Datasets:

tomekkorbak

/

detoxify-pile-chunk3-5500000-5550000

like 0

[ "Company Info:\nWilliam M. Cobb & Associates, Inc.\n12770 Coit Road, Suite 907\nDallas, TX 75251\n972-385-0354 (phone)\n972-788-5165 (fax)\n\nContact:\nRandal M. Brush\nSenior Vice President\n\nIf you need any other info just let me know." ]

{ "pile_set_name": "Enron Emails" }

[ 0.0005886671133339405 ]

[ "\n241 F.Supp.2d 572 (2002)\nOCEAN WINDS COUNCIL OF CO-OWNERS, INC., ", "Plaintiff,\nv.\nAUTO-OWNERS INSURANCE COMPANY, Defendant.", "\nNo. ", "CIV.A.2:98-2970-18.", "\nUnited States District Court, D. South Carolina, Charleston Division.", "\nDecember 16, 2002.", "\n*573 William Jefferson Leath, Jr., Timothy David St. Clair, Charleston, SC, for plaintiff.", "\nWalter Henry Bundy, Jr., Mt. Pleasant, SC, Mark Crawford, Jems Dowell Gandy, III, John C. Hayes, IV, Robert H. Hood, Charleston, SC, for defendant.", "\n\n\n*574 ORDER\nNORTON, District Judge.", "\n\nI. Background\nThis is a first-party insurance action brought by the insured, Ocean Winds Council of Co-Owners, Inc. (\"Ocean Winds\"), against its insurer, Auto-Owners Insurance Co. (\"Auto Owners\"). ", "The following facts are undisputed. ", "Ocean Winds is the owner of a condominium complex that was insured from July 1993 to July 2001 under a policy issued by Auto-Owners. ", "Ocean Winds filed a claim with Auto Owners on October 27, 1997, requesting payment for water and termite damage to the buildings under the following \"collapse\" provision in the insurance contract:\n5. ", "Additional Coverages\nd. Collapse\nWe will pay for loss or damage caused by or resulting from risks of direct physical loss involving collapse of a building or any part of the building caused only by one or more of the following:\n. . . . .", "\n(2) hidden decay;\n(3) hidden insect or vermin damage\n. . . . .", "\nCollapse does not include settling, cracking, shrinkage, bulging, or expansion.", "\nAfter Auto Owners did not pay the claim, Ocean Winds filed suit on September 4, 1998, in state court alleging the following causes of action: bad faith refusal to pay benefits; breach of contract; improper claims practices in violation of S.C.Code § 38-59-20 (West 2002); and attorneys' fees under S.C.Code § 38-59-40 (West 2002).[1]\nAfter defendant removed the action, this court certified the following question to the South Carolina Supreme Court:\nIn order to trigger coverage [under the policy quoted above], is it required: 1) that the building or part of the building fall to the ground or be reduced to flattened rubble; or 2) that the building manifest substantial structural impairment, but has not yet fallen to the ground or been reduced to flattened rubble.", "\nThe South Carolina Supreme Court held on June 17, 2002, that \"a requirement of imminent collapse is the most reasonable construction of the policy clause covering `risks of direct physical loss involving collapse.' ", "We define imminent collapse to mean collapse is likely to happen without delay.\" ", "Ocean Winds Council of Co-Owners, Inc. v. Auto-Owner Ins. ", "Co., 350 S.C. 268, 565 S.E.2d 306, 308 (2002). ", "During the time in which the certified question was pending, plaintiff paid to have the buildings repaired at a cost of $582,494.52. (", "Pl's.", "Mem.", "Opp.", "Summ. ", "J. Ex. ", "A.)\n\nII. ", "Summary Judgment Standard\nSummary judgment shall be granted when the \"pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.\" ", "Fed.", "R.Civ.", "P. 56(c). \"", "At the summary judgment stage the judge's function is not himself to weigh the evidence and determine the truth of the matter but to determine whether there is a genuine issue for trial.\" ", "Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. ", "2505, 91 L.Ed.2d 202 (1986). ", "There is no requirement *575 that the trial judge make findings of fact. ", "Id. at 250, 106 S.Ct. ", "2505. ", "Rather, the threshold inquiry is whether \"there are any genuine factual issues that properly can be resolved only by a finder of fact because they may reasonably be resolved in favor of either party.\" ", "Id. In other words, \"to grant summary judgment the court must determine that no reasonable jury could find for the nonmoving party on the evidence before it.\" ", "See Perini Corp. v. Perini Const, Inc., 915 F.2d 121, 124 (4th Cir. ", "1990). ", "An issue of fact concerns material facts only if establishment of the fact might affect the outcome of the lawsuit under governing substantive law. ", "Anderson, 477 U.S. at 248, 106 S.Ct. ", "2505. ", "All facts and reasonable inferences therefrom are viewed in the light most favorable to the nonmoving party. ", "Matsushita Elec. ", "Indus. ", "Co. v. Zenith Radio Corp., 475 U.S. 574, 587-88,106 S.Ct. ", "1348, 89 L.Ed.2d 538 (1986).", "\n\"[A] party seeking summary judgment always bears the initial responsibility of informing the district court of the basis for its motion, and identifying those portions of `the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any,' which it believes demonstrate the absence of a genuine issue of material fact.\" ", "Celotex Corp. v. Catrett, 477 U.S. 317, 324, 106 S.Ct. ", "2548, 91 L.Ed.2d 265 (1986) (quoting Fed. ", "R.Civ.", "P. 56(c)). ", "Once this burden has been met, the nonmoving party must \"go beyond the pleadings and by her own affidavits, or by the `depositions, answers to interrogatories, and admissions on file,' designate `specific facts showing that there is a genuine issue for trial' \" Id. (quoting Fed.", "R.Civ.", "P. 56(e)); see also Pleasurecrft Marine Engine Co. v. Thermo Power Corp., 272 F.3d 654, 658 (4th Cir.2001).", "\n\nIII. ", "Law/Analysis\n\nA. Breach of Contract\nDefendant argues that there is no genuine issue of material fact whether the buildings were in danger of \"imminent collapse\" as defined by the South Carolina Supreme Court. ", "As an initial matter, the parties differ on the standard set forth by the South Carolina Supreme Court. ", "Plaintiff highlights a passage from the court's opinion that says the policy \"appears to cover even the threat of loss from collapse.\" ", "565 S.E.2d at 308 (emphasis added). ", "Plaintiff is correct that the court used this language to explain that actual collapse was not necessary to trigger coverage under the policy. ", "Id. However, the court further defined this \"threat\" to include only \"imminent collapse,\" which means \"collapse is likely to happen without delay.\" ", "Id. The court explained that \"collapse coverage should not be converted into a maintenance agreement by allowing recovery for damage which, while substantial, does not threaten collapse.\" ", "Id. Thus, the precise standard set forth by the South Carolina Supreme Court is whether the buildings were under a threat of \"imminent collapse,\" which means \"collapse is likely to happen without delay.\" ", "Id.\nApplying this standard, defendant points to the following passage from plaintiffs expert, Mr. Peter E. Sherrat: \"I am not sure whether imminent means tomorrow or not, but in a short timeframe. ", "And whether that timeframe is in three months or six months, but it was not at the stage meaning the next day I thought it was going to fall down.", "\"[2] In addition, defendant *576 argues that in the five years between the time the suit was filed and the repairs were made, the buildings did not collapse and were never vacated for safety reasons. ", "Plaintiff submits affidavits from Robert Norman, Robert Sisnroy, and Brett E. Carlson purporting to show that the collapse of the buildings was imminent. (", "Pl's Opp. ", "Mem. ", "Exs. ", "A, B. and C.) However, defendant points out that each of these affiants testified that the buildings were under a threat of collapse with a significant weather or seismic event. (", "Nornan Aff. ¶ ", "12, Sisnroy Aff. ¶ ", "5; Carlson Aff. ¶ ", "4.) ", "This testimony is not sufficient to create a genuine issue of material fact whether the buildings were faced with \"imminent collapse without delay.\" ", "Many buildings are subject to collapse in the event of a \"significant weather or seismic event.\" ", "This type of structural threat, although it may be serious and require repair, does not rise to the level of \"imminent collapse\" as defined by the South Carolina Supreme Court. ", "Therefore, viewing the evidence in the light most favorable to plaintiff, it has failed to meet its burden of presenting evidence showing there is a genuine issue of material fact whether the condominium complex was under the threat of imminent collapse without delay. ", "As a result, there is no genuine issue of material fact whether defendant breached its insurance contract.", "\n\nB. Bad Faith\nDefendant has moved for summary judgment on plaintiffs bad faith refusal to pay cause of action on the grounds that it cannot be held liable for a bad faith refusal to pay when it had no duty to pay under the contract. ", "Plaintiff argues that even if there was no breach of the insurance contract, defendant exercised bad faith in processing its claim. \"[", "T]here is an implied covenant of good faith and fair dealing in every insurance contract that neither party will do anything to impair the other's rights to receive benefits under the contract.\" ", "Tadlock Painting Co. v. Maryland Cas. ", "Co., 322 S.C. 498, 473 S.E.2d 52, 53 (1996) (internal citations and quotation marks omitted.) ", "In general, \"[a]n insured may recover damages for a bad faith denial of coverage if he or she proves there was no reasonable basis to support the insurer's decision to deny benefits under a mutually binding insurance contract.\" ", "Cock-N-Bull Steak House, Inc. v. Gen. Ins. ", "Co., 321 S.C. 1, 466 S.E.2d 727, 730 (1996). ", "However, \"the benefits due an insured are not limited solely by those expressly set out in the contract.\" ", "Tadlock, 473 S.E.2d at 55. \"[", "T]he covenant of good faith and fair dealing extends not just to the payment of a legitimate claim, but also to the manner in which it is processed.\" ", "Mixson, Inc. v. American Loyalty Ins. ", "Co., 562 S.E.2d at 662 (citing Tadlock, 473 S.E.2d at 52). \"[", "I]f an insured can demonstrate bad faith or unreasonable action by the insurer in processing a claim under their mutually binding insurance contract, he can recover consequential damages in a tort action.\" ", "Tadlock, 473 S.E.2d at 53 (quoting Nichols, 306 S.E.2d at 619) (emphasis in original). ", "Thus, breach of an express contractual provision is not a prerequisite to bringing a bad faith cause of action. ", "Tadlock, 473 S.E.2d at 54.", "\nPlaintiff argues that defendant exercised bad faith in its processing of plaintiffs claim. ", "Defendant responds that the delay was caused by Auto Owner's contesting the coverage under the policy. ", "However, plaintiff has submitted expert testimony from Thomas Hesse saying that Auto Owner's (1) failed to provide a copy of the reservation of rights letter to the insured; (2) failed to provide a proof of loss form to the insured as required by it claims handling guidelines manual; (3) *577 failed to respond to correspondence from the insured attorney; and (4) failed to deny the claim or give any reasons for denying the claim. (", "Hesse Depo. ", "at 14-15.) ", "Hesse testified that \"the biggest evidence of lack of good faith in this case is just the delay in getting this case resolved.\" (", "Hesse Depo. ", "at 27.) ", "Hesse further testified that Auto Owner's \"never has outright denied the claim or gave [sic] any reason why it was denying the claim.\" (", "Hesse Depo. ", "at 15.) ", "In addition, Hesse cited Auto Owners's \"attitude about the claim, that they are not there to try to resolve the claim one way or the other. ", "It is trying to find ways to not provide the coverage.\" (", "Hesse Depo. ", "at 27.) ", "Specifically, Hesse pointed out that the internal files showed that the Auto Owners was considering using the fact that Ocean Winds had not presented a \"proof of loss\" as a reason for denying the claim when Auto Owners, when \"by both their guidelines and also by the policy, they are the ones that are supposed to provide the proof of loss form to the insured.\" (", "Hesse Depo. ", "at 27.) ", "Therefore, Hesse's testimony is sufficient to create a genuine is of fact as to whether defendant acted in bad faith in processing these claims.", "\nDefendant suggests that \"bad faith refusal to pay benefits\" and \"breach of the implied warranty of good faith and fair dealing\" are separate causes of action. ", "Specifically, defendant argues that a bad faith refusal to pay benefits cause of action requires that benefits be due under the insurance contract, whereas a breach of the implied warranty of good faith and fair dealing cause of action addresses bad faith processing of claims. ", "Accordingly, defendant contends that plaintiff only alleged a bad faith refusal to pay benefits cause of action in its Complaint, but has now attempted to transform this claim into a breach of the implied warranty of good faith and fair dealing cause of action by arguing that the claim was processed in bad faith even if coverage did not exist. ", "However, bad faith refusal to pay benefits and breach of implied warranty of good faith and fair dealing are not separate causes of action. ", "Rather, all bad faith actions—including claims based on bad faith processing of the claims when there is no breach of the insurance contract— arise out of the implied warranty of good faith and fair dealing. ", "See Tadlock, 473 S.E.2d at 52-53. ", "Therefore, although the label is admittedly confusing, plaintiffs \"bad faith refusal to pay benefits\" cause of action, construed under the liberal pleading standards of Fed.", "R.Civ.", "P. 8, encompassed its theory that defendant failed to process his claim in good faith even if there were no coverage under the contract.", "\n\nC. Improper Claim Practices Act\nPlaintiff also brought causes of action under sections 38-59-20 and 38-59-40 of the Improper Claim Practices Act (\"ICPA\"). ", "Section 38-59-20 provides in relevant part:[3]\nAny of the following acts by an insurer doing accident and health insurance, property insurance, casualty insurance, surety insurance, marine insurance, or title insurance business, if committed without just cause and performed with such frequency as to indicate a general business practice, constitutes improper claim practices:\n. . . . .", "\n(2) Failing to acknowledge with reasonable promptness pertinent communications with respect to claims arising under its policies, including third-party *578 claims arising under liability insurance policies.", "\n. . . . .", "\n(4) Not attempting in good faith to effect prompt, fair, and equitable settlement of claims, including third-party liability claims, submitted to it in which liability has become reasonably clear.", "\n. . . . .", "\n(8) Any other practice which constitutes an unreasonable delay in paying or an unreasonable failure to pay or settle in full claims, including third-party liability claims, arising under coverages provided by its policies.", "\nS.C.Code § 38-59-20 (West 2002).", "\nIn Masterclean, Inc. v. Star Ins. ", "Co., 347 S.C. 405, 556 S.E.2d 371, 377 (S.C. 2001), the South Carolina Supreme Court held that \"third parties do not have a private right of action under S.C.Code Ann. § ", "38-59-20.\" ", "Instead, they are entitled to administrative relief before the Chief Insurance Commissioner. ", "556 S.E.2d at 377. ", "Although this court is not aware of any South Carolina court decision that has specifically held that a first-party claimant also does not have a private right of action under section 38-59-20, some courts have issued broad statements that the Improper Claims Practices Act \"does not create a private cause of action.\" ", "Gaskins v. Southern Farm Bureau Cas. ", "Ins. ", "Co., 343 S.C. 666, 541 S.E.2d 269, 272 (S.C.App.2000) (third-party case); accord Swinton v. Chubb & Son, Inc., 283 S.C. 11, 320 S.E.2d 495, 497 (S.C.App.1984) (holding in a third-party case that § 38-37-1110, the predecessor statute, recodified as § 38-59-20 by 1987 Act No. ", "155, § 1, \"does not establish ... a private right of action for any violation thereof) (omissions in original). ", "Further, the text of section 38-59-20 does not draw any distinction between first-party and third-party claims. ", "Thus, this court concludes based on South Carolina court decisions in the third party context that the South Carolina courts would also conclude that section 38-59-20 does not create a statutory private cause of action for first-party claimants.[4]\nPlaintiff also seeks attorneys fees pursuant to S.C.Code § 38-59^0 (West 2002), under which \"[a]n insurer is liable to the policy holder for all reasonable attorneys' fees for the prosecution of the case against the insurer if the trial judge finds the refusal to pay the policyholder's claim was without reasonable cause or in bad faith.\" ", "Mixson, 562 S.E.2d at 663 (citing section 38-59-40). ", "Determination of an insurer's liability for attorneys' fees pursuant to section 38-59-40 is a matter for decision by the trial judge. ", "Dorman v. Allstate Ins. ", "Co., 332 S.C. 176, 504 S.E.2d 127, 130 (S.C.App.1998) (internal citations omitted). \"", "In making this determination, the trial judge must ascertain whether or not an insurer's refusal to pay a claim was without reasonable cause or in bad faith.\" ", "Id. As explained above, a genuine issue of material facts exists as to whether defendant acted in bad faith in its processing of plaintiffs claim. ", "Thus, the court will determine the issue of attorneys fees at trial.", "\n\nIII. ", "Conclusion\nIt is therefore,\nORDERED, for the foregoing reasons, that defendant's motion for summary judgment is GRANTED as to plaintiffs causes of action for breach of contract and improper *579 claim practices under S.C.Code § 38-59-20.", "\nIt is FURTHER ORDERED that defendant's motion for summary judgment is DENIED as to plaintiffs causes of action for bad faith refusal to pay benefits and attorneys' fees under S.C.Code § 38-59-40.", "\nAND IT IS SO ORDERED.", "\nNOTES\n[1] The Complaint also included a request for a declaratory judgment, which was later withdrawn.", "\n[2] Defendant quoted this passage in its Memorandum of Law in Support of Its Motion For Summary Judgment, but did not cite the deposition page number. ", "However, plaintiff has not contested its accuracy.", "\n[3] The Improper Claims Practices Act contains other provisions, but plaintiff has only presented evidence in support of provisions two, four and eight. (", "Hesse Depo. ", "at 15-16.)", "\n[4] As explained above, South Carolina courts have expressly recognized a bad faith cause of action for first-party claimants under the common law, so the significance of an additional statutory cause of action would be limited.", "\n" ]

