrats may face in the 2024 Senate races and their narrow path to Senate '\n",
            " 'control after 2024. It also mentions the importance of the GOP nominating '\n",
            " 'bad candidates and luck for Democrats to have a better chance of regaining '\n",
            " 'the Senate.')\n",
            "----------------------------------SOURCE DOCUMENTS---------------------------\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('Even with an additional senator going into 2023, the 2024 map is still so '\n",
            " 'bad for Democrats that keeping the Senate for years to come will be a fairly '\n",
            " 'tough order. The party’s prospects might rest more upon limiting the damage '\n",
            " 'in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a '\n",
            " 'bad 2024 could make it very difficult for Democrats to regain the Senate '\n",
            " 'before 2030 or 2032.')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('Even with an additional senator going into 2023, the 2024 map is still so '\n",
            " 'bad for Democrats that keeping the Senate for years to come will be a fairly '\n",
            " 'tough order. The party’s prospects might rest more upon limiting the damage '\n",
            " 'in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a '\n",
            " 'bad 2024 could make it very difficult for Democrats to regain the Senate '\n",
            " 'before 2030 or 2032.')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('My colleague Geoffrey Skelley already took an initial look at the 2024 '\n",
            " 'Senate map, which is bad for Democrats, but I wanted to dig slightly deeper '\n",
            " 'and also consider 2026 and 2028. Like Geoffrey, I’ll be focused on the '\n",
            " 'Senate, because that’s the weakest link in Democrats’ effort to win back the '\n",
            " 'trifecta. Despite some disadvantages in the Electoral College, Democrats '\n",
            " 'obviously have no trouble winning the presidency. And they nearly held onto '\n",
            " 'the House this year despite losing the popular vote for the House. The '\n",
            " 'current House map doesn’t have much — if any — of a Republican skew, and '\n",
            " 'Democrats should be able to win back the House if the national political '\n",
            " 'environment is reasonably good for them.')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('My colleague Geoffrey Skelley already took an initial look at the 2024 '\n",
            " 'Senate map, which is bad for Democrats, but I wanted to dig slightly deeper '\n",
            " 'and also consider 2026 and 2028. Like Geoffrey, I’ll be focused on the '\n",
            " 'Senate, because that’s the weakest link in Democrats’ effort to win back the '\n",
            " 'trifecta. Despite some disadvantages in the Electoral College, Democrats '\n",
            " 'obviously have no trouble winning the presidency. And they nearly held onto '\n",
            " 'the House this year despite losing the popular vote for the House. The '\n",
            " 'current House map doesn’t have much — if any — of a Republican skew, and '\n",
            " 'Democrats should be able to win back the House if the national political '\n",
            " 'environment is reasonably good for them.')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('That bleak picture may shape the next few years of political maneuvering. '\n",
            " 'When Vox’s Dylan Matthews suggested on Twitter that liberal Justices Sonia '\n",
            " 'Sotomayor (age 68) and Elena Kagan (age 62) should retire while Democrats '\n",
            " 'have their Senate majority and be replaced by younger justices, it didn’t go '\n",
            " 'over well. But it’s a perfectly rational suggestion if Democrats don’t feel '\n",
            " 'like gambling with their judicial future. (Consider how consequential Ruth '\n",
            " 'Bader Ginsburg’s decision not to retire has been for liberals.) Democrats '\n",
            " 'have a narrow path to Senate control after 2024, but it’s narrow indeed, and '\n",
            " 'one that might require the GOP continuing to nominate bad candidates — and a '\n",
            " 'fair share of luck.\\n'\n",
            " 'publish_date: 1670515535000.0\\n'\n",
            " 'full_text: 2028 Election\\n'\n",
            " 'Are The Democrats Screwed In The Senate After 2024?\\n'\n",
            " 'No, but the party faces an uphill battle.')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('Suppose, though, that Democrats win the presidency but lose the Senate in '\n",
            " '2024. It’s really not as crazy as it sounds. If every state votes '\n",
            " 'identically to how it did in the 2020 presidential election — and every '\n",
            " 'Senate race follows the presidential vote — then Democrats would come out of '\n",
            " '2024 with the presidency but only 48 Senate seats after losing West '\n",
