Upload folder using huggingface_hub

README.md

@@ -25,26 +25,38 @@ model.print_topics()
 The model is structured as follows:
 
 ```
-SemanticSignalSeparation(decomposition=FastICA(n_components=10),
-                         vectorizer=CountVectorizer(min_df=10,
-                                                    stop_words='english'))
+ClusteringTopicModel(clustering=KMeans(n_clusters=20),
+                     dimensionality_reduction=PCA(n_components=5),
+                     feature_importance='c-tf-idf',
+                     vectorizer=CountVectorizer(min_df=10,
+                                                stop_words='english'))
 ```
 
 ## Topics
 The topics discovered by the model are the following:
 
-| Topic ID | Highest Ranking | Lowest Ranking |
-| - | - | - |
-| 0 | goaltenders, nhl, bullpen, sabres, goaltender, puckett, leafs, braves, pitchers, canucks | accelerator, malaysia, automobile, accelerators, mazda, automobiles, automotive, vehicle, silicon, britain |
-| 1 | saturn, suzuki, symptoms, jupiter, bmw, exhaust, volvo, engine, mazda, propulsion | wiretapping, wiretaps, nsa, spying, eavesdropping, wiretap, security, encryption, enforcement, safeguarding |
-| 2 | drawbacks, advantages, productivity, efficiency, innovation, economical, disadvantages, proponents, competitiveness, economically | address, instructions, serial, arrived, codes, configured, 9591, 16550, contacting, recieved |
-| 3 | publishes, archives, publisher, scholars, manuscripts, npr, affiliated, revelations, discusses, archive | motorcycling, motorcycles, speeding, motorcycle, driving, motorcyclist, riding, harleys, braking, vehicles |
-| 4 | motherboard, ram, motherboards, processor, cmos, hardware, chipset, chipsets, amd, mb | yale, sunroof, damphousse, library, npr, billboards, balloon, schools, kerosene, nicholas |
-| 5 | palestinians, palestinian, gazans, gaza, genocide, israelis, atrocities, israeli, hamas, holocaust | motorola, mastercard, technician, smartdrive, telephony, transmissions, phones, electronically, voyager, cruising |
-| 6 | spectrometer, makefile, biochemistry, dblspace, bibliography, booklet, bookstores, circumference, nutritional, statistically | uh, um, em, yeah, oh, er, ah, yer, yo, ye |
-| 7 | theology, theological, scripture, theologians, christianity, biblical, agnosticism, devout, agnostic, christians | missiles, munitions, soviets, artillery, bunker, missile, explosives, tactical, grenades, soviet |
-| 8 | causes, metabolism, obstruction, bugging, xsession, disabling, debugger, behaviour, syndrome, occurs | prices, pricing, price, affordable, cheap, forsale, inexpensive, cost, purchases, priced |
-| 9 | xcreatewindow, programmable, bitmap, bitmaps, colormaps, freeware, gui, imagewriter, colormap, adobe | discrepancy, inaccuracies, defective, debacle, unrecognized, faulty, misconception, sceptical, refutation, warranted |
+| Topic ID | Highest Ranking |
+| - | - |
+| 0 | ax, max, g9v, b8f, jpeg, pl, a86, db, 1d9, file |
+| 1 | drive, scsi, price, card, sale, 00, shipping, ram, pc, offer |
+| 2 | pathetic, path, patient, patience, paths, pathology, patrick, patent, patently, patriot |
+| 3 | key, encryption, government, clipper, chip, keys, law, use, nsa, escrow |
+| 4 | people, right, don, think, just, government, like, say, does, rights |
+| 5 | game, team, year, 25, play, games, players, 10, 55, season |
+| 6 | dos, windows, image, file, edu, ftp, version, files, available, program |
+| 7 | god, jesus, bible, people, christ, believe, christians, christian, faith, say |
+| 8 | mr, president, people, fbi, gun, think, did, don, batf, know |
+| 9 | space, use, new, launch, used, like, don, know, just, 00 |
+| 10 | god, jews, people, church, does, did, christian, greek, just, israel |
+| 11 | car, just, like, don, people, think, money, insurance, make, time |
+| 12 | software, windows, thanks, know, version, does, ftp, available, xfree86, pc |
+| 13 | ax, edu, information, pub, space, ftp, data, mail, file, entry |
+| 14 | hockey, game, games, team, season, nhl, la, league, don, pts |
+| 15 | armenian, armenians, turkish, people, said, israel, jews, genocide, israeli, armenia |
+| 16 | 00, car, new, 50, price, bike, good, like, 1st, 10 |
+| 17 | like, just, time, problem, don, use, know, vitamin, good, think |
+| 18 | drive, scsi, card, disk, windows, controller, drives, use, bus, ide |
+| 19 | ax, max, edu, com, b8f, ah, 145, a86, pl, air |
 
 ## Package versions
 

model.joblib

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:414ec7e8a0e127ce291f85425722f0df7219e16eb236a972dd5314e6d2cdb2f8
-size 145878291
+oid sha256:7c44d3bc14662cd8548ed94cbb5294496d2d8c20874fb2f5189797b0a0fbce92
+size 139475171