{ "pile_set_name": "FreeLaw" }

[ 0.0007217532838694751, 0.0008183066383935511, 0.0013787426287308335, 0.0008883342379704118, 0.0007037832983769476, 0.0007329118088819087, 0.000692663830704987, 0.0007144397823140025, 0.0007691086502745748, 0.0006083524785935879, 0.0006253553438000381, 0.0007146314601413906, 0.0007567951106466353, 0.018261410295963287, 0.0034009700175374746, 0.0008103047148324549, 0.0007894622976891696, 0.0007553569739684463, 0.0007337552960962057, 0.0008668419905006886, 0.0006387130706571043, 0.0006203093798831105, 0.006289532873779535, 0.0012244058307260275, 0.0027753503527492285, 0.0009020013967528939, 0.0009593204595148563, 0.0008550651255063713, 0.0005613923422060907, 0.0022228683810681105, 0.0010373276891186833, 0.0006817776593379676, 0.000640416459646076, 0.0010075594764202833, 0.0007177218794822693, 0.0006429373170249164, 0.0009328110609203577, 0.0009804071160033345, 0.0005861147656105459, 0.0008839588263072073, 0.0007634854409843683, 0.0007576036150567234, 0.0005897827795706689, 0.0007756659178994596, 0.0009804071160033345, 0.0007800838793627918, 0.0009282667888328433, 0.00507537554949522, 0.0007591987960040569, 0.0007390322280116379, 0.0005940741393715143, 0.00086874671978876, 0.0006789308972656727, 0.0010373276891186833, 0.0006979253375902772, 0.000805189716629684, 0.0010373276891186833, 0.0006162792560644448, 0.0012426747707650065, 0.0006388738402165473, 0.0005885128630325198, 0.0006431530346162617, 0.0006308164447546005, 0.0006882656016387045, 0.0008669712697155774, 0.0007508347043767571, 0.0009644938400015235, 0.0005769242998212576, 0.0006259486544877291, 0.0005943278083577752, 0.0006783229764550924, 0.004915359430015087, 0.0012244058307260275, 0.0015471611404791474, 0.0005923784337937832, 0.0007269193301908672, 0.0006966192158870399, 0.0006784325814805925, 0.0008846686687320471, 0.0007147360593080521, 0.0007712091319262981, 0.0008496786467730999, 0.0006791308987885714, 0.0007355185225605965, 0.0007682373397983611, 0.0006900273729115725, 0.0006992848357185721, 0.0008481809636577964, 0.0006215628236532211, 0.0009427821496501565, 0.3208395540714264, 0.0006803883006796241, 0.0006319077219814062, 0.0006907981587573886, 0.000617418612819165, 0.0009634941234253347, 0.0007034691516309977, 0.001627274788916111, 0.0006226520636118948, 0.0008000725065357983, 0.0007569682784378529, 0.000752823194488883, 0.0006606057868339121, 0.0007158914813771844, 0.0007947107660584152, 0.0006721565150655806, 0.0005785101093351841, 0.0007947107660584152, 0.0007035116432234645, 0.0006964553031139076, 0.0007947107660584152, 0.0007089528953656554, 0.0005962844588793814, 0.0006087656947784126, 0.0007947107660584152, 0.0007035116432234645, 0.0005844426923431456, 0.0007947107660584152, 0.0007035116432234645, 0.0006980490288697183, 0.0007233317592181265, 0.0007866210653446615, 0.0007393537671305239, 0.0009876638650894165, 0.0008482447592541575, 0.0006445146864280105, 0.0006775803049094975, 0.0010373276891186833, 0.0006264562252908945, 0.0007272951770573854, 0.0006684139952994883, 0.0006358832470141351, 0.007180466782301664, 0.0006912898388691247, 0.007180466782301664, 0.0007775189005769789, 0.0006523836054839194, 0.0009820706909522414, 0.0007839002646505833, 0.0008730935514904559, 0.0006400336278602481, 0.0007811457617208362, 0.0006786704761907458, 0.0008844727417454123, 0.0008363372762687504, 0.0007999631343409419, 0.0007192371995188296, 0.0007148917065933347, 0.0006502855103462934, 0.0007019680342637002, 0.0005901349941268563, 0.0011162369046360254, 0.0006509207305498421, 0.0007280701538547873, 0.0007180410902947187, 0.0005902000120840967, 0.0012426747707650065, 0.0006806133897043765, 0.0007823238847777247, 0.0010100718354806304, 0.0006053477409295738, 0.000590000650845468, 0.0006375429220497608, 0.0006308300071395934, 0.0007947107660584152, 0.0006683930405415595, 0.00064614083385095, 0.001994825666770339 ]