            " 'Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from '\n",
            " 'the GOP in the 2026 midterm while controlling the presidency? Unlikely — but '\n",
            " 'then again, Democrats gaining Senate seats this year seemed unlikely and '\n",
            " 'they did it.\\n'\n",
            " '2028\\n'\n",
            " 'Republican pickup opportunities\\n'\n",
            " 'Platinum tier: None\\n'\n",
            " 'Gold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), '\n",
            " 'Georgia (Warnock) \\n'\n",
            " 'Silver tier: New Hampshire (Hassan)\\n'\n",
            " 'Bronze tier: Oregon (Wyden)')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('Suppose, though, that Democrats win the presidency but lose the Senate in '\n",
            " '2024. It’s really not as crazy as it sounds. If every state votes '\n",
            " 'identically to how it did in the 2020 presidential election — and every '\n",
            " 'Senate race follows the presidential vote — then Democrats would come out of '\n",
            " '2024 with the presidency but only 48 Senate seats after losing West '\n",
            " 'Virginia, Montana and Ohio. Could Democrats pick up two Senate seats from '\n",
            " 'the GOP in the 2026 midterm while controlling the presidency? Unlikely — but '\n",
            " 'then again, Democrats gaining Senate seats this year seemed unlikely and '\n",
            " 'they did it.\\n'\n",
            " '2028\\n'\n",
            " 'Republican pickup opportunities\\n'\n",
            " 'Platinum tier: None\\n'\n",
            " 'Gold tier: Nevada (Cortez Masto), Pennsylvania (Fetterman), Arizona (Kelly), '\n",
            " 'Georgia (Warnock) \\n'\n",
            " 'Silver tier: New Hampshire (Hassan)\\n'\n",
            " 'Bronze tier: Oregon (Wyden)')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('Beyond that, any other races look like an extreme stretch for Democrats. '\n",
            " 'Maybe Josh Hawley in Missouri could be vulnerable in a 2018-type '\n",
            " 'environment, but that’s less likely in a presidential year and Hawley’s '\n",
            " 'approval rating isn’t that bad. Put him in the “honorable mention” tier, I '\n",
            " 'guess.\\n'\n",
            " 'Given all this, I’d imagine the modal outcome from 2024 is something like '\n",
            " 'Republicans picking up about three Senate seats. There’s uncertainty around '\n",
            " 'that; it’s surely within the realm of possibility that Democrats hold the '\n",
            " 'Senate, but it’s also possible that Republicans could wind up with 54 or 55 '\n",
            " 'seats. And of course, all of these races are correlated. If the GOP is doing '\n",
            " 'well enough to win the presidency in 2024, they should have no trouble also '\n",
            " 'winning the Senate.9\\n'\n",
            " '2026\\n'\n",
            " 'Republican pickup opportunities\\xa0\\n'\n",
            " 'Platinum tier: None\\n'\n",
            " 'Gold tier: Georgia (Ossoff), Michigan (Peters)\\n'\n",
            " 'Silver tier: New Hampshire (Shaheen)\\n'\n",
            " 'Bronze tier: Minnesota (Smith), Virginia (Warner)')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('Beyond that, any other races look like an extreme stretch for Democrats. '\n",
            " 'Maybe Josh Hawley in Missouri could be vulnerable in a 2018-type '\n",
            " 'environment, but that’s less likely in a presidential year and Hawley’s '\n",
            " 'approval rating isn’t that bad. Put him in the “honorable mention” tier, I '\n",
            " 'guess.\\n'\n",
            " 'Given all this, I’d imagine the modal outcome from 2024 is something like '\n",
            " 'Republicans picking up about three Senate seats. There’s uncertainty around '\n",
            " 'that; it’s surely within the realm of possibility that Democrats hold the '\n",
            " 'Senate, but it’s also possible that Republicans could wind up with 54 or 55 '\n",
            " 'seats. And of course, all of these races are correlated. If the GOP is doing '\n",
            " 'well enough to win the presidency in 2024, they should have no trouble also '\n",
            " 'winning the Senate.9\\n'\n",
            " '2026\\n'\n",
            " 'Republican pickup opportunities\\xa0\\n'\n",
            " 'Platinum tier: None\\n'\n",
            " 'Gold tier: Georgia (Ossoff), Michigan (Peters)\\n'\n",
            " 'Silver tier: New Hampshire (Shaheen)\\n'\n",
            " 'Bronze tier: Minnesota (Smith), Virginia (Warner)')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('I’m not that impressed by long-term projections of Democratic doom in the '\n",
            " 'Senate. Mostly that’s because I’m not that impressed by long-term political '\n",
            " 'projections in general. Political coalitions change; not many people would '\n",
            " 'have had it on their bingo card that Democrats would obtain a Senate '\n",