[ "---\nabstract: |\n This paper explores modifications to a feedforward five-layer spiking convolutional network (SCN) of the ventral visual stream \\[Masquelier, T., Thorpe, S., Unsupervised learning of visual features through spike timing dependent plasticity. ", "PLoS Computational Biology, 3(2), 247-257\\]. ", "The original model showed that a spike-timing-dependent plasticity (STDP) learning algorithm embedded in an appropriately selected SCN could perform unsupervised feature discovery. ", "The discovered features where interpretable and could effectively be used to perform rapid binary decisions in a classifier.", "\n\n In order to study the robustness of the previous results, the present research examines the effects of modifying some of the components of the original model. ", "For improved biological realism, we replace the original non-leaky integrate-and-fire neurons with Izhikevich-like neurons. ", "We also replace the original STDP rule with a novel rule that has a probabilistic interpretation. ", "The probabilistic STDP slightly but significantly improves the performance for both types of model neurons. ", "Use of the Izhikevich-like neuron was not found to improve performance although performance was still comparable to the IF neuron. ", "This shows that the model is robust enough to handle more biologically realistic neurons. ", "We also conclude that the underlying reasons for stable performance in the model are preserved despite the overt changes to the explicit components of the model.", "\nauthor:\n- |\n Amirhossein Tavanaei:[^1]\\\n **\\\n **\\\n Timothée Masquelier:[^2]\\\n **\\\n **\\\n **\\\n **\\\n **\\\n **\\\n Anthony S. Maida:[^3]\\\n **\\\n **\ntitle: '**Acquisition of Visual Features Through Probabilistic Spike-Timing-Dependent Plasticity**'\n---\n\n[^4]\n\nIntroduction\n============\n\nThe pattern recognition functionality found in the ventral visual pathway of the primate brain is supported by a multi-layer, spiking biological network which is able to learn to recognize new patterns. ", "An early and highly influential model of this computation is found in [@Serre2007a]. ", "They introduced a hierarchy of simple/complex cells (model neurons) that incorporated feature extraction and location invariance pooling over larger and larger receptive field sizes. ", "The design has been used in many subsequent deep networks and is a member of the class of convolutional networks, that dates back to [@LeCun1998a] and [@Fukushima1980a].", "\n\nThe model in [@Serre2007a] did not use spiking neurons and thus could not support spike-timing-dependent plasticity (STDP) learning, which is known to occur in many parts of the mammalian brain. ", "Furthermore, under some conditions the recognition process in primates is so rapid that there is barely enough time for even one spike to travel from the retina to the decision region in the inferior temporal cortex [@Thorpe1996a]. ", "The first artificial spiking neural network (SNN) that included STDP and was designed to address the speed of recognition observed in the human and primate brain is that of Masquelier and Thorpe [@Masquelier2007a]. ", "Empirical data imposes the constraint that the majority of information used in recognition is contained in the first neural spike for any neuron in the feedforward pipeline. ", "The paper [@Masquelier2007a] presented a model that demonstrated how this was computationally possible and the information provided by the first spike was sufficient to drive an STDP-based feature discovery algorithm to support the recognition process.", "\n\nThe initial work was performed on the Caltech data set which presented images in a canonical viewpoint. ", "Later work in [@Kheradpisheh2015a] showed that this algorithm would scale to larger data sets which showed the same object from multiple views and at multiple sizes. ", "The recognition performance compared favorably with other approaches, specifically, [@Serre2007a] and [@Krizhevsky2012a].", "\n\nOne limitation of the work in [@Masquelier2007a] is that, despite the key observations underlying the design of the model, it raises new questions about biological realism that beg to be answered. ", "Specifically, there are two limitations that the work presented here considers. ", "First, spike generation in [@Masquelier2007a] is performed by non-leaky integrate-and-fire (IF) neurons. ", "This may not be a serious limitation, given the short time scale under which the model operates. ", "However, it would be prudent to examine the model’s performance using a more realistic spike-generation mechanism. ", "This is possible by replacing the IF neuron with a simplified Izhikevich model neuron [@Izhikevich2003a; @Izhikevich2007a].", "\n\nThe second question is why the STDP version used in [@Masquelier2007a] is effective. ", "Put differently, can we understand why this, or alternative STDP mechanisms, perform effective feature discovery within this framework? ", "The work in [@Kheradpisheh2015a] shows that indeed that the STDP used in [@Masquelier2007a] is an effective feature discovery mechanism. ", "Furthermore, a theory of how global properties of the network emerge through STDP was developed in [@Masquelier2010a].", "\n\nThe present paper takes preliminary steps to develop a probabilistic or Bayesian variant of the original model in [@Masquelier2007a]. ", "In particular, we are inspired by the papers of Nessler et al. [", "@Nessler2009a; @Nessler2013a]. ", "Those papers show that an appropriately chosen STDP rule with a particular probabilistic interpretation, when embedded in an appropriately designed winner-take-all (WTA) network, is able to approximate online expectation maximization. ", "Their paper also includes proof-of-concept experiments showing that implementations of the model are able to perform pattern classification.", "\n\nOne drawback of the model in [@Nessler2013a] is that its trainable ‘excitatory’ connections use negative weights. ", "This is because the trained weights converge to the log of a probability value (explained later). ", "This gives an elegant probabilistic interpretation to the weights but since the logarithm of a probability value is negative, the weights are necessarily negative. ", "The authors note that theoretically the weights can be shifted to positive values by choosing a positive parameter, $c$, equal to the magnitude of the largest negative weight in the implementation. ", "The present paper seeks a solution where the (excitatory) weights are positive.", "\n\nAlthough the network in [@Masquelier2007a] has five layers, it can be viewed functionally as consisting of three feedforward-connected modules: feature extraction, feature discovery, and classification. ", "The feature extraction and discovery components are duplicated across five spatial scales. ", "The feature discovery module uses a simple form of spike-time-dependent plasticity (STDP) based learning. ", "We compare the IF neuron with a simplified Izhikevich regular-spiking (RS) neuron which models the spike generation dynamics of one type of neocortical pyramidal cell, the principle neuron in the neocortex.", "\n\nThe present research studies variants of this network when STDP is modified and when the model neurons are modified. ", "Specifically, we modified the original implementation in [@Masquelier2007a]. ", "The motivation with respect to STDP was to obtain some insight as to why the learning rule is able to support feature discovery and to assess the effectiveness of our new probabilistic STDP rule in feature discovery. ", "The motivation with regard to using different model neurons was to incrementally improve the biological realism of the model and to examine the robustness of STDP-based feature acquisition with a different model neuron. ", "Finally, if our manipulations do not have an effect on the qualitative performance of the model, either positive or negative, then we have established that the model has some degree of robustness.", "\n\nBackground\n==========\n\nSTDP modifies synaptic weights in response to local timing differences between pre- and postsynaptic spikes. ", "The existence of several variants of STDP within synapses is well established in biological networks [@Markram2011b]. ", "In most cases, however, there is little consensus about its larger role in learning for biological networks. ", "Idealized models of STDP have been studied extensively.", "\n\nIn recent literature, there have been at least two approaches to addressing the question of why biological STDP might be effective in learning. ", "The first and most established approach seeks to obtain an idea of what kinds of temporal spike codes STDP produces [@Song2001a; @Guyonneau2005a] and what kind of codes STDP can learn to recognize [@Masquelier2009a].", "\n\nThe second approach assumes that the relevant biological networks implement a Bayesian computation. ", "Empirical studies have provided evidence that the some brain areas may perform a Bayesian analysis of sensory stimuli [@Doya2007a; @Kording2004a]. ", "Also, it is claimed that the spike generation process is stochastic with exponential function upon the membrane potential [@Rezende2011a; @Jolivet2006a]. ", "Probabilistic interpretations of STDP could form a theoretical link to the broader area of machine learning. ", "Nessler et al. (", "2009, 2013) showed that a version of the STDP rule can perform Bayesian computation in a spiking winner-take-all (WTA) network [@Nessler2009a; @Nessler2013a].", "\n\nBased on the Nessler et al. (", "2013) STDP rule, Kappel et al. (", "2014) developed a generic cortical microcircuit to approximately implement the hidden Markov model (HMM) learning [@Kappel2014a]. ", "In addition, a modified variant of this probability-based STDP has been employed in a Gaussian mixture model (GMM) training of the HMM equipped by embedded SNNs in its states [@Tavanaei2015b].", "\n\nThe present work is inspired by the last approach and seeks to study a probabilistic rule in the context of the network used in [@Masquelier2007a].", "\n\nNetwork Architecture\n====================\n\nThe SNN architecture in the previous study (Masquelier and Thorpe, 2007 [@Masquelier2007a]) is shown in Fig.", " \\[fig:SNNArchitecture\\]. ", "The network consists of four layers hierarchy of spiking neurons (S1-C1-S2-C2) in temporal domain. ", "Simple cells (S) compute a linear sum (integration) while complex cells (C) compute nonlinear max pooling. ", "The last layer (layer 5) performs a classification over visual features received from C2.", "\n\nLayer S1 detects edges of the image with four orientations using intensity-to-latency conversion. ", "The WTA between S1 cells is implemented to find the best matching orientation. ", "C1 maps subsample S1 maps by taking the maximum response over a square neighborhood. ", "S2 is selective to intermediate complexity visual features, defined as a combination of oriented edges. ", "C2 takes the maximum response of S2 over all positions and scales. ", "The maximum operation of the complex cells propagates the first spike emitted by a given group of afferent neurons because of the time-to-first-spike framework supposed in this architecture.", "\n\nSTDP learning occurs in the weights projecting from layer C1 to layer S2. ", "The STDP has the effect of concentrating large synaptic weights on afferent neurons that systematically fire early, while postsynaptic spike latencies decrease. ", "The synaptic weights are updated and duplicated at all positions and scales. ", "The synaptic weights are considered between 0 and 1 to ensure excitatory synapses. ", "A simplified STDP has been used in this structure in which the time window is supposed to cover the whole spike wave.", "\n\nWhen a cell fires, local inhibition prevents other cells in that layer from firing at the same scale and within a square neighborhood of the firing position.", "\n\nThe final membrane potential measured in the C2 cells (and is contrast invariant) is used for training the radial basis function (RBF) classifier. ", "Images are processed by the network one by one and the neurons in each layer are allowed to fire only once. ", "Finally, the classifier layer utilizes RBF classification to recognize the patterns according to the visual features received from C2.", "\n\n![", "5-layer hierarchical spiking neural network developed in [@Masquelier2007a]. ", "There is one S1-C1-S2 pathway for each processing scale. ", "C2 cells select the maximum response received from S2. ", "Finally, visual features are classified by the last layer. ", "STDP learning is applied to the synapses connecting C1 and S2 neuron layers. ", "The synaptic weight change in the SNN proposed in [@Masquelier2007a] is computed by a factor of $w(1-w)$. In the current investigation, we change the STDP rule and neuron models to study the effect on performance. (", "from [@Masquelier2007a])[]{data-label=\"fig:SNNArchitecture\"}](Figure1.jpg){width=\"9cm\"}\n\nThe STDP rules tested\n=====================\n\nIn a later section, we will examine network classification performance using a novel STDP rule. ", "Here, we explain the original STDP rule used in [@Masquelier2007a], along with the new rule. ", "Neither rule needs to compute the exact time difference between pre- and postsynaptic spikes.", "\n\nThe original STDP rule\n----------------------\n\nThe original STDP rule given in [@Masquelier2007a] is shown below. ", "$w_{ki}$ denotes the weight from presynaptic neuron $i$ to postsynaptic neuron $k$. $$\\Delta w_{ki}=\n\\begin{cases}\n&a^{+}\\cdot w_{ki}\\cdot(1-w_{ki}), \\ \\ t_{k}^\\mathrm{f} - t_{i}^\\mathrm{f} \\ge 0\\\\\n&-a^{-}\\cdot w_{ki}\\cdot(1-w_{ki}), \\ \\ \\mathrm{otherwise}\\\\\n\\end{cases}\n\\label{MasquelierThorpeSTDP}$$\n\nThe first case describes LTP and the second describes LTD. ", "$t_{i}^\\mathrm{f}$ and $t_{k}^\\mathrm{f}$ denote firing times of units $i$ and $k$, respectively. ", "Weights fall in the range (0, 1). ", "Note that the quantity $w_{ki}\\cdot (1-w_{ki})$ is the slope of the sigmoidal function. ", "Therefore, if $a^{+} = 1$ and $a^{-}=1$, weight changes will be largest when $w_{ki}$ is $0.5$ and weight changes will become arbitrarily small as $w_{ki}$ approaches the bounds 0 or 1. ", "In the most of simulations, the magnitudes of amplification parameters $a^{+}$ and $a^{-}$ are kept in a ratio of 4/3.", "\n\nProbabilistic STDP variant\n--------------------------\n\nThe proposed new rule yields weights with a probabilistic interpretation after convergence. ", "Our new rule is modified from [@Nessler2013a]. ", "The rule from [@Nessler2013a], when embedded in an appropriate winner-take-all network with Poisson spiking neurons can approximate expectation maximization learning. ", "It is given below. ", "$$\\Delta w_{ki}=\n\\begin{cases}\n&e^{-w_{ki}}-1, \\ \\ 0< t_{k}^\\mathrm{f} - t_{i}^\\mathrm{f} < \\epsilon\\\\\n&-1, \\ \\ \\ \\ \\ \\ \\ \\ \\ \\ \\mathrm{otherwise}\\\\\n\\end{cases}\n\\label{nessler}$$\n\nLTP occurs if the presynaptic neuron fires briefly (e.g., within $\\epsilon=10$ ms) before the postsynaptic neuron. ", "Otherwise LTD occurs. ", "The rule is unusual in that the magnitude of the LTP weight adjustment is exponentially related to the current value of the weight. ", "Additionally, weight values are always negative. ", "After convergence, the weight value is equal to the log of the probability that the presynaptic neuron will have fired within the $\\epsilon$ interval, given that the postsynaptic neuron has fired.", "\n\nOur SNN uses positive weights for the excitatory neurons in the feedforward layers and therefore the rule (\\[nessler\\]) is not applicable. ", "Therefore, we modified (\\[nessler\\]) to obtain (\\[AT\\_STDP\\]) below.", "\n\n$$\\Delta w_{ki}=\n\\begin{cases}\n&a^+\\cdot e^{-w_{ki}}, \\ \\ t_{k}^\\mathrm{f} - t_{i}^\\mathrm{f} \\ge 0, \\ \\ \\ (\\mathrm{LTP})\\\\\n&-a^-, \\ \\ \\ \\ \\ \\ \\ \\ \\ \\mathrm{otherwise}, \\ \\ \\ \\ (\\mathrm{LTD})\\\\\n\\end{cases}\n\\label{AT_STDP}$$\n\nwhere $a^+$ and $a^-$ are the amplification parameters used in (\\[MasquelierThorpeSTDP\\]). ", "The synaptic weights are constrained to be positive during the training process.", "\n\nProbabilistic perspective\n-------------------------\n\nTo interpret the behavior of the STDP rule probabilistically, we follow the approach of Nessler et al.", " (2013), p. 21, formula 28 [@Nessler2013a]. ", "At equilibrium, the expected weight change will be zero. ", "This can be written and then simplified as shown below. ", "The notation $z_k=1$ means that the postsynaptic neuron fired on a given stimulus presentation. ", "The notation $y_k=1$ means that the presynaptic neuron fired before the postsynaptic neuron on the same stimulus presentation. ", "$$\\begin{array}{l}\nE[\\Delta w_{ki}]=0\n\\Leftrightarrow\\\\\na^+ \\cdot p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1) e^{-w_{ki}} - a^- \\cdot p_\\mathbf{w}^{*}(y_i = 0 | z_k = 1) = 0\n\\Leftrightarrow\\\\\na^+ \\cdot p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1) e^{-w_{ki}} + a^-\\cdot p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1) = a^- \n\\Leftrightarrow\\\\\ne^{-w_{ki}}=\\frac{a^- -a^- \\cdot p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1)}{a^+ \\cdot p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1)}\n\\Leftrightarrow\\\\\nw_{ki}=\\ln \\big(\\frac{a^+}{a^-} \\cdot \\frac{p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1)}{1-p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1)} \\big) \\Leftrightarrow \\nonumber\n\\end{array}$$ $$\\label{eqn:equilibrium}\nw_{ki}=\\ln \\big (\\frac{a^+}{a^-}\\big ) + \\ln(\\textit{\\textbf{Odds Ratio}})$$ where, $p^*$ denotes an equilibrium probability.", "\n\nThe synaptic weight, $w_{ki}$, in (\\[eqn:equilibrium\\]) specifies the log odds (logit) of the probability. ", "The probability, $p^*$, denotes the casual effect of presynaptic neuron in postsynaptic spike in which the STDP rule is triggered. ", "The odds ratio is the ratio of the probability that the event of interest occurs to the probability that it does not [@Bland2000a]. ", "The event of interest is the presynaptic activation right before the postsynaptic neuron fires. ", "Positive log odds ($p_\\mathbf{w}^{*}(y_i = 1 | z_k = 1) \\geq \\frac{a^- }{a^- +a^+}$) specifies a probable condition in which the synaptic weight is subject to LTP. ", "The parameter $\\frac{a^-}{a^- +a^+}$ determines a stochastic threshold for postsynaptic firing. ", "The negative log odds triggers LTD until the synaptic weight reaches its minimum efficacy. ", "The minimum efficacy is when the presynaptic neuron does not have an effect on the postsynaptic membrane potential (disconnected synapse).", "\n\nA number of previous empirical studies have provided evidence that Bayesian analysis of sensory stimuli occurs in the brain [@Doya2007a; @Kording2004a; @Rao2002a]. ", "In Bayesian inference, the hidden causes are inferred using prior knowledge and the likelihood of the observations to obtain a posterior probability. ", "In the case of two classes $M_1$ and $M_2$, the posterior probability of class $M_1$ is introduced by the sigmoid function as follows [@Bishop2006a] $$p(M_1|x)=\\frac{1}{1+e^{-a}}$$ where $a$ is $$a=\\mathrm{ln} \\frac{p(x|M_1)p(M_1)}{p(x|M_2)p(M_2)}$$ which is interpreted as log odds by $$a=\\mathrm{ln} \\frac{p(M_1|x)}{p(M_2|x)}=\\mathrm{ln} \\frac{p(M_1|x)}{1-p(M_1|x)}\n\\label{eq:oddBishop}$$ Equation (\\[eq:oddBishop\\]) is analogous to the synaptic weight in (\\[eqn:equilibrium\\]).", "\n\nLimitations and convergence\n---------------------------\n\nWe consider the maximum possible weight value after a series of the LTP events during training. ", "Does an excitatory synaptic weight converge to some constant or grow without bound as a function of LTP events? ", "To address this, we start with initial weight, $w^0$. The next synaptic weights after performing $n$ LTP rules are $w^1, w^2, ..., w^n$. According to (\\[AT\\_STDP\\]), $$w^{i+1}=w^i+a^+\\cdot e^{-w^i}\n\\label{Convergence2}$$ The maximum weight change happens when $w^i=0$. So, if we suppose $w^0=0$ (maximum change), the final synaptic weight is obtained by: $$\\label{Wn1}\n\\begin{array}{l}\nw^n=w^0+a^+\\cdot e^{-w^0}+...+a^+\\cdot e^{-w^{n-1}}=a^+\\cdot \\sum_{i=0}^{n-1}e^{-w^i}\n\\end{array}$$ And to find an upper bound for the synaptic weight in $n$ LTP operations, we have: $$\\label{upperbound1}\n\\begin{array}{l}\nw^1=a^+, \\\\ \nw^2=a^+ + a^+\\cdot e^{-a^+} < 2a^+,\\\\\nw^3 < 2a^++a^+\\cdot e^{-2a^+} < 3a^+, \\\\ \nand ... \\\\ \nw^{n-1} < (n-2)a^++a^+\\cdot e^{-(n-2)a^+} < (n-1)a^+ \n\\end{array}$$ Therefore, $$w^n < a^+\\cdot \\sum_{i=0}^{n-1}e^{-ia^+}=a^+\\cdot \\frac{1-e^{-na^+}}{1-e^{-a^+}}\n\\label{upperbound2}$$ Finally, a synaptic weight after infinite iterations of the LTP adjustments (with $0<a^+<1$) will fall in range $(0, 1.582)$ where, $1.582$ is an upper bound if $a^+ = 1$.\n\nIzhikevich’s Neuron Model\n=========================\n\nIzhikevich’s model for the RS spiking neuron concisely, but accurately, captures the spike-generation dynamics of one subtype of pyramidal neuron [@Izhikevich2003a; @Izhikevich2007a]. ", "We used a simplified version of this so that it would be compatible with the SNN implementation (We only use the first generated spike). ", "Its spike-generation dynamics is specified in Equation Set (\\[eqn:Izhikevich1\\]) using state variables $V$ and $U$.\n\n\\[eqn:Izhikevich1\\] $$\\begin{aligned}\nC\\frac{dV}{dt}&=k(V-V_{\\mathrm{rest}})(V-V_{\\mathrm{th}})-U+I_{\\mathrm{tot}}\\\\\n\\frac{dU}{dt}&=a[b(V-V_{\\mathrm{rest}})-U]%\\nonumber\\end{aligned}$$\n\n$V$ denotes the membrane potential and $U$ specifies a recovery factor and keeps the membrane potential near the resting value, $V_{\\mathrm{rest}}$.\n\nThe many different parameters: capacity ($C$), threshold ($V_{\\mathrm{th}}$), $a$, $b$, $c$, $d$, and $k$ customize the spike-generation dynamics. ", "The model parameters to obtain an RS spiking neuron are: $a=0.03, b=-2, U_0=0, \nV_\\mathrm{rest}=0, V_\\mathrm{th}=0.49, C=100, K=0.7$; where $V_\\mathrm{th}$ and $V_\\mathrm{rest}$ were changed according to the model requirements.", "\n\nThe spike time occurs when $V$ reaches $V_\\mathrm{th}$. $I_{\\mathrm{tot}}$ is set to the net synaptic input.", "\n\nExperiments and Results\n=======================\n\nWe performed two experiments. ", "Both experiments compared four network architectures. ", "Specifically, we studied four types of the 5-layer SNNs as follows:\n\n- [**SNN 1:** The original 5-layer convolutional network of IF spiking neurons in [@Masquelier2007a] with the original parameters.]{}", "\n\n- [**SNN 2:** The SNN equipped with our probabilistic STDP learning rule.]{}", "\n\n- [**SNN 3:** The original SNN using Izhikevich-like neurons instead of IF neurons. ]{}", "\n\n- [**SNN 4:** The SNN using Izhikevich-like neurons and equipped with the probabilistic STDP rule.]{}", "\n\nIn all of the SNNs, the learning parameter $a^+=2^{-6}$ at the beginning and is multiplied by $2$ every $400$ postsynaptic spikes, until a maximum value of $2^{-2}$. $a^-$ is adjusted such that $a^+/a^-=4/3$.\n\n![", "Sample images from the Caltech dataset. ", "The top row shows face examples and the bottom row shows background examples.[]{data-label=\"fig:noNoisesamples\"}](DataSet.jpg){width=\"8.5cm\"}\n\nExperiment 1\n------------\n\n### Method\n\nTo compare with the original work [@Masquelier2007a], we evaluated the SNN using a subset of the Caltech dataset (http://www.vision.caltech.edu/archive.html) containing faces (positive class) and background images (negative class) that was used in the original model. ", "We also compared motorbikes to background images. ", "Images were converted to gray-scale as described in [@Masquelier2007a]. ", "Examples images for face and background appear in Fig.", " \\[fig:noNoisesamples\\] (more examples, including motorbikes, are given in [@Masquelier2007a]). ", "The dataset was equally divided into training and testing sets.", "\n\nIn one simulation, the SNNs were applied on the face/background (non-face) training samples. ", "In a second simulation, the bike/background data was used.", "\n\nThese simulations produced ten class-specific C2 features. ", "Each C2 cell represents a combination of edges in respect to an input image. ", "A representation of the input images can be reconstructed by convolving the weight matrix with a set of kernels representing oriented bars [@Masquelier2007a]. ", "During the training process, initial iterations do not show a clear reconstructed preferred stimulus. ", "After training, the C2 features are able to represent the face images. ", "Thus, the C2 features developed selectivity to face features. ", "Learning was stopped after 1000 iterations while the weight matrices were saved after each 100 iterations to track the learning process.", "\n\nThe trained network was evaluated using the testing set. ", "Following [@Masquelier2007a], we used two measures for evaluating the model: 1) the accuracy rate when the false positive rate equals the missed rate (equilibrium point); and, 2) A term of area under the receiver operator characteristic (ROC).", "\n\n### Results and discussion\n\nFig.", " \\[fig:barchartsAcc\\] shows the maximum accuracy rate obtained from the SNN models (**SNN 1** through **SNN 4**). ", "It also includes standard errors of the mean. ", "These were computed by running each simulation nine times ($n=9$) with a different set of initial values for the weights. ", "Different subsamples for training and testing data were not used in this first experiment but they were used in second experiment. ", "Fig.", " \\[fig:barchartsAcc\\] shows a trend that the probabilistic STDP rule gives better accuracy regardless of whether the IF neuron (**SNN1** vs **SNN2**) or the Izhikevich-like neuron is used (**SNN3** vs **SNN4**). ", "The overall performance for the motorbikes data was lower. ", "Fig.", " \\[fig:barchartsROC\\] illustrates the same pattern of results for the maximum ROC obtained from the SNN models for both face and motorbike datasets.", "\n\nTwo-tailed t-tests were performed to assess the statistical significance of the accuracy trends. ", "The following comparisons were performed: **SNN1** versus **SNN2**, **SNN3** versus **SNN4**, and **SNN1** versus **SNN4** for both faces and motorbikes. ", "The results are shown in Figs.", " \\[fig:barchartsAcc\\] and \\[fig:barchartsROC\\] in terms of $p<0.1$ (\\*), $p<0.05$ (\\*\\*), $p<0.01$ (\\*\\*\\*), and $p<0.005$ (\\*\\*\\*\\*). ", "The differences reached more significance for the network that used Izhikevich-like neurons. ", "Thus, we claim that our probabilistic variant STDP rule significantly improves performance when Izhikevich-like neurons are used, either with faces or motorbikes.", "\n\n![", "Summary of performance for face and bike recognition in term of accuracy rate in equilibrium point (%).[]{data-label=\"fig:barchartsAcc\"}](ACC_Bar.jpg){width=\"7cm\"}\n\n![", "Summary of performance for face and bike recognition in term of area under the ROC curve (%).[]{data-label=\"fig:barchartsROC\"}](ROC_Bar.jpg){width=\"7cm\"}\n\nFig.", " \\[fig:reconstruction\\] shows reconstructions for the ten C2 cells on the face data for the four SNNs. ", "They indicate that face-like features were discovered by all SNN variants. ", "Thus, that property of the original SNN in [@Masquelier2007a] is shown to be robust under the changes introduced in this experiment.", "\n\n![", "Reconstructions for the 10 features having developed selectivity to face features. (", "A) through (D) specify the reconstructions of **SNN 1** through **SNN 4** respectively.[]{data-label=\"fig:reconstruction\"}](SNN_Constructors_All.jpg){width=\"8.5cm\"}\n\nFig.", " \\[fig:reconstructionIter\\] shows the reconstructions for a selected face feature during the training process (100 to 1000 iterations) for each of the SNNs. ", "Qualitatively, the results are similar for all of the SNNs.", "\n\n![", "Reconstructions of a selected face feature for **SNN 1** through **SNN4** during the training process in 100 through 1000 iterations.[]{data-label=\"fig:reconstructionIter\"}](Iteration_Constructor.jpg){width=\"8.5cm\"}\n\nFig.", " \\[fig:cmp3\\] shows the accuracy results for all of SNNs applied to the faces data. ", "The most salient trend is that **SNN 3** and **SNN 4**, which both use the Izhikevich-like units, perform better relatively early in the training, at iterations 200 and 300, but the advantage is lost later in the training. ", "Also, **SNN 3** seems to perform slightly worse than the others. ", "Standard errors were not computed, but this was done in Experiment 2.", "\n\nExperiment 2\n------------\n\nThe results in Experiment 1 need further exploration because, first, the results are possibly compromised by ceiling effects (especially for the faces data) and, second, we did not perform cross-validation by using different samples of training and testing data.", "\n\n### Method\n\nTo address the possible ceiling effect, we used noisy test images (Gaussian noise with zero mean and standard deviation of $0.3$) for the testing phase. ", "The purpose was to reduce the accuracy rate so that ceiling effects were no longer an issue. ", "Five sets of runs ($n=5$) were performed for both faces and motorbikes. ", "Both testing and training images sets had 218 items. ", "For each simulation, 175 items were sampled from each set to be used in that simulation. ", "Standard errors were computed for each of 40 data points.", "\n\n### Results and discussion\n\nFig.", " \\[fig:noisyFaces\\_All\\] shows the face results for all of SNNs (**SNN 1** through **SNN 4**) including standard errors. ", "Fig.", " \\[fig:noisyMotorbikes\\_All\\] shows the results for motorbikes. ", "In both figures the accuracy rates are below 97 percent, so using the noisy images eliminated the complication of a ceiling effect. ", "For the faces (Fig.", " \\[fig:noisyFaces\\_All\\]), all of the trends shown in Fig.", " \\[fig:cmp3\\] are preserved in these results. ", "Most notably, the trend where the probabilistic STDP variant improves accuracy regardless of whether IF units or Izhikevich-like units are used still persists. ", "For the motorbikes, the proposed probabilistic STDP performs better; however, the accuracy performances are close across all manipulations. ", "From the data taken as a whole, we can also conclude that the original model is robust across the manipulations.", "\n\n![", "Accuracy rate of **SNN 1** through **SNN 4** at equilibrium point in 100 through 1000 learning iterations. (", "Experiment 1)[]{data-label=\"fig:cmp3\"}](AJAB1.jpg){width=\"8.5cm\"}\n\n![", "Accuracy of face/non-face recognition for noisy images applied to **SNN 1** through **SNN 4** for 100 through 1000 learning iterations. (", "Experiment 2)[]{data-label=\"fig:noisyFaces_All\"}](AJAB2.jpg){width=\"8.5cm\"}\n\n![", "Accuracy rates of motorbike/non-motorbike recognition for noisy images applied to **SNN 1** through **SNN 4** for 100 through 1000 learning iterations. (", "Experiment 2)[]{data-label=\"fig:noisyMotorbikes_All\"}](AJAB3.jpg){width=\"8.5cm\"}\n\nControl Experiments\n-------------------\n\nExperiments 1 and 2 strongly suggest that the proposed probabilistic STDP rule improves the network performance in comparison to the original one. ", "We ran two additional control experiments to improve our confidence in this result. ", "The first experiment controlled for weight selection and the second experiment tested several other object categories.", "\n\nFigs.", " \\[fig:reconstruction\\] and \\[fig:reconstructionIter\\] show more selected weights for probabilistic STDP (lines in face features). ", "Hence, is the advantage of the probabilistic rule because it selected more weights? ", "To address this, ten experiments with different $a^+/a^-$ ratios were run on the face recognition task to assess the effect of the number of selected weights on accuracy. ", "Fig.", " \\[fig:weightsvsacc\\] shows the accuracy rate versus average number of the selected weights. ", "This figure shows that the probabilistic rule consistently outperforms the classic rule for a given number of selected weights.", "\n\n![", "Accuracy rates of **SNN 1** and **SNN 2** versus average number of selected weights after training. ", "Data points were obtained by 10 different training runs for each SNN with $a^+/a^-=4/9, 4/8, 4/7, 4/6, 4/5, 4/4, 4/3, 4/2, 4,$ and $8$. The trend lines show better performance of the probabilistic variation of the STDP introduced in this paper.[]{data-label=\"fig:weightsvsacc\"}](weightsvsacc.jpg){width=\"8.5cm\"}\n\nAdditionally, a subset of the 3D-Object images provided by Savarese et al. ", "at CVGLab, Stanford University [@Savarese2007a] was used for new experiments. ", "The accuracy rates of **SNN 1** through **SNN 4** are shown in Fig.", " \\[fig:3dres\\]. ", "This figure shows better performance of probabilistic STDP (**SNN 2** and **SNN 4**) than original one (**SNN 1** and **SNN 3** respectively) in 12 out of 16 cases.", "\n\n![", "Accuracy rates of the SNNs for recognizing eight sets of 3D-objects provided in [@Savarese2007a].[]{data-label=\"fig:3dres\"}](3dresults.jpg){width=\"10cm\"}\n\nConclusion\n==========\n\nThis paper studied the performance of the spiking network given in [@Masquelier2007a] when the original STDP learning rule was replaced with a novel rule that had a probabilistic interpretation and also when the model neuron was changed from an IF neuron to an Izhikevich-like neuron. ", "Our findings are the following. ", "First, our experiments introduced a probabilistic STDP variant which improved the accuracy rate. ", "Most notably, the magnitude of the weight adjustments for LTP was an exponential function of the weight magnitude. ", "This is an entirely different type of rule. ", "Second, when we replaced the non-leaky IF neurons with Izhikevich-like neurons, accuracy performance remained robust. ", "This is significant because the IF neuron is the simplest possible spike generator whereas the Izhikevich neuron has a much more complicated spike generation mechanism.", "\n\nOur main conclusion is that our probabilistic variant of STDP improved performance for instances of the model that used either IF neurons or Izhikevich-like neurons. ", "A secondary conclusion is that the original model is robust against more biologically realistic neurons (Izhikevich-like RS neurons). ", "Although Izhikevich-like neuron did not improve the performance, considering the nature of the manipulations, new knowledge has in fact been obtained. ", "We are not aware of other work that has studied variations of this model.", "\n\nA priori, one would expect that the most likely effect of any drastic change, like those listed above, would impair network performance, but instead, the performance remained robust. ", "Future researchers experimenting with this model, for instance to build a deep spiking network, can have some measure of confidence that the performance of the individual S/C modules will remain robust when placed in a larger context.", "\n\nSerre, T., Wolf, W., Bileschi, S. Riesenhuber, M., Poggio, T., Robust object recognition with cortex-like mechanisms, *IEEE Transactions on Pattern Analysis and Machine Intelligence*, 29(3), 411-426, 2007.", "\n\nLeCun, Y., Bengio, Y., Convolutional networks for images, speech, and time series. ", "In *Handbook of brain theory and neural networks (1st ed)*, 255-258, MIT Press, 1995.", "\n\nFukushima, K., Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. *", "Biological Cybernetics*, 36, 193-202, 1980.", "\n\nThorpe, S., Fize, D., Marlot, C., Speed of processing in the human visual system. *", "Nature*, 381, 520-522, 1996.", "\n\nMasquelier, T., Thorpe. ", "S. J., Unsupervised learning of visual features through spike timing dependent plasticity. *", "PLoS Computational Biology*, 3(2), e31, 247-257, 2007.", "\n\nKheradpisheh, S. R., Ganjtabesh, M., Masquelier, T., Bio-inspired Unsupervised Learning of Visual Features Leads to Robust Invariant Object Recognition. ", "arXiv preprint arXiv:1504.03871, 2015, to appear in *Neurocomputing*.", "\n\nKrizhevsky, A., Sutskever, I., Hinton, G., Image[N]{}et classification with deep convolutional neural networks. *", "Advances in neural information processing systems (NIPS) 25*, 1097-1105, 2012.", "\n\nIzhikevich, Eugene M., Simple model of spiking neurons. *", "IEEE Transactions on neural networks*, 14(6), 1569-1572, 2003.", "\n\nIzhikevich, Eugene M., *Dynamical systems in neuroscience*. ", "MIT press, 2007.", "\n\nMasquelier, Timothée, and Simon J. Thorpe, Learning to recognize objects using waves of spikes and spike-timing-dependent plasticity. *", "The 2010 International IEEE Joint Conference on Neural Networks (IJCNN)*, 2010.", "\n\nNessler, B., Pfeiffer, M., Maass, W., STDP enables spiking neurons to detect hidden causes of their inputs. *", "Advances in neural information processing systems*, 1357-1365, 2009.", "\n\nNessler, B., Pfeiffer, M., Buesing L., Maass, W., Bayesian computation emerges in generic cortical microcircuits through spike-timing-dependent plasticity. *[", "PL]{}o[S]{} Computational Biology*, 9(4): e1003037, 1-30, 2013.", "\n\nMarkram, H., Gerstner, W., Sjöström, P. J., Spike-timing-dependent plasticity: A comprehensive overview. *", "Frontiers in Synaptic Neuroscience*, 4(2), 1-3, 2011.", "\n\nSong, Sen, and Larry F. Abbott, Cortical development and remapping through spike timing-dependent plasticity. *", "Neuron*, 32(2), 339-350, 2001.", "\n\nGuyonneau, R., VanRullen, R., Thorpe, S., Neurons tune to the earliest spikes through STDP. *", "Neural Computation*, 17, 858-579, 2005.", "\n\nMasquelier, Timothée, Rudy Guyonneau, and Simon J. Thorpe, Competitive STDP-based spike pattern learning. *", "Neural computation*, 21(5), 1259-1276, 2009.", "\n\nDoya Kenji, *Bayesian brain: Probabilistic approaches to neural coding*, MIT press, 2007.", "\n\nKörding, Konrad P., and Daniel M. Wolpert, Bayesian integration in sensorimotor learning. *", "Nature*, 427, 244-247, 2004.", "\n\nRezende, D. J., Wierstra, D., Gerstner, W., Variational learning for recurrent spiking networks. *", "Advances in Neural Information Processing Systems*, 136-144, 2011.", "\n\nJolivet, R., Rauch, A., Lüscher, H. R., Gerstner, W, Predicting spike timing of neocortical pyramidal neurons by simple threshold models. *", "Journal of computational neuroscience*, 21(1), 35-49, 2006.", "\n\nKappel, David, Bernhard Nessler, and Wolfgang Maass, STDP installs in winner-take-all circuits an online approximation to hidden [M]{}arkov model learning. *[", "PL]{}o[S]{} Computational Biology*, 10(3): e1003511, 1-22, 2014.", "\n\nTavanaei, Amirhossein, and Anthony S. Maida, Studying the interaction of a hidden Markov model with a Bayesian spiking neural network. *", "IEEE 25th International Workshop on Machine Learning for Signal Processing (MLSP)*, Boston, MA, USA, 2015.", "\n\nBland, J. Martin, and Douglas G. Altman, Statistics Notes: The odds ratio. *", "BMJ*, 320(7247), 1468, 2000.", "\n\nRao RPN, Olshausen BA, Lewicki MS, *Probabilistic Models of the Brain*. ", "MIT Press, 2002.", "\n\nBishop, Christopher M., *Pattern recognition and machine learning*. ", "Springer, New York, 2006.", "\n\nSavarese S., Fei-Fei L., 3D generic object categorization, localization and pose estimation, *IEEE 11th International Conference on Computer Vision (ICCV)*, 1–8, 2007.", "\n\n[^1]: tavanaei@louisiana.edu\n\n[^2]: timothee.masquelier@alum.mit.edu\n\n[^3]: maida@cacs.louisiana.edu\n\n[^4]: 2016 IEEE. ", "Personal use of this material is permitted. ", "Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.", "\n" ]

{ "pile_set_name": "ArXiv" }

[ 0.0006445152102969587, 0.0006595486192964017, 0.000619100930634886, 0.000563603884074837, 0.0005405564443208277, 0.0008427341817878187, 0.0006744723068550229, 0.000578354112803936, 0.000852357828989625, 0.000586442241910845, 0.0005377110792323947, 0.0010640104301273823, 0.000566322123631835, 0.0006311860051937401, 0.0006284427945502102, 0.0008711663540452719, 0.0005573268281295896, 0.0006230569561012089, 0.0005972451763227582, 0.0005529586342163384, 0.0005267694941721857, 0.0006251187296584249, 0.000571601907722652, 0.0005674903513863683, 0.0005696580628864467, 0.0007904596859589219, 0.0006033056415617466, 0.0005766909453086555, 0.0007556169293820858, 0.000610583636444062, 0.0005664835916832089, 0.000611916824709624, 0.0005534804076887667, 0.0005503291031345725, 0.000550821830984205, 0.0008757184259593487, 0.0005886968574486673, 0.0005483647692017257, 0.0008620157022960484, 0.000637025514151901, 0.0006627861876040697, 0.000539036060217768, 0.0005927091115154326, 0.0005632465472444892, 0.0005666450597345829, 0.000657017168123275, 0.0008306764648295939, 0.000563350971788168, 0.0005637919530272484, 0.0005375106702558696, 0.0006249686703085899, 0.0005900908727198839, 0.0006682172534056008, 0.0006093830452300608, 0.000564809306524694, 0.0005491953925229609, 0.0005633820546790957, 0.000584011955652386, 0.0005668300436809659, 0.0006639633793383837, 0.0005870730965398252, 0.0006124524516053498, 0.0006591995479539037, 0.0007559413788840175, 0.0006160666816867888, 0.0006207263213582337, 0.0006786782760173082, 0.0005624672630801797, 0.0005443712580017745, 0.0007497614715248346, 0.0009687762358225882, 0.0006871219375170767, 0.0007101450464688241, 0.0005620160372927785, 0.0006326456204988062, 0.0005650459206663072, 0.0007725886534899473, 0.0006182583165355027, 0.0006731148459948599, 0.000621089362539351, 0.0005823014071211219, 0.0010485027451068163, 0.0006409923080354929, 0.0007431615376845002, 0.0006210201536305249, 0.00069630634970963, 0.0005717177409678698, 0.000859270163346082, 0.0005654543056152761, 0.0019054226577281952, 0.0006317032966762781, 0.0005936891539022326, 0.0006043766043148935, 0.0005735690938308835, 0.0006528407684527338, 0.0005736680468544364, 0.0006068615475669503, 0.0005438516382128, 0.0006398517871275544, 0.0006577448220923543, 0.0202504750341177, 0.002327066147699952, 0.0006613727891817689, 0.0014927037991583347, 0.0010206904262304306, 0.000632844923529774, 0.0006846291944384575, 0.0006301969406194985, 0.0007075563771650195, 0.0005833741743117571, 0.006442993879318237, 0.0006692226161248982, 0.0005902844131924212, 0.0006411737413145602, 0.001049865037202835, 0.0006714947521686554, 0.0006404918385669589, 0.007275629788637161, 0.0005496229860000312, 0.0007183507550507784, 0.0006258475477807224, 0.0006292189937084913, 0.000553001300431788, 0.0008074012584984303, 0.0009251229930669069, 0.1607530117034912, 0.0009230430005118251, 0.0010203924030065536, 0.0007342079188674688, 0.000703513331245631, 0.0016305673634633422, 0.0009637971525080502, 0.0007023015641607344, 0.0009975124849006534, 0.0005919686518609524, 0.0006256522028706968, 0.004841722548007965, 0.0005796379409730434, 0.0008616199484094977, 0.007003344129770994, 0.0006452263914979994, 0.0010482791112735868, 0.003791102906689048, 0.002513969549909234, 0.0006576376617886126, 0.000577156082727015, 0.0006111477850936353, 0.0006967929075472057, 0.0008667309884913266, 0.0009054344845935702, 0.00125025468878448, 0.0005851812311448157, 0.0005996398977003992, 0.0006604515365324914, 0.00058875692775473, 0.0005532844224944711, 0.0006795050576329231, 0.0005645600613206625, 0.0005758359329774976, 0.0005604540347121656, 0.0006139371544122696, 0.0005460151587612927, 0.0005660275346599519, 0.0005728121614083648, 0.0005414111656136811, 0.0005907802260480821, 0.0005905632860958576, 0.0005427516880445182, 0.0005591337685473263, 0.0007207099697552621, 0.0006926974165253341, 0.000763148651458323, 0.0006074240081943572, 0.0005690058460459113, 0.000829310214612633, 0.0013373462716117501, 0.0006807797471992671, 0.000829310214612633, 0.0006580034969374537, 0.0005837954813614488, 0.0021340905223041773, 0.00058859953423962, 0.0019213129999116063, 0.0005927184247411788, 0.0006298039224930108, 0.0019054226577281952, 0.0006038028514012694, 0.0005763268563896418, 0.0006727749132551253, 0.0007642497657798231, 0.0005920942639932036, 0.0019054226577281952, 0.0005665693897753954, 0.0007407706580124795, 0.000608624133747071, 0.0009727649739943445, 0.0019054226577281952, 0.0006283536204136908, 0.0006492587272077799, 0.000804231851361692, 0.011573641560971737, 0.0006301893736235797, 0.0005660804454237223, 0.0006081921164877713, 0.0008404927793890238, 0.0009688141872175038, 0.0005722421919927001, 0.0005676602013409138, 0.0006750166648998857, 0.0007207099697552621, 0.0023781678173691034, 0.000829310214612633, 0.0008828825666569173, 0.0006254796171560884, 0.0006807135650888085, 0.0007783395703881979, 0.0006788551690988243, 0.0006339728715829551, 0.0006262118113227189, 0.0005389885627664626, 0.0019054226577281952, 0.0007019395707175136, 0.0006560636102221906, 0.0006626747199334204, 0.0006714531336911023, 0.0007226816960610449, 0.000592305907048285, 0.0005358288181014359, 0.0005916866357438266, 0.0007231991039589047, 0.0006836446700617671, 0.0006858909036964178, 0.0006456224946305156, 0.000829310214612633, 0.0006262389360927045, 0.0006162293720990419, 0.0019054226577281952, 0.0007247542380355299, 0.0006138871540315449, 0.0005729230470024049, 0.0007311468943953514, 0.0009309101733379066, 0.0007624518475495279, 0.0019054226577281952, 0.0006331477779895067, 0.0006164264632388949, 0.0005841937381774187, 0.0005946105229668319, 0.0007009489345364273, 0.0007711330545134842, 0.000807201664429158, 0.0006123785860836506, 0.0006417114054784179, 0.0006597982137463987, 0.0005750845302827656, 0.00056502252118662, 0.0005204994231462479, 0.0006859268760308623, 0.0006524299387820065, 0.0007367131765931845, 0.0005933577194809914, 0.0007252329960465431, 0.0006439298158511519, 0.000618160585872829, 0.000745507248211652, 0.0006348773604258895, 0.0006223880336619914, 0.0006514054257422686, 0.0011895117349922657, 0.000647537992335856, 0.0006131109548732638, 0.0007840157486498356, 0.0006071730749681592, 0.000626727647613734, 0.000693037873134017, 0.000687693536747247, 0.000568477320484817, 0.0008054111385717988, 0.0006359760300256312, 0.0006987472297623754, 0.0006573434802703559, 0.000620444247033447, 0.000598185695707798, 0.0006674292962998152, 0.0006768177263438702, 0.0006390297203324735, 0.0006830986239947379, 0.0006439173012040555, 0.0006328030140139163, 0.0007219541585072875, 0.0006595225422643125, 0.0007040536729618907, 0.0007193769561126828, 0.0006246140692383051, 0.0007150395540520549, 0.0006138505996204913, 0.0008173605310730636, 0.0006619106279686093, 0.0006335008074529469, 0.0005757638136856258, 0.0006299885571934283, 0.0006888062926009297, 0.000927812303416431, 0.0006856703548692167, 0.0006277517531998456, 0.0006674731848761439, 0.000632396899163723, 0.0008329369593411684, 0.0005545389140024781, 0.000581527769099921, 0.001994825666770339 ]