            " 'majority in 2020 by winning two runoffs in Georgia, and then would actually '\n",
            " 'expand on that majority in 2022 following another runoff in Georgia despite '\n",
            " 'their party controlling the White House.\\n'\n",
            " 'With that said, while I don’t think we can say very much about what politics '\n",
            " 'will look like 10 or 20 years from now, a medium-term look-ahead is '\n",
            " 'possible. Do Democrats have a reasonable hope of regaining a trifecta — '\n",
            " 'control of the House, Senate and presidency — at any point in the near '\n",
            " 'future (2024, 2026 or 2028)? And how long will they hold onto control of the '\n",
            " 'Senate and the presidency, allowing them to appoint Supreme Court justices '\n",
            " 'without Republican help?')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('I’m not that impressed by long-term projections of Democratic doom in the '\n",
            " 'Senate. Mostly that’s because I’m not that impressed by long-term political '\n",
            " 'projections in general. Political coalitions change; not many people would '\n",
            " 'have had it on their bingo card that Democrats would obtain a Senate '\n",
            " 'majority in 2020 by winning two runoffs in Georgia, and then would actually '\n",
            " 'expand on that majority in 2022 following another runoff in Georgia despite '\n",
            " 'their party controlling the White House.\\n'\n",
            " 'With that said, while I don’t think we can say very much about what politics '\n",
            " 'will look like 10 or 20 years from now, a medium-term look-ahead is '\n",
            " 'possible. Do Democrats have a reasonable hope of regaining a trifecta — '\n",
            " 'control of the House, Senate and presidency — at any point in the near '\n",
            " 'future (2024, 2026 or 2028)? And how long will they hold onto control of the '\n",
            " 'Senate and the presidency, allowing them to appoint Supreme Court justices '\n",
            " 'without Republican help?')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('That bleak picture may shape the next few years of political maneuvering. '\n",
            " 'When Vox’s Dylan Matthews suggested on Twitter that liberal Justices Sonia '\n",
            " 'Sotomayor (age 68) and Elena Kagan (age 62) should retire while Democrats '\n",
            " 'have their Senate majority and be replaced by younger justices, it didn’t go '\n",
            " 'over well. But it’s a perfectly rational suggestion if Democrats don’t feel '\n",
            " 'like gambling with their judicial future. (Consider how consequential Ruth '\n",
            " 'Bader Ginsburg’s decision not to retire has been for liberals.) Democrats '\n",
            " 'have a narrow path to Senate control after 2024, but it’s narrow indeed, and '\n",
            " 'one that might require the GOP continuing to nominate bad candidates — and a '\n",
            " 'fair share of luck.\\n'\n",
            " 'url: '\n",
            " 'https://fivethirtyeight.com/features/democrats-senate-chances-2024-and-beyond/\\n'\n",
            " \"categories: ['news']\")\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('_id: 63927168098523fa9d5ef047\\n'\n",
            " 'title: Are The Democrats Screwed In The Senate After 2024?\\n'\n",
            " 'author: Nate Silver\\n'\n",
            " 'description: Are The Democrats Screwed In The Senate After 2024?\\n'\n",
            " 'No, but the party faces an uphill battlee')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('I don’t want to focus too much on any one individual race, but it’s probably '\n",
            " 'safe to assume that Joe Manchin has the toughest reelection campaign of all. '\n",
            " 'Meanwhile, maybe Sherrod Brown can feel a bit more comfortable following Tim '\n",
            " 'Ryan’s relatively good performance in Ohio this year, but Ryan did lose, and '\n",
            " 'Brown may face a tougher opponent than Ryan did in J.D. Vance. Democrats '\n",
            " 'also run the risk of retirements here; neither Manchin nor Jon Tester has '\n",
            " 'officially yet announced they will seek reelection.\\n'\n",
            " 'Beyond the platinum tier, Democrats face some other risks. Kyrsten Sinema is '\n",
            " 'likely to face a serious primary challenge. Democrats keep winning close '\n",
            " 'races for Congress in Nevada but it’s a very purple state. Tammy Baldwin, '\n",
            " 'Debbie Stabenow and Bob Casey won by fairly comfortable margins in 2018 and '\n",
            " 'will likely be fine in the event of a decent-to-good Democratic year, but if '\n",
            " '2024 turns into a good Republican year, they could be in trouble.\\n'\n",
            " 'Democratic pickup opportunities')\n",
            "\n",
            "> Are The Democrats Screwed In The Senate After 2024?\n",