[ "chelsea sweet - i am a horny hot blond chick, so come over and let's just fuckSex Category: Babes\n\n0\n\n0\n\n5\n\nDani Wild - Party girl with natural boobs and bubble butt! ", "I am the real thing and love to raise a little hell and have some fun. ", "I am spunky and outgoing, and yes - I am exhibitionist. ", "Come to see me play!Sex Category: Babes\n\namanda - hi i am amanda - i like to have sex and show you how to make loveSex Category: Babes\n\n0\n\n0\n\n8\n\nhalo sweet - I am a firey redhead in every sense of the word! ", "I am very sexual and love trying new things! ", "Wanna watch me? ", "I can be your escort to fulfilling your ultimate sex dream desire!Sex Category: Babes" ]

{ "pile_set_name": "Pile-CC" }

[ 0.9758355021476746, 0.12564274668693542, 0.0009572634007781744, 0.4412011206150055, 0.13991966843605042, 0.017205968499183655, 0.7950841784477234 ]

[ "By now, this news is a few days old, but there was something I wanted to check on before noting it. ", "Here’s the story:\n\nNASA’s Kepler spacecraft has discovered the first confirmed planetary system with more than one planet crossing in front of, or transiting, the same star.", "\n\nThe transit signatures of two distinct planets were seen in the data for the sun-like star designated Kepler-9. ", "The planets were named Kepler-9b and 9c. ", "The discovery incorporates seven months of observations of more than 156,000 stars as part of an ongoing search for Earth-sized planets outside our solar system. ", "The findings will be published in this week’s issue of the journal Science.", "\n\nKepler’s ultra-precise camera measures tiny decreases in stars’ brightness that occur when a planet transits them. ", "The size of the planet can be derived from these temporary dips.", "\n\nThe distance of the planet from a star can be calculated by measuring the time between successive dips as the planet orbits the star. ", "Small variations in the regularity of these dips can be used to determine the masses of planets and detect other non-transiting planets in the system.", "\n\nWhat struck me as interesting about this is how fast the planets are orbiting, which, apparently, related to how close in to the star they are. ", "I wonder what the typical planetary system looks like, in terms of the distribution of larger vs. smaller planet. ", "This could have implications as we play the game of messing around with Drake equation like calculations (but for life in general).", "\n\nWould it be possible to assert that a small (“Class M” if you will) planet close enough to a certain size star is more likely to have life, and that large gaseous Jovian planets swooping around in the inner regions of a solar system would reduce the likelihood of that? ", "Or, are Jovian ‘inner’ planets likely to have numerous “Class M” satellites, like this one?", "\n\nMy understanding (I asked NASA) is that we basically don’t know this at the present time but characterizing the overall pattern of ‘solar systems’ is an objective of the mission. ", "So stay tuned.", "\n\nRelated\n\nComments\n\nWould it be possible to assert that a small (“Class M” if you will) planet close enough to a certain size star is more likely to have life, and that large gaseous Jovian planets swooping around in the inner regions of a solar system would reduce the likelihood of that?", "\n\nYes, mostly. ", "At least, life based on the kind of chemistry our life uses. ", "We need liquid water, which puts restrictions on temperature – not too hot, not too cold – the “Goldilocks zone”.", "\n\nIt’s generally assumed that wild swings in temperature would be problematic for getting life started, too – so eccentric orbits are presumed to be less hospitable. ", "And big Jovians close to the star can destabilize smaller planets, which would tend to kick them out of the “Goldilocks zone”.", "\n\nOf course, these assumptions may be too restrictive. ", "There may be “life as we don’t know it.” ", "But until we find some, we won’t know about it. ", "🙂\n\nMy intuition tells me that our current methods for detecting exoplanets are biased in favor of close-in planets. ", "Our main methods are by detecting the gravitational pull (proportional to 1/r^2 where r is the distance from the star to the planet’s orbit) or by detecting a transit (geometry favors close-in planets as they are more likely and more frequently aligned). ", "I know of only one system that has been detected by direct imaging, and that is the very nearby Fomalhaut system, where we can resolve the angular separation between the star and its planets in Neptune-like orbits. ", "It’s also obvious that all three methods favor big planets, which are brighter, more massive, and more likely to be observed transiting.", "\n\nI’ve been trying to put together a presentation on exoplanets and the subject keeps moving faster than I can keep up. ", "From what I read, the large, closely orbiting hot jupiters are thought to have migrated inward from Jupiter-to-Neptune orbits. ", "This migration would disturb and probably destroy any earthlike planets. ", "So, I’m very eager to see earth-size planets in earth-like orbits where there is also a hot jupiter, as that would turn over the understanding of how these large planets got where there are and thus increase the chance of finding earthlike planets.", "\n\nAbout this blog\n\nThe science is mostly climate change, life science, evolution, and science education. ", "The science policy and politics is mostly about climate change and the evolution-creationism false controversy. ", "The technology is mostly about Linux, regular normal person computer use, kids programming, and now and then, household items.", "\n\nAs an Anthropologist and Africanist, I often write about those topics as well.", "\n\nUsually, I write about one or two topics for a while then shift to something different. ", "This is not systematic, and has to do with context and external forces such as: Is this a presidential election year? ", "Are we having an El Nino? ", "Is there a fight going on somewhere about teaching creationism in a public school? ", "Did I just get a cool new robot toy? ", "Like that.", "\n\nSo, if you don't find the most recent few posts interesting, have a look at the topics below. ", "But, of course, chances are you got here with a Google search and you'll never read this \"about\" thing." ]

{ "pile_set_name": "Pile-CC" }

[ 0.000558197614736855, 0.0006016486440785229, 0.0005912692286074162, 0.0006240550428628922, 0.0005716324085369706, 0.0005849066074006259, 0.0006495908019132912, 0.0007437004242092371, 0.0006743604317307472, 0.0006622513174079359, 0.0006010520155541599, 0.0005731373676098883, 0.0007340441807173193, 0.0007112431921996176, 0.0007917091716080904, 0.0005336691974662244, 0.000776499102357775, 0.0006523214979097247, 0.0007222947897389531, 0.0007304270984604955, 0.0018218758050352335, 0.0006128638051450253, 0.005721177440136671, 0.0006202576914802194, 0.001935113687068224, 0.0006271889433264732, 0.0007989952573552728, 0.000600638275500387, 0.0005742035573348403, 0.0006798454560339451, 0.0006225425167940557, 0.0005557063268497586, 0.011080224998295307, 0.000550940225366503, 0.0005713132559321821, 0.0006161635392345488, 0.000704420090187341, 0.0005758518818765879, 0.000579029438085854, 0.0006047635688446462, 0.0009215406607836485, 0.0008372626034542918, 0.0010578848887234926, 0.0013685534941032529, 0.0005646386416628957, 0.000759266025852412 ]

[ "Q:\n\nclassic asp multiple button click\n\nI have 2 asp pages.", "\n\nFeedback_Administration_Update.asp\nFeedback_Administration.asp\n\nMy HTML form:\n<form ACTION=\"Feedback_Administration_Update.asp\" METHOD=POST NAME=\"form\">\n <input type=\"button\" value=\"Submit\" onclick=\"javascript:Checkit();\" name=\"btnSubmit\">\n <input type=\"reset\" value=\"Reset\" onclick=\"javascript:document.location=location\">\n <input type=\"button\" value=\"Send Response to Requester\" name=\"btnSubmit\" onclick=\"javascript:Checkit();\"/>\n</form>\n\nCode in Feedback_Administration_Update.asp file:\ndim buttonPressed\nbuttonPressed=Request.", "Form(\"btnSubmit\")\nselect case buttonPressed\n case \"Submit\"\n 'update database code \n Response.", "Redirect \"feedback_administration.asp?ID=\" & regEx.", "Replace((request.form(\"ID\")), \"''\") & \"&lstPages=\" & Trim(Request(\"lstPages\")) & \"&view=\" & Request(\"view\")& \"&strSuccess1=\" &\"Success\"\n case \"Send Response to Requester\"\n 'mail task code \nEnd Select\n\nI am not sure why they are not working on clicking any button.", "\nWhen I click on submit the page, it should update the code, when I click on submit Send Response to Requester, it should send email.", "\n\nA:\n\nThe value of button inputs is not sent to the server. ", "You need to use ordinary submit buttons, and to have validation code you can cancel their onclick event:\n<input type=\"submit\" value=\"Submit\" onclick=\"return Checkit();\" name=\"btnSubmit\" />\n<input type=\"submit\" value=\"Send Response to Requester\" name=\"btnSubmit\" onclick=\"return Checkit();\" />\n\nTo make it work, change your JavaScript function in two ways. ", "First, don't make it submit the form and second have it return true upon successful validation and false otherwise, e.g.:\nfunction Checkit() {\n //validate your stuff..\n if ([all is good]) {\n //don't submit the form here, the submit button will do that.", "\n return true;\n }\n\n //indicate that something was wrong, cancel form submission:\n return false;\n}\n\nHaving this in place will cause Request.", "Form(\"btnSubmit\") to hold the desired value.", "\n\n" ]

{ "pile_set_name": "StackExchange" }

[ 0.0008372846059501171, 0.0007427428499795496, 0.0006166973034851253, 0.0006984709179960191, 0.0006344016874209046, 0.0005637430585920811, 0.0006692273309454322, 0.0007234889199025929, 0.0008237141300924122, 0.0010762772290036082, 0.0006820395938120782, 0.001994825666770339 ]

[ "Entrepreneurs thinking outside the box to help first home buyers jump on the property ladder is one way to help solve the housing affordability problem in cities such as Sydney, according to one NSW politician.", "\n\nAt the launch of digital property platform Kohab, which has been described as “the Tinder for real estate”, NSW parliamentary secretary for finance, service and property Alister Henskens said private initiatives such as these were allowing Sydneysiders build equity in real estate where they would otherwise be locked out of the market.", "\n\nKohab is an online marketplace that matches people with other buyers who want to co-own a property, either to live in or rent out as an investment, that would otherwise be unaffordable to purchase on their own.", "\n\n“[These platforms] will allow people to, in a more manageable way, start getting equity in real estate at an earlier point in time,” Mr Henskens said.", "\n\n“I’ve held four or five properties in my life, the nature of the exercise is that you want to get into the real estate market, build some equity and use that equity to buy your next property,” he said.", "\n\nWhen asked what more could be done to help with housing affordability in Sydney, Mr Henskens said the problem was that not enough people were prepared to move out to cheaper areas.", "\n\n“People in the 50s or 60s would go out west and travel into work. ", "Now people are less willing to do that … they want their first home to be close to the city and that’s obviously where the highest prices are,” he said.", "\n\nIn terms of the growing intergenerational divide of home ownership between Gen Y and their parents, Mr Henskens believed the combination of first home owner schemes, stamp duty concessions and entrepreneurial platforms like Kohab would “slowly remediate the gap over time”.", "\n\nKohab isn’t the first online platform to tap into the sharing economy to allow first home buyers to fulfil their dream of owning at a time when young Australians are increasingly priced out of the market.", "\n\nOther platforms that offer co-ownership of property include fractional investments like DomaCom, BrickX and CoVesta.", "\n\n“If you’ve got two or three people buying together who couldn’t have possibly bought on their own then all of a sudden they’re able to get into a property and they’re able to share the costs of that property,” Kohab founder Mr Dawson said. “", "Their need for a deposit is halved or [cut by a] third depending on how many people they buy with”.", "\n\nHe argued the model was different to existing fractional investment models because Kohab was not a fund manager or an investment platform and didn’t charge management or transaction fees.", "\n\n“From a platform point of view, we don’t operate through a trust – it’s real property which transfers to other people. ", "So effectively all parties could agree to sell the property on the open market and put it on the listing portal. ", "One person might want to buy the other person out and they’ll negotiate a price and they’ll move that ownership over. ", "It’s around having your name on the title and splitting the title according to your ownership,” Mr Dawson.", "\n\n“I think it comes down to practical basis, we’re basically saying two to four is fairly common but I guess five couples potentially buying something together as a second home or a holiday place, that’s possible as well,” Mr Dawson said.", "\n\nReal Estate Institute of New South Wales president Leanne Pilkington said this latest offering was a more-secure route for many family and friends who had already gone down that path in an ad hoc way.", "\n\n“I think a lot of parents will take a lot of advantage of it because we already see a lot of parents helping their children buy property. ", "What choice do you have when the median price in Sydney is a million dollars,” Mrs Pilkington said.", "\n\n“You can’t do it on your own as first home buyer so I think it will give families that security that they can do it properly,” she said.", "\n\nMr Dawson rejected claims his co-ownership business would further widen the gap between the “haves and have nots”.", "\n\n“I can’t see how that would happen because we’re moving people that are renting into buying and then obviously, there will be more rental properties available so I don’t see a knock-on effect at all,” he said." ]

{ "pile_set_name": "OpenWebText2" }

[ 0.0005784235545434058, 0.0006648311973549426, 0.0011775054736062884, 0.0005904320860281587, 0.0007953319582156837, 0.0006769861211068928, 0.0008163237362168729, 0.0008090973133221269, 0.0006390315247699618, 0.0008175444090738893, 0.0007119508809410036, 0.0008195848786272109, 0.0008439351222477853, 0.000640318205114454, 0.0006326826987788081, 0.0006281099049374461, 0.0009566625813022256, 0.0008749098051339388, 0.0005885504651814699, 0.0005747535615228117, 0.0007391194812953472, 0.0010186561848968267, 0.0008317140745930374, 0.0010058057960122824, 0.0008322491776198149 ]

[ "This invention relates to providing a system for improved risk assessment in market trading. ", "More particularly, this invention relates to providing a system for risk assessment from multiple independent risk factors across multiple time frames.", "\nIn trading any market vehicle, over a period of time, technicians and economists seek to identify risks as early as possible. ", "However, early pin-point detection, optimal sequencing, categorization, and confirmation of the risks are difficult in real-time (as they happen); this leads to delayed entry and exit in market trades. ", "Further, multiple confirmations of risks delay entries and exits in market trades even longer.", "\nA system is needed to identify and confirm risks in real-time, as the risks develop. ", "Further, a system is needed that will identify and confirm risks in user-desired time frames and at any stage of a trend." ]

{ "pile_set_name": "USPTO Backgrounds" }

[ 0.0005637838621623814, 0.0005633229739032686, 0.0005324208177626133, 0.000550011289305985, 0.0005903665442019701, 0.0005295120063237846, 0.0005393623723648489 ]

[ "Denis Bertolini\n\nDenis Bertolini (born 13 December 1977 in Rovereto) is an Italian former professional road racing cyclist.", "\n\nMajor results\n2001\n 1st Road race, Mediterranean Games\n2004\n 1st Stages 2 & 6 Circuit de Lorraine\n 1st Stage 2 Peace Race\n\nReferences\n\nCategory:1977 births\nCategory:Living people\nCategory:Italian male cyclists\nCategory:Mediterranean Games gold medalists for Italy\nCategory:Mediterranean Games medalists in cycling\nCategory:Competitors at the 2001 Mediterranean Games" ]

{ "pile_set_name": "Wikipedia (en)" }

[ 0.000933848787099123, 0.000606697634793818 ]

[ "Q:\n\nValue not being returned\n\nbeing new to this I am probably doing something silly somewhere which is breaking the code. ", "Have taken a look at some similar posts but found nothing that helps. ", "\nSo, the issue I am having is when I attempt to print newmarbles at the end it is saying it is not defined. ", "I assume I am doing something wrong in terms of returning the value? ", "Thanks in advance for any help. ", "\nimport random\n\ndef playNovice(marbles):\n aimove = random.randint(1, (marbles/2))\n print(\"AI Move\", aimove)\n newmarbles = marbles - aimove\n return newmarbles\n\ndef userPlay(marbles):\n usermove = int(input(\"Enter your move: \"))\n while usermove > marbles / 2 or usermove == 0:\n print(\"Invalid Move\")\n usermove = int(input(\"Enter your move: \"))\n else:\n newmarbles = marbles - usermove\n return newmarbles\n\ndifficulty = input(\"Which difficulty? ", "novice or expert: \")\nmarbles = 100\nplayNovice(marbles)\nuserPlay(marbles)\nprint(newmarbles)\n\nA:\n\nThe scope of newmarbles is local to playNovice(). ", "You should assign the returned value to a variable, which may have the same name:\nnewmarbles = playNovice(marbles)\nuserPlay(marbles)\nprint(newmarbles)\n\n" ]

{ "pile_set_name": "StackExchange" }

[ 0.01784900575876236, 0.0006063623004592955, 0.0007402197807095945, 0.0007220690022222698, 0.0005264387000352144, 0.0008204290643334389, 0.0006697569624520838, 0.0006489500519819558 ]

[ "/**\n * Licensed to the Apache Software Foundation (ASF) under one\n * or more contributor license agreements. ", " See the NOTICE file\n * distributed with this work for additional information\n * regarding copyright ownership. ", " The ASF licenses this file\n * to you under the Apache License, Version 2.0 (the\n * \"License\"); you may not use this file except in compliance\n * with the License. ", " You may obtain a copy of the License at\n *\n * http://www.apache.org/licenses/LICENSE-2.0\n *\n * Unless required by applicable law or agreed to in writing,\n * software distributed under the License is distributed on an\n * \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n * KIND, either express or implied. ", " See the License for the\n * specific language governing permissions and limitations\n * under the License.", "\n */\n\npackage org.apache.storm.spout;\n\nimport java.nio.", "ByteBuffer;\nimport java.util.", "List;\n\nimport org.apache.storm.tuple.", "Fields;\n\nimport static org.apache.storm.utils.Utils.tuple;\n\npublic class RawScheme implements Scheme {\n private static final long serialVersionUID = 6098042939916415521L;\n\n @Override\n public List<Object> deserialize(ByteBuffer ser) {\n byte[] bytes = new byte[ser.remaining()];\n ser.get(bytes);\n return tuple(bytes);\n }\n\n @Override\n public Fields getOutputFields() {\n return new Fields(\"bytes\");\n }\n}\n" ]

{ "pile_set_name": "Github" }

[ 0.0006217397167347372, 0.0005401495727710426, 0.0005986080504953861, 0.0005496976082213223, 0.0005568802589550614, 0.0014924280112609267, 0.0007840732578188181, 0.02987099252641201, 0.0018729271832853556 ]

[ "package scalatron.scalatron.impl\n\n/** This material is intended as a community resource and is licensed under the\n * Creative Commons Attribution 3.0 Unported License. ", "Feel free to use, modify and share it.", "\n */\n\nimport java.io.", "File\n\n\ncase class PluginDirectory(pluginDirectory: File) {\n case class VersionedPlugin(version: Int, plugin: File)\n\n val botfile_r = \"\"\"ScalatronBot_(\\d+)\\.jar\"\"\".r\n val botfile_f = \"ScalatronBot_%d.jar\"\n\n def mostRecentPlugin: Option[File] = {\n mostRecentPluginWithVersion.map(_.plugin)\n }\n\n def nextFilename: File = {\n val nextVersion = mostRecentPluginWithVersion.map(_.version).getOrElse(0) + 1\n val filePath = pluginDirectory.getCanonicalPath + \"/\" + botfile_f.format(nextVersion)\n new File(filePath)\n }\n\n private def mostRecentPluginWithVersion : Option[VersionedPlugin] = {\n val botnums = enumFiles\n\n if (botnums.isEmpty) {\n None\n } else {\n val most_recent = botnums.maxBy(_.version)\n Some(most_recent)\n }\n }\n\n private def enumFiles: Array[VersionedPlugin] = {\n val files = pluginDirectory.listFiles()\n val botnums = files.filter(file => botfile_r.findFirstIn(file.getName) match {\n case Some(num) => true\n case None => false\n }).map(file => botfile_r.findFirstIn(file.getName) match {\n case Some(num) => {\n val botfile_r(num2) = file.getName\n VersionedPlugin(num2.toInt, file)\n }\n })\n botnums\n }\n}\n" ]

{ "pile_set_name": "Github" }

[ 0.000633631250821054, 0.0005673175910487771, 0.0008099164115265012, 0.005044224672019482 ]

[ "module github.com/tetafro/godot\n\ngo 1.13\n" ]

{ "pile_set_name": "Github" }

[ 0.002369348658248782 ]

[ "Crabro cingulatus\n\nCrabro cingulatus is a species of square-headed wasp in the family Crabronidae. ", "It is found in Central America and North America.", "\n\nReferences\n\nFurther reading\n\n \n\nCategory:Crabronidae\nCategory:Insects described in 1867" ]

{ "pile_set_name": "Wikipedia (en)" }

[ 0.018193326890468597, 0.0010005959775298834, 0.0005939897382631898 ]

[ "Reference ranges and impact of selected confounders on classic serum and urinary renal markers in neonatal period.", "\nRenal injury in term and pre-term neonates may be an antecedent to chronic kidney disease in the child's further life. ", "The aim of the paper was to compile a reference range for selected serum and urinary renal markers in the neonatal period for full-term (FT) and pre-term (PT) newborns. ", "The prospective study included 23 FT infants (birth weight - BW≥2500g and gestational age - GA≥37Hbd) and 32 PT children (BW<2500g and GA<37Hbd) in good general condition, without acute kidney injury (AKI) or sepsis. ", "Between the 4th and 28th DOL, urinary concentrations of the studied renal markers (uCr, uNa, uOsm) were determined on a daily basis, while serum creatinine (SCr) was assessed minimum every 48-72h. ", "The mean GA and BW of the FT and PT infants were respectively as follows: 38.5±1.7Hbd; 3433±495.2g and 32.7±2.6Hbd; 1836.7±419.8g. For serum glomerular (SCr, eGFR) and tubular markers (FENa, RFI), the median values with normal ranges were compiled. ", "For urinary renal markers (uCr, uNa, uOsm) and those values standardized for kg of body weight, percentile tables for 4-28DOL were elaborated. ", "The study has resulted in determining the normal ranges of serum glomerular and tubular renal markers, as well as percentile tables of selected urinary renal parameters during the neonatal period. ", "The percentile tables may prove to be helpful for further standardization of other urinary parameters per urinary creatinine in neonatal population." ]

{ "pile_set_name": "PubMed Abstracts" }

[ 0.000578611041419208, 0.0031557593028992414, 0.0005980981513857841, 0.0019685097504407167, 0.0006034838734194636, 0.0006633510929532349, 0.0006591709679923952, 0.0006593760917894542, 0.0016496841562911868 ]

[ "(Photo by JIM WATSON/AFP via Getty Images)\n\n(CNSNews.com) - President Donald Trump celebrated having confirmed more than 150 judges to the federal bench during a ceremony at the White House on Wednesday.", "\n\n\n\n“From the beginning of my campaign, I promised to appoint judges who will adhere to the true and original meaning of our Constitution. ", "This afternoon, that promise was made, and that promise was kept, and I think many times over. ", "Later today, the Senate will vote on our 158th new judge,” Trump said.", "\n\n\n\n\n\n\n\n\n\n“We will have confirmed as I said 112 judges to the federal district courts, 44 judges to the federal circuit courts and two outstanding judges to the United States Supreme Court, and that number’s rapidly rising. ", "Now one out of every four circuit judges currently on the bench was appointed by this administration, and that number’s now exceeded that by quite a bit. ", "No president in history has confirmed so many judges - not even close - in such a short period of time,” he said.", "\n\n\n\nIn total, Trump has 158 judges appointed to the federal courts. ", "Approximately, one out of every four active judges on the U.S. Court of Appeals has been appointed by Trump.", "\n\n\n\nThe president also bragged about appointing judges who are younger than former President Barack Obama.", "\n\n\n\n“The average age of my newly appointed circuit court judges is less than 50. ", "They’re young, smart, and 10 years younger than President Obama’s nominees. ", "The pace of appointments is only accelerating. ", "It’s going very quickly, and then you know at some point, we’re going to have to slow it down, because we’re not going to have any, any openings,” Trump said.", "\n\n\n\n“We’re on track to have more judges confirmed this year than in the prior two years combined for us, and will be a record in history the number of judges we’re getting in just a short period of time,” the president said.", "\n\n\n\n“Once all pending nominations are confirmed, I will have appointed about 219 judges, and that number will go up very substantially from there, but the more gains we make, the more aggressive the hard left has become,” Trump warned.", "\n\n\n\nHe said liberals are using nationwide injunctions to invalidate the president’s executive actions.", "\n\n\n\n“One of their favorite new tactics is the nationwide injunction. ", "This very destructive practice enables any one of 673 district judges to single-handedly invalidate critical executive actions and stop the enforcement of duly enacted law,” Trump said.", "\n\n\n\n“We’ve had it many times, especially when it comes to our border, and when you see what’s going on on the other side of the border, people are starting to say maybe he is right about the wall, and maybe he is right about having a very strong border,” the president added.", "\n\n\n\n“The activist left only needs to find one partisan resistance judge anywhere in the country to agree with them. ", "On perhaps no issue have these district court injunctions been more brazen and damaging than in the area of immigration. ", "In case after case, judges are prohibiting us from enforcing the plain letter of the law,” he said.", "\n\n\n\nFor example, Trump said, a district judge’s ruling declaring the law that allows the expedited removal of illegal aliens apprehended within two years of entering the United States could only be applied to illegals caught within 14 days.", "\n\n\n\n“That’s a tiny group of people. ", "Just a small fraction of what the law expressly allows,” the president said.", "\n\n\n\nHe said the left wing’s assault on the Constitution has become a personal assault on judicial nominees themselves.", "\n\n\n\n“Nominees have been attacked in hearings for their religious beliefs. ", "Our entire nation was stunned by the cruel campaign of lies, and slander, and abuse that was hurled at Justice Brett Kavanaugh, but we refused to cave in,” Trump said. ", "But Republicans will never back down, he said.", "\n\n\n\n“While I'm your president, we will continue to nominate highly qualified judges who truly love our Constitution, and I know that Leader McConnell and Chairman Graham are absolutely committed to getting them confirmed quickly, and they have been going quickly,” the president said.", "\n\n\n\n“Of course, I'd like to get some other people confirmed too, please. ", "Some of the other people. ", "Mitch would say, ‘Well, we got to do the judges first.’ ", "And I always agree with him. ", "I have a lot of people waiting over there, but he says, ‘I want those judges.’ ", "And I agree with him,” Trump said.", "\n\n\n\n“Every judge we have placed on the bench will hear thousands of cases over their careers. ", "They will ensure equal justice for every party before them, regardless of age, race, income, religion, beliefs, or background. ", "And they will safeguard the birthright of liberty that belongs to every man, woman, child in our land,” the president added.", "\n\n\n\n\n\n" ]

{ "pile_set_name": "OpenWebText2" }

[ 0.0005468683084473014, 0.0005452483310364187, 0.0005691695841960609, 0.0007844376377761364, 0.0006071884417906404, 0.0007316183764487505, 0.0006639193743467331, 0.0007387596415355802, 0.0006736337090842426, 0.0005916167283430696, 0.0009761670953594148, 0.001025311998091638, 0.0006370099727064371, 0.0007618109812028706, 0.0005397566128522158, 0.0006943875341676176, 0.0017260080203413963, 0.000613694719504565, 0.0012949962401762605, 0.0005721031338907778, 0.0006624050438404083, 0.0006498232833109796, 0.0006614200538024306, 0.0006986716762185097, 0.0009819242404773831, 0.0005994904204271734, 0.0007807446527294815, 0.0009925143094733357, 0.059686388820409775, 0.025431308895349503, 0.0005360528011806309, 0.000562764354981482, 0.000683559337630868, 0.0006445597973652184, 0.0006083390908315778, 0.000593601493164897, 0.0006463286699727178, 0.0007829663809388876, 0.0007389364182017744, 0.0011422302341088653, 0.001994825666770339 ]