            "('I don’t want to focus too much on any one individual race, but it’s probably '\n",
            " 'safe to assume that Joe Manchin has the toughest reelection campaign of all. '\n",
            " 'Meanwhile, maybe Sherrod Brown can feel a bit more comfortable following Tim '\n",
            " 'Ryan’s relatively good performance in Ohio this year, but Ryan did lose, and '\n",
            " 'Brown may face a tougher opponent than Ryan did in J.D. Vance. Democrats '\n",
            " 'also run the risk of retirements here; neither Manchin nor Jon Tester has '\n",
            " 'officially yet announced they will seek reelection.\\n'\n",
            " 'Beyond the platinum tier, Democrats face some other risks. Kyrsten Sinema is '\n",
            " 'likely to face a serious primary challenge. Democrats keep winning close '\n",
            " 'races for Congress in Nevada but it’s a very purple state. Tammy Baldwin, '\n",
            " 'Debbie Stabenow and Bob Casey won by fairly comfortable margins in 2018 and '\n",
            " 'will likely be fine in the event of a decent-to-good Democratic year, but if '\n",
            " '2024 turns into a good Republican year, they could be in trouble.\\n'\n",
            " 'Democratic pickup opportunities')\n",
            "----------------------------------SOURCE DOCUMENTS---------------------------\n"
          ]
        }
      ],
      "source": [
        "from pprint import pprint\n",
        "\n",
        "# Print question, answer, and evaluations\n",
        "print(\"\\n\\n> Question:\")\n",
        "pprint(query)\n",
        "print(\"\\n> Answer:\")\n",
        "pprint(answer)\n",
        "print(\"\\n> Eval:\")\n",
        "pprint(evals)\n",
        "\n",
        "# Print the relevant sources used for the answer\n",
        "print(\"----------------------------------SOURCE DOCUMENTS---------------------------\")\n",
        "for document in sources:\n",
        "    print(\"\\n> \" + document.metadata[\"source\"])\n",
        "    pprint(document.page_content[:1000])\n",
        "print(\"----------------------------------SOURCE DOCUMENTS---------------------------\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "UV7_OQHg7WVW"
      },
      "outputs": [],
      "source": [
        "import gradio as gr"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "n1wKmRl37WVW"
      },
      "outputs": [],
      "source": [
        "def get_response_from_query(db, query, k=3):\n",
        "\n",
        "\n",
        "    docs = db.similarity_search(query, k=k)\n",
        "\n",
        "    docs_page_content = \" \".join([d.page_content for d in docs])\n",
        "\n",
        "    # llm = BardLLM()\n",
        "    llm = ChatOpenAI(model_name=\"gpt-3.5-turbo-16k\",temperature=0)\n",
        "\n",
        "    prompt = PromptTemplate(\n",
        "        input_variables=[\"question\", \"docs\"],\n",
        "        template=\"\"\"\n",
        "        A bot that is open to discussions about different cultural, philosophical and political exchanges. I will use do different analysis to the articles provided to me. Stay truthful and if you weren't provided any resources give your oppinion only.\n",
        "        Answer the following question: {question}\n",
        "        By searching the following articles: {docs}\n",
        "\n",
        "        Only use the factual information from the documents. Make sure to mention key phrases from the articles.\n",
        "\n",
        "        If you feel like you don't have enough information to answer the question, say \"I don't know\".\n",
        "\n",
        "        \"\"\",\n",
        "    )\n",
        "\n",
        "    chain = LLMChain(llm=llm, prompt=prompt)\n",
        "    # chain = RetrievalQAWithSourcesChain.from_chain_type(llm=llm, prompt=prompt,\n",
        "    #                                                     chain_type=\"stuff\", retriever=db.as_retriever(), return_source_documents=True)\n",
        "\n",
        "    response = chain.run(question=query, docs=docs_page_content,return_source_documents=True)\n",
        "    r_text = str(response)\n",
        "\n",
        "    ##evaluation part\n",
        "\n",
        "    prompt_eval = PromptTemplate(\n",
        "        input_variables=[\"answer\", \"docs\"],\n",
        "        template=\"\"\"\n",
        "       You job is to evaluate if the response to a given context is faithful.\n",
        "\n",
        "        for the following: {answer}\n",
        "        By searching the following article: {docs}\n",
        "\n",
        "       Give a reason why they are similar or not, start with a Yes or a No.\n",
        "        \"\"\",\n",
        "    )\n",
        "\n",
        "    chain_part_2 = LLMChain(llm=llm, prompt=prompt_eval)\n",
        "\n",
        "\n",
        "    evals = chain_part_2.run(answer=r_text, docs=docs_page_content)\n",
        "\n",
        "    return response,docs,evals"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7VMW3kDRG2lM"
      },