[ "“President Boyd K. Packer: Farewell to a Master Teacher,” Ensign, August 2015, 1–16\n\n“As one of the Twelve Apostles, I bear witness of the Lord Jesus Christ. ", "He lives. ", "He is our Redeemer and our Savior. … ", "He presides over this Church. ", "He is no stranger to His servants. ", "As we move into the future with quiet confidence, His Spirit will be with us.” ", "1\n\nPhotograph courtesy of Deseret News\n\nPresident Boyd K. Packer, who served as a member of the Quorum of the Twelve Apostles since 1970 and as President of the quorum since 2008, passed away on July 3, 2015. ", "He is survived by his wife, Donna Smith Packer, and their 10 children.", "\n\nMembers of The Church of Jesus Christ of Latter-day Saints lost more than a dedicated leader when President Boyd K. Packer, President of the Quorum of the Twelve Apostles, passed away. ", "They lost a master teacher who taught even the most complex gospel topics with clarity.", "\n\nPresident Packer dedicated his life to serving his Heavenly Father and to teaching Heavenly Father’s children, beginning with his own family, continuing with a career in seminary, and then in his worldwide ministry as a General Authority. ", "He looked to the Savior as his model for teaching. ", "In his 1975 book Teach Ye Diligently, he wrote:\n\n“Consider, please, that He has declared, ‘What manner of men ought ye to be? ", "Verily I say unto you, even as I am.’ (", "3 Nephi 27:27.) …", "\n\n“I do not hesitate to admit that I desire to teach as He taught. ", "Though that may be far beyond my capabilities, He is, nevertheless, the ideal. …", "\n\n“… It is not untoward for any of us to aspire to be like Him.", "”2\n\nAs a young man, President Packer pledged to God full and unwavering lifelong loyalty.3 He lived out that pledge and left a legacy that will be a spiritual landmark for other disciples of Jesus Christ. ", "While the world in general might not be aware of all of Boyd K. Packer’s contributions, his works are known to the Source of their inspiration.", "\n\nEarly Years Photographs courtesy of President Boyd K. Packer, except as noted Boyd Kenneth Packer was born on September 10, 1924, the 10th of 11 children born to Ira Wight Packer and Emma Jensen Packer. ", "Theirs was a family in which hard work and obedience to gospel principles were simply the way one lived. ", "Ira was a skilled mechanic who owned an auto-repair garage in Brigham City, Utah, about 60 miles (100 km) north of Salt Lake City. ", "Both of Ira’s parents came from farming backgrounds. ", "Emma grew up with a love of nature and the outdoors that would carry over into the life of her son Boyd. ", "His love of the outdoors, nature, and animals was expressed in much of his artwork, particularly his detailed, sensitive carvings of birds. ", "President Packer’s carving “Male Blue Jay on Norway Maple Leaves” is on display at a Brigham Young University museum. ", "His ancestors were pioneers, British on his father’s side and Scandinavian on his mother’s. ", "The brass door knocker on President Packer’s front door and on the doors of his grandchildren’s homes is modeled after a yoked ox carving he made. ", "For members of the Packer family, it symbolizes two important things: first, their pioneer ancestry and, second, willingness and faith to take on the yoke of the Master (see Matthew 11:29–30).4 “Sometimes in my growing years I thought we were poor,” he once wrote in a short life history. “", "I later learned that that was not true. ", "We just didn’t have any money. ", "We were always rich in the things most significant in our lives.", "”5 At age five, Boyd contracted polio. ", "His illness was diagnosed at the time as pneumonia, and he seemed to recover with no significant aftereffects. ", "But the polio would come to be a challenge later in life. ", "Boyd posing with his B-24 bomber crew Because of World War II, he was unable to serve a mission. ", "In 1943 he joined the United States Army Air Corps, graduating as a pilot in 1944. ", "He was trained to fly bombers and spent almost a year stationed in Japan after the war was over. ", "Because of pain he suffered while serving as a pilot, X-rays were taken, showing evidence of his earlier polio in the malformed bones of his knees and hip. (", "In the last years of his life, the effects of that polio left him using a wheelchair.) ", "While in the military, he found ample opportunity to study the gospel, reading the Book of Mormon several times. ", "He referred to that book as “the single most powerful influence in my life.", "”6 More than once he told the story of gaining his own witness of the truth of the gospel in an isolated bunker one night on a tiny island in the Pacific Ocean. ", "He walked away from that bunker a different man because he knew. ", "What had been a belief and a hope had crystalized into certainty. ", "He knew.7 Boyd baptized Chio Sato; her husband was also baptized that day. ", "These were the first baptisms in Japan since 1924. ", "He also found ways to be a missionary. ", "Boyd K. Packer was one of those Latter-day Saint servicemen who helped reintroduce the gospel in Japan in the mid-1940s. ", "He was instrumental in bringing Tatsui Sato into the Church, a man whose later work in translation helped many Japanese members enjoy the scriptures and temple ceremonies in their own language.", "\n\nMarriage and Family Returning home, Boyd enrolled in Weber College (now Weber State University) in Ogden, Utah. ", "There he met Donna Smith, whom he married on July 28, 1947, in the Logan Utah Temple. ", "They would become parents of 10 children: Allan, Kenneth, David, Laurel, Russell, Spencer, Gayle, Kathleen, Lawrence, and Eldon. ", "President Packer always credited his wife for the success of their home life. ", "He described her as a dynamic woman who had “a great and powerful motivating influence.", "”8 Her influence on him was described well by a longtime friend, Elder A. Theodore Tuttle (1919–86), of the First Quorum of the Seventy. ", "He and President Packer served together as teachers and administrators in the Church Educational System. ", "In a foreword to one of President Packer’s books, his friend Theodore wrote: “There is one source of inspiration available to Elder Packer that is not available to anyone else. ", "That is Donna. … ", "She is the one who has borne their ten children—and borne much of the responsibility of rearing them, due to his call as ‘special witness.’ ", "She it is who is his sweetheart, his friend, and his quiet support.", "”9 President and Sister Packer were wonderful examples of unity. ", "Their son Allan explains: “If you see Dad, you see Mom. ", "If you see Mom, you see Dad. ", "They’re together; they’re unified. ", "They influenced each other, especially, of course, in the home.” ", "His role as father was always a top priority, so he found time for his children. ", "If he was carving wood or painting and his children wanted to participate, he would let them. ", "His son Allan remembers: “He would stop and give us a brush with some paint, and we could work. ", "And then when we would lose interest, … he would go about repairing. ", "And so behind some of the layers of paint are some contributions that the kids made.", "”10 Boyd K. Packer (left) as a young teacher with his seminary class in Brigham City, Utah. ", "Photograph courtesy of Deseret News As a young father, Boyd served the Church in local callings and also his community as a city councilman in Brigham City. ", "At the same time, he was employed by the Church Educational System. ", "He had received an associate’s degree from Weber College (1948) and a bachelor’s degree from Utah State University (1949). ", "He received a master’s degree also from Utah State University (1953) and a doctorate from Brigham Young University (1962). ", "In 1955 he was appointed as an assistant administrator of seminaries and institutes for the Church. ", "He was serving in that role when he was called as an Assistant to the Twelve in October 1961, shortly after his 37th birthday. ", "While he was an Assistant to the Twelve, he also served as president of the New England Mission, where his integrity and ability helped him make many friends for the Church among civic and business leaders. ", "Despite the heavy demands of these callings, President Packer continued to put family first. ", "Said his son Allan: “He always … knew that the family was the eternal organization—and so that had priority. ", "If we called his office, his secretary had instructions that no matter when we called or what meeting he was in, we were always to be able to be put through. ", "And that’s what we experienced if we needed to even just say hi. ", "Sometimes we were a little embarrassed if we were interrupting a meeting with several of the senior Brethren, but that was the instruction—it was that he was always available.", "”11\n\nCalled to the Quorum of the Twelve Apostles In 1970 he was called to the Quorum of the Twelve Apostles following the death of President David O. McKay. ", "His service in the quorum began as a new President of the Church was sustained: President Joseph Fielding Smith. ", "Elder Packer’s later callings to preside over the quorum would also coincide with the sustaining of new Church Presidents. ", "In June 1994, when President Howard W. Hunter retained President Thomas S. Monson as Second Counselor in the First Presidency, Elder Packer, the next senior Apostle, became Acting President. ", "In February 2008, he was set apart as President of the quorum after President Monson became President of the Church. ", "A 1970 article reporting on his call to the Quorum of the Twelve quoted Elder Packer regarding how he had planned his life: “A number of years ago I chose several basic objectives in life—things that I wanted to be and do. ", "First, I wanted to be a good father. … ", "Livelihood, hobbies, even social opportunities had to be weighed against whether or not they related to that ideal. ", "I soon learned that the perfect plan for fatherhood was the gospel. … ", "Home is the center of the gospel—and of my life. … ", "The Packers with their first two children, Allan and Kenneth “The second goal that I had was that I … wanted to be good—good for something. ", "Mostly I wanted to be a good son, to both my earthly father and my Heavenly Father. ", "I have never thought that I deserved to have good children unless I could be one myself. ", "I’ve had an idea that we contribute to the glory of our Father in heaven when we add in our own person one more worthy individual. ", "I’ve felt that I wasn’t worthy to get what I wasn’t willing to give.” ", "Elder Packer spoke also of how we can receive help in pursuing objectives like these: “It seems to me that there is a great power in the Church—in all of us—that is untapped because we are always setting about to do things in our way, when the Lord’s way would accomplish much greater returns. … ", "Why don’t we talk to our Father? ", "In specifics? ", "About real problems? ", "As often as we would with our earthly father if he were nearby?”12 President Packer’s service in general Church leadership brought many years of schooling under the hand of the Lord. ", "His ministry spanned generations and continents. ", "One woman recalled how Elder Packer, speaking in the pioneer tabernacle in her southeastern Idaho town in the 1960s, touched her life when she was a teen. ", "He called on young people in the congregation to go home and express love for their parents and thank them for their sacrifices. ", "Expressions of love like this were not common in her home nor in the homes of her peers. ", "Elder Packer’s message caused her to consider what it was costing her parents, in labor and self-denial, to raise a large family on the income from a small farm. ", "It gave her new appreciation for her pioneer ancestry and spiritual heritage. ", "Elder Packer and his wife, Donna, hold a family home evening with their children in 1970, when he was called to be an Apostle. ", "Pictured from left are Eldon, Gayle, Lawrence, Kathleen, Spencer, Russell, Laurel, David, and Allan. ", "One son, Kenneth, was serving a mission in Australia. ", "Photograph courtesy of Deseret News In Germany in the early 1970s, Elder Packer and his wife had to travel by train from Munich to Berlin overnight to fulfill one of his Church assignments. ", "Two young missionaries had taken them to the train. ", "As it was pulling out of the station, one of those young elders asked through the window if Elder Packer had any German money. ", "When Elder Packer responded no, the missionary, running alongside the train, handed him a 20-mark note. ", "A few hours later, that 20-mark note proved crucial to finishing their journey safely. ", "It helped change the mind of an East German soldier, who could have put Sister Packer off the train and arrested her because her older type of passport was not accepted by his government. ", "The young missionary who had handed that 20-mark note to a visiting Apostle was David A. Bednar, who, some 30 years later, would serve with President Packer in the Quorum of the Twelve. ", "The point of the story, President Packer said, was that when we are in the service of the Lord, we need not worry about being able to fulfill our assignments because His hand will be with us.13 Anyone who followed the ministry of Boyd K. Packer knew that the hand of the Lord was with him. ", "The love he possessed for his fellow Latter-day Saints could be felt when he ministered among them. ", "On a narrow, hilly spot in the Mayan highlands, in the Momostenango Guatemala Stake, there is a small meetinghouse that was dedicated by Elder Packer in the mid-1980s. ", "The members who were present fondly recall how the strength of his witness touched their spirits, and they still remember the love they felt as an Apostle of the Lord walked among them. ", "Elder Packer visits with a group of children in Guatemala in 1979. ", "Photograph courtesy of Church History Library\n\nTeacher and Witness The Brethren who served with Boyd K. Packer in the Quorum of the Twelve came to know his ability to teach gospel principles and his steadfastness in following them. “", "Elder Packer is very much a teacher,” commented President James E. Faust (1920–2007), who served as a counselor in the First Presidency. “", "While all of the Twelve are teachers, he’s a teacher in the Twelve.” ", "Elder Packer’s love of and use of the scriptures in leadership roles influenced the direction of the entire Church, President Faust said.14 Known for teaching from the scriptures, then-Elder Packer meets with fellow Apostles Elder James E. Faust (left) and Elder Dallin H. Oaks in 1987. ", "Elder Russell M. Nelson once noted that when the Quorum of the Twelve was weighing a problem, Elder Packer would often search for relevant teachings from the Book of Mormon to help find a solution. “", "Without the Book of Mormon, Elder Packer couldn’t be the prophet he is. ", "He is a gifted seer.” ", "His teaching from the scriptures was characterized by “deep comprehension,” Elder Nelson said. “", "No one has plumbed the depths of this man.", "”15 Once Elder Packer was asked if it is difficult for 12 men with strong intellects and widely varying backgrounds to agree on the direction of the affairs of the Church. ", "Members of the Twelve, he explained, “share something that no other men on earth share.” ", "They bring great individual strengths and abilities to bear on their collective responsibilities—and then they unanimously bend their wills to divine inspiration received. “", "We’re all individual, and we’re all resolute, but we’re all one.", "”16 In his later years, the expression of his apostolic witness seemed to become somehow more urgent for him and ever more plain. ", "President Packer’s voice of testimony was a reliable constant in the Church, a word of witness that never faltered. ", "It was never hard for the hearer to discern that Boyd K. Packer knew. ", "After the sustaining of President Monson in the April 2008 general conference, President Packer’s report of what happened showed both reverence for Church callings and certainty as to their divine origin. “", "There was no question about what would be done, no hesitancy,” he said. “", "In that sacred meeting, Thomas Spencer Monson was sustained by the Quorum of the Twelve Apostles as the President of the Church.” ", "The process was not new, President Packer explained. “", "The Lord Himself set in motion this pattern of administration.", "”17 Accompanied by Elder Russell M. Nelson, President Packer returned to his hometown for the groundbreaking of the Brigham City Utah Temple. ", "Photograph courtesy of Deseret News In 2010, President Packer returned to his hometown of Brigham City to break ground for the Brigham City Utah Temple on the site of his old elementary school. ", "Two years later he presided over its dedication. ", "Throughout the process he showed his characteristic humility. “", "I didn’t propose there be a temple in Brigham City,” he said. “", "The Brethren brought that up. ", "My contribution was not objecting. ", "The same was true of the dedication; I didn’t assign myself to that. ", "I am glad I was assigned to it. ", "I am grateful.", "”18\n\nAlways Striving to Improve Throughout his life, President Packer was constantly striving to improve. ", "He spoke of the living Apostles, including himself, as “ordinary people” who might wonder why they should be called to this holy office. “", "There are many qualifications that I lack. ", "There is so much in my effort to serve that is wanting. ", "There is only one single thing, one qualification that can explain it. ", "Like Peter and all of those who have since been ordained, I have that witness. “", "I know that God is our Father. ", "He introduced His Son, Jesus Christ, to Joseph Smith. ", "I declare to you that I know that Jesus is the Christ. ", "I know that He lives. … ", "He made His Atonement. ", "Of Him I bear witness.", "”19 Known for teaching that reverence invites revelation, President Packer was one of the first General Authorities to arrive at general conference. ", "Photograph by Scott G. Winterton, Deseret News Many years ago, after receiving a personal witness on that tiny island in the Pacific, he wanted to give something back. ", "He wanted to formalize his commitment as a disciple of Jesus Christ. “", "It had become critically important,” he remembered, “to establish this intention between me and the Lord so that I knew that He knew which way I had committed my agency. ", "I went before Him and said, ‘I’m not neutral, and you can do with me what you want. ", "If you need my vote, it’s there. ", "I don’t care what you do with me and you don’t have to take anything from me because I give it to you—everything, all I own, all I am.", "’”20 Those who loved and honored Boyd K. Packer will have every confidence that the Lord has accepted his offering, an offering magnified through a lifetime of sacred service." ]

{ "pile_set_name": "OpenWebText2" }

[ 0.0007216149824671447, 0.015241291373968124, 0.04214100167155266, 0.002975199604406953, 0.002227540360763669, 0.0005550189525820315, 0.0005890532629564404, 0.002111508511006832, 0.0010616290383040905, 0.0009358256938867271, 0.0010860265465453267, 0.0008680677856318653, 0.0008857125067152083, 0.0009574973955750465, 0.002268161391839385, 0.0007639530813321471, 0.0006465680198743939, 0.001231098547577858, 0.0006576405139639974, 0.0005777576589025557, 0.0006635651807300746, 0.000677315576467663, 0.0007649671169929206, 0.0007509567076340318, 0.0008693852578289807, 0.000704424106515944, 0.0006537934532389045, 0.0010055599268525839, 0.002828258089721203, 0.0008404427208006382, 0.0006878891726955771, 0.0011261473409831524, 0.0005766183603554964, 0.03887878358364105, 0.0006507541984319687, 0.004207585006952286, 0.0009455253020860255, 0.000713521905709058, 0.0010664951987564564, 0.0027672534342855215, 0.0010733952512964606, 0.0005813625757582486, 0.0007689177291467786, 0.0006001206929795444, 0.0011803241213783622, 0.0006752602057531476, 0.0006117614102549851, 0.0006338858511298895, 0.0008528981124982238, 0.0007698521949350834, 0.0006064265035092831, 0.0005962467403151095, 0.000632061914075166, 0.0013104903046041727, 0.0006969834794290364, 0.0006834935629740357, 0.0006199482595548034, 0.0009569829562678933, 0.0005975380772724748, 0.009005461819469929, 0.0011845523258671165, 0.0009839283302426338, 0.0006452644593082368, 0.005362425930798054, 0.029653234407305717, 0.0012415864039212465, 0.000602782005444169, 0.0009995844447985291, 0.0007483363151550293, 0.0007452455465681851, 0.0007123295799829066, 0.0007572828326374292, 0.0006257726927287877, 0.0006159073091112077, 0.0006552555714733899, 0.0006188460974954069, 0.0006261571543291211, 0.0007757808780297637, 0.0006473026587627828, 0.000609547714702785, 0.0008276760927401483, 0.0006809766637161374, 0.0005672411061823368, 0.0005517216632142663, 0.0005603135796263814, 0.0006720494711771607, 0.0006908089853823185, 0.0008169712382368743, 0.0006557201268151402, 0.0006187594262883067, 0.0006719851517118514, 0.0008167315390892327, 0.0005958544206805527, 0.0014560561394318938, 0.0012046816991642118, 0.0005969449994154274, 0.0007460862398147583, 0.002960497047752142, 0.0006285608396865427, 0.001894779852591455, 0.0005638115108013153, 0.0023021630477160215, 0.0006622239015996456, 0.0007833947893232107, 0.0006758625386282802, 0.0005716809537261724, 0.0006427963962778449, 0.0006320914253592491, 0.0006714915507473052, 0.0009717117063701153, 0.000562479835934937, 0.0006761511322110891, 0.0006468219216912985, 0.0006213508313521743, 0.0005720706540159881, 0.0007042122888378799, 0.0009163302602246404, 0.0007105633267201483, 0.0005508113536052406, 0.001566202030517161, 0.0006398609257303178, 0.000627150118816644, 0.0006586647941730917, 0.0005856878706254065, 0.000511219201143831, 0.0006814616499468684, 0.0006362278363667428, 0.0006138336029835045, 0.0015274035977199674, 0.0006097560399211943, 0.000602323969360441, 0.0017912578769028187, 0.002396403579041362, 0.0006192827131599188, 0.0009263737592846155, 0.0005639719893224537, 0.001110654673539102, 0.0005327594699338078, 0.003927553538233042, 0.0006576423184014857, 0.0008651495445519686, 0.0008291118429042399, 0.0005562537699006498, 0.0007231394993141294, 0.0006372899515554309, 0.0006920072482898831, 0.0006533985724672675, 0.0005989755736663938, 0.0006010706420056522, 0.0005616120179183781, 0.000617574027273804, 0.0006529406527988613, 0.0005859073135070503, 0.0009139346075244248, 0.0006079000304453075, 0.0005675250431522727, 0.0005532106151804328, 0.0009241319494321942, 0.0005730887642130256, 0.0006962286424823105, 0.0007804228807799518, 0.000651493261102587, 0.0008583266753703356, 0.0032082239631563425, 0.0008087033638730645, 0.01225154846906662, 0.0025290437042713165, 0.014138715341687202, 0.0005724418442696333, 0.0006279363296926022, 0.0005711722769774497, 0.0007386365323327482, 0.0005811310256831348, 0.00128582410980016, 0.0006304264534264803, 0.33874160051345825, 0.0005662253242917359 ]

[ "/*\r\n * (C) 2003-2006 Gabest\r\n * (C) 2006-2013 see Authors.txt\r\n *\r\n * This file is part of MPC-HC.", "\r\n *\r\n * MPC-HC is free software; you can redistribute it and/or modify\r\n * it under the terms of the GNU General Public License as published by\r\n * the Free Software Foundation; either version 3 of the License, or\r\n * (at your option) any later version.", "\r\n *\r\n * MPC-HC is distributed in the hope that it will be useful,\r\n * but WITHOUT ANY WARRANTY; without even the implied warranty of\r\n * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. ", " See the\r\n * GNU General Public License for more details.", "\r\n *\r\n * You should have received a copy of the GNU General Public License\r\n * along with this program. ", " If not, see <http://www.gnu.org/licenses/>.", "\r\n *\r\n */\r\n\r\n#pragma once\r\n\r\n#include <afxstr.h>\r\n#include \"NullRenderers.h\"\r\n#include \"HdmvClipInfo.h\"\r\n#include \"H264Nalu.h\"\r\n#include \"MediaTypeEx.h\"\r\n#include \"vd.h\"\r\n#include \"text.h\"\r\n\r\n#define LCID_NOSUBTITLES -1\r\n\r\nextern void DumpStreamConfig(TCHAR* fn, IAMStreamConfig* pAMVSCCap);\r\nextern int CountPins(IBaseFilter* pBF, int& nIn, int& nOut, int& nInC, int& nOutC);\r\nextern bool IsSplitter(IBaseFilter* pBF, bool fCountConnectedOnly = false);\r\nextern bool IsMultiplexer(IBaseFilter* pBF, bool fCountConnectedOnly = false);\r\nextern bool IsStreamStart(IBaseFilter* pBF);\r\nextern bool IsStreamEnd(IBaseFilter* pBF);\r\nextern bool IsVideoRenderer(IBaseFilter* pBF);\r\nextern bool IsAudioWaveRenderer(IBaseFilter* pBF);\r\nextern IBaseFilter* GetUpStreamFilter(IBaseFilter* pBF, IPin* pInputPin = NULL);\r\nextern IPin* GetUpStreamPin(IBaseFilter* pBF, IPin* pInputPin = NULL);\r\nextern IPin* GetFirstPin(IBaseFilter* pBF, PIN_DIRECTION dir = PINDIR_INPUT);\r\nextern IPin* GetFirstDisconnectedPin(IBaseFilter* pBF, PIN_DIRECTION dir);\r\nextern void NukeDownstream(IBaseFilter* pBF, IFilterGraph* pFG);\r\nextern void NukeDownstream(IPin* pPin, IFilterGraph* pFG);\r\nextern IBaseFilter* FindFilter(LPCWSTR clsid, IFilterGraph* pFG);\r\nextern IBaseFilter* FindFilter(const CLSID& clsid, IFilterGraph* pFG);\r\nextern IPin* FindPin(IBaseFilter* pBF, PIN_DIRECTION direction, const AM_MEDIA_TYPE* pRequestedMT);\r\nextern CStringW GetFilterName(IBaseFilter* pBF);\r\nextern CStringW GetPinName(IPin* pPin);\r\nextern IFilterGraph* GetGraphFromFilter(IBaseFilter* pBF);\r\nextern IBaseFilter* GetFilterFromPin(IPin* pPin);\r\nextern IPin* AppendFilter(IPin* pPin, CString DisplayName, IGraphBuilder* pGB);\r\nextern IPin* InsertFilter(IPin* pPin, CString DisplayName, IGraphBuilder* pGB);\r\nextern void ExtractMediaTypes(IPin* pPin, CAtlArray<GUID>& types);\r\nextern void ExtractMediaTypes(IPin* pPin, CAtlList<CMediaType>& mts);\r\nextern void ShowPPage(CString DisplayName, HWND hParentWnd);\r\nextern void ShowPPage(IUnknown* pUnknown, HWND hParentWnd);\r\nextern CLSID GetCLSID(IBaseFilter* pBF);\r\nextern CLSID GetCLSID(IPin* pPin);\r\nextern bool IsCLSIDRegistered(LPCTSTR clsid);\r\nextern bool IsCLSIDRegistered(const CLSID& clsid);\r\nextern void CStringToBin(CString str, CAtlArray<BYTE>& data);\r\nextern CString BinToCString(const BYTE* ptr, size_t len);\r\ntypedef enum {\r\n CDROM_NotFound,\r\n CDROM_Audio,\r\n CDROM_VideoCD,\r\n CDROM_DVDVideo,\r\n CDROM_Unknown\r\n} cdrom_t;\r\nextern cdrom_t GetCDROMType(TCHAR drive, CAtlList<CString>& files);\r\nextern CString GetDriveLabel(TCHAR drive);\r\nextern bool GetKeyFrames(CString fn, CUIntArray& kfs);\r\nextern DVD_HMSF_TIMECODE RT2HMSF(REFERENCE_TIME rt, double fps = 0); // used to remember the current position\r\nextern DVD_HMSF_TIMECODE RT2HMS_r(REFERENCE_TIME rt); // used only for information (for display on the screen)\r\nextern REFERENCE_TIME HMSF2RT(DVD_HMSF_TIMECODE hmsf, double fps = 0);\r\nextern void memsetd(void* dst, unsigned int c, size_t nbytes);\r\nextern void memsetw(void* dst, unsigned short c, size_t nbytes);\r\nextern bool ExtractBIH(const AM_MEDIA_TYPE* pmt, BITMAPINFOHEADER* bih);\r\nextern bool ExtractBIH(IMediaSample* pMS, BITMAPINFOHEADER* bih);\r\nextern bool ExtractAvgTimePerFrame(const AM_MEDIA_TYPE* pmt, REFERENCE_TIME& rtAvgTimePerFrame);\r\nextern bool ExtractDim(const AM_MEDIA_TYPE* pmt, int& w, int& h, int& arx, int& ary);\r\nextern bool MakeMPEG2MediaType(CMediaType& mt, BYTE* seqhdr, DWORD len, int w, int h);\r\nextern bool CreateFilter(CStringW DisplayName, IBaseFilter** ppBF, CStringW& FriendlyName);\r\nextern IBaseFilter* AppendFilter(IPin* pPin, IMoniker* pMoniker, IGraphBuilder* pGB);\r\nextern CStringW GetFriendlyName(CStringW DisplayName);\r\nextern HRESULT LoadExternalObject(LPCTSTR path, REFCLSID clsid, REFIID iid, void** ppv);\r\nextern HRESULT LoadExternalFilter(LPCTSTR path, REFCLSID clsid, IBaseFilter** ppBF);\r\nextern HRESULT LoadExternalPropertyPage(IPersist* pP, REFCLSID clsid, IPropertyPage** ppPP);\r\nextern void UnloadExternalObjects();\r\nextern CString MakeFullPath(LPCTSTR path);\r\nextern CString GetMediaTypeName(const GUID& guid);\r\nextern GUID GUIDFromCString(CString str);\r\nextern HRESULT GUIDFromCString(CString str, GUID& guid);\r\nextern CString CStringFromGUID(const GUID& guid);\r\nextern CStringW UTF8To16(LPCSTR utf8);\r\nextern CStringA UTF16To8(LPCWSTR utf16);\r\nextern CStringW UTF8ToStringW(const char* S);\r\nextern CStringW LocalToStringW(const char* S);\r\nextern CString ISO6391ToLanguage(LPCSTR code);\r\nextern CString ISO6392ToLanguage(LPCSTR code);\r\nextern LCID ISO6391ToLcid(LPCSTR code);\r\nextern LCID ISO6392ToLcid(LPCSTR code);\r\nextern CString ISO6391To6392(LPCSTR code);\r\nextern CString ISO6392To6391(LPCSTR code);\r\nextern CString LanguageToISO6392(LPCTSTR lang);\r\nextern int MakeAACInitData(BYTE* pData, int profile, int freq, int channels);\r\nextern BOOL CFileGetStatus(LPCTSTR lpszFileName, CFileStatus& status);\r\nextern bool DeleteRegKey(LPCTSTR pszKey, LPCTSTR pszSubkey);\r\nextern bool SetRegKeyValue(LPCTSTR pszKey, LPCTSTR pszSubkey, LPCTSTR pszValueName, LPCTSTR pszValue);\r\nextern bool SetRegKeyValue(LPCTSTR pszKey, LPCTSTR pszSubkey, LPCTSTR pszValue);\r\nextern void RegisterSourceFilter(const CLSID& clsid, const GUID& subtype2, LPCTSTR chkbytes, LPCTSTR ext = NULL, ...);\r\nextern void RegisterSourceFilter(const CLSID& clsid, const GUID& subtype2, const CAtlList<CString>& chkbytes, LPCTSTR ext = NULL, ...);\r\nextern void UnRegisterSourceFilter(const GUID& subtype);\r\nextern LPCTSTR GetDXVAMode(const GUID* guidDecoder);\r\nextern void DumpBuffer(BYTE* pBuffer, int nSize);\r\nextern CString ReftimeToString(const REFERENCE_TIME& rtVal);\r\nextern CString ReftimeToString2(const REFERENCE_TIME& rtVal);\r\nextern CString DVDtimeToString(const DVD_HMSF_TIMECODE& rtVal, bool bAlwaysShowHours = false);\r\nextern REFERENCE_TIME StringToReftime(LPCTSTR strVal);\r\nextern COLORREF YCrCbToRGB_Rec601(BYTE Y, BYTE Cr, BYTE Cb);\r\nextern COLORREF YCrCbToRGB_Rec709(BYTE Y, BYTE Cr, BYTE Cb);\r\nextern DWORD YCrCbToRGB_Rec601(BYTE A, BYTE Y, BYTE Cr, BYTE Cb);\r\nextern DWORD YCrCbToRGB_Rec709(BYTE A, BYTE Y, BYTE Cr, BYTE Cb);\r\nextern void TraceFilterInfo(IBaseFilter* pBF);\r\nextern void TracePinInfo(IPin* pPin);\r\nextern void SetThreadName(DWORD dwThreadID, LPCSTR szThreadName);\r\nextern void HexDump(CString fName, BYTE* buf, int size);\r\nextern void CorrectComboListWidth(CComboBox& m_pComboBox);\r\n\r\nextern void getExtraData(const BYTE* format, const GUID* formattype, const size_t formatlen, BYTE* extra, unsigned int* extralen);\r\nextern void audioFormatTypeHandler(const BYTE* format, const GUID* formattype,\r\n DWORD* pnSamples, WORD* pnChannels,\r\n WORD* pnBitsPerSample, WORD* pnBlockAlign,\r\n DWORD* pnBytesPerSec);\r\n\r\ntypedef enum {\r\n PICT_NONE,\r\n PICT_TOP_FIELD,\r\n PICT_BOTTOM_FIELD,\r\n PICT_FRAME\r\n} FF_FIELD_TYPE;\r\n\r\nclass CPinInfo : public PIN_INFO\r\n{\r\npublic:\r\n CPinInfo() {\r\n pFilter = NULL;\r\n }\r\n ~CPinInfo() {\r\n if (pFilter) {\r\n pFilter->Release();\r\n }\r\n }\r\n};\r\n\r\nclass CFilterInfo : public FILTER_INFO\r\n{\r\npublic:\r\n CFilterInfo() {\r\n pGraph = NULL;\r\n }\r\n ~CFilterInfo() {\r\n if (pGraph) {\r\n pGraph->Release();\r\n }\r\n }\r\n};\r\n\r\n#define BeginEnumFilters(pFilterGraph, pEnumFilters, pBaseFilter) \\\r\n{ \\\r\n CComPtr<IEnumFilters> pEnumFilters; \\\r\n if (pFilterGraph && SUCCEEDED(pFilterGraph->EnumFilters(&pEnumFilters))) \\\r\n { \\\r\n for (CComPtr<IBaseFilter> pBaseFilter; S_OK == pEnumFilters->Next(1, &pBaseFilter, 0); pBaseFilter = NULL) \\\r\n {\r\n\r\n#define EndEnumFilters }}}\r\n\r\n#define BeginEnumCachedFilters(pGraphConfig, pEnumFilters, pBaseFilter) \\\r\n{ \\\r\n CComPtr<IEnumFilters> pEnumFilters; \\\r\n if (pGraphConfig && SUCCEEDED(pGraphConfig->EnumCacheFilter(&pEnumFilters))) \\\r\n { \\\r\n for (CComPtr<IBaseFilter> pBaseFilter; S_OK == pEnumFilters->Next(1, &pBaseFilter, 0); pBaseFilter = NULL) \\\r\n {\r\n\r\n#define EndEnumCachedFilters }}}\r\n\r\n#define BeginEnumPins(pBaseFilter, pEnumPins, pPin) \\\r\n{ \\\r\n CComPtr<IEnumPins> pEnumPins; \\\r\n if (pBaseFilter && SUCCEEDED(pBaseFilter->EnumPins(&pEnumPins))) \\\r\n { \\\r\n for (CComPtr<IPin> pPin; S_OK == pEnumPins->Next(1, &pPin, 0); pPin = NULL) \\\r\n {\r\n\r\n#define EndEnumPins }}}\r\n\r\n#define BeginEnumMediaTypes(pPin, pEnumMediaTypes, pMediaType) \\\r\n{ \\\r\n CComPtr<IEnumMediaTypes> pEnumMediaTypes; \\\r\n if (pPin && SUCCEEDED(pPin->EnumMediaTypes(&pEnumMediaTypes))) \\\r\n { \\\r\n AM_MEDIA_TYPE* pMediaType = NULL; \\\r\n for (; S_OK == pEnumMediaTypes->Next(1, &pMediaType, NULL); DeleteMediaType(pMediaType), pMediaType = NULL) \\\r\n {\r\n\r\n#define EndEnumMediaTypes(pMediaType) \\\r\n } \\\r\n if (pMediaType) \\\r\n DeleteMediaType(pMediaType); \\\r\n } \\\r\n}\r\n\r\n#define BeginEnumSysDev(clsid, pMoniker) \\\r\n{ \\\r\n CComPtr<ICreateDevEnum> pDevEnum4$##clsid; \\\r\n pDevEnum4$##clsid.", "CoCreateInstance(CLSID_SystemDeviceEnum); \\\r\n CComPtr<IEnumMoniker> pClassEnum4$##clsid; \\\r\n if (SUCCEEDED(pDevEnum4$##clsid->CreateClassEnumerator(clsid, &pClassEnum4$##clsid, 0)) \\\r\n && pClassEnum4$##clsid) \\\r\n { \\\r\n for (CComPtr<IMoniker> pMoniker; pClassEnum4