      "outputs": [],
      "source": [
        "from linkify_it import LinkifyIt"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "IdKyZMdyjQ1g"
      },
      "outputs": [],
      "source": [
        "#!pip list"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 610
        },
        "id": "YqMJNvrl7WVW",
        "outputId": "f22b7e2c-2372-4032-fefc-037a4f4d45c7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Colab notebook detected. To show errors in colab notebook, set debug=True in launch()\n",
            "Running on public URL: https://91a238021b8734d96f.gradio.live\n",
            "\n",
            "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<div><iframe src=\"https://91a238021b8734d96f.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": []
          },
          "metadata": {},
          "execution_count": 38
        }
      ],
      "source": [
        "def greet(query):\n",
        "\n",
        "    answer,sources,evals = get_response_from_query(db,query,2)\n",
        "    return answer,sources,evals\n",
        "examples = [\n",
        "    [\"How to be happy\"],\n",
        "    [\"Climate Change Challenges in Europe\"],\n",
        "    [\"Philosophy in the world of Minimalism\"],\n",
        "    [\"Hate Speech  vs Freedom of Speech\"],\n",
        "    [\"Articles by Noam Chomsky on US Politics\"],\n",
        "    [\"The importance of values and reflection\"]\n",
        "    ]\n",
        "demo = gr.Interface(fn=greet, title=\"cicero-semantic-search\", inputs=\"text\",\n",
        "                    outputs=[gr.components.Textbox(lines=3, label=\"Response\"),\n",
        "                             gr.components.Textbox(lines=3, label=\"Source\"),\n",
        "                             gr.components.Textbox(lines=3, label=\"Evaluation\")],\n",
        "                   examples=examples)\n",
        "\n",
        "demo.launch(share=True)\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "UnjSpVxc7WVW",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3e8b0fe4-de94-4cfb-d6f4-3d4f5b10ad41"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded as API: https://91a238021b8734d96f.gradio.live/ ✔\n",
            "(\"Based on the provided articles, the concerns of Democrats for the 2024 Senate seem to revolve around the challenging electoral map that they will face. The articles mention that even with an additional senator going into 2023, the 2024 map is still unfavorable for Democrats, making it difficult for them to keep the Senate for years to come. The party's prospects may rely on limiting the damage in 2024 to have a chance of regaining the Senate in 2026 or 2028. However, a bad outcome in 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.\\n\\nIn summary, the concerns of Democrats for the 2024 Senate primarily stem from the unfavorable electoral map and the potential long-term implications if they are unable to limit the damage in that election.\", \"[Document(page_content='Even with an additional senator going into 2023, the 2024 map is still so bad for Democrats that keeping the Senate for years to come will be a fairly tough order. The party’s prospects might rest more upon limiting the damage in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a bad 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0}), Document(page_content='Even with an additional senator going into 2023, the 2024 map is still so bad for Democrats that keeping the Senate for years to come will be a fairly tough order. The party’s prospects might rest more upon limiting the damage in 2024 so that it has a chance to regain the Senate in 2026 or 2028. But a bad 2024 could make it very difficult for Democrats to regain the Senate before 2030 or 2032.', metadata={'source': 'Are The Democrats Screwed In The Senate After 2024?', 'row': 0})]\", 'Yes, the response is faithful to the given context. The response accurately reflects the concerns of Democrats for the 2024 Senate, mentioning the unfavorable electoral map and the potential long-term implications if they are unable to limit the damage in that election. The response also includes similar language and structure as the provided articles.')\n"
          ]
        }
      ],
      "source": [
        "from gradio_client import Client\n",
        "\n",
        "client = Client(\"https://91a238021b8734d96f.gradio.live/\")\n",
        "result = client.predict(\n",
        "\t\t\"Can you tell me about the concerns of Democrats for 2024 Senate\",\t# str  in 'query' Textbox component\n",
        "\t\tapi_name=\"/predict\"\n",
        ")\n",
        "print(result)"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "KcuKWqwj_AgQ"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "environment": {
      "kernel": "python3",
      "name": "pytorch-gpu.1-13.m108",
      "type": "gcloud",
      "uri": "gcr.io/deeplearning-platform-release/pytorch-gpu.1-13:m108"
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.10"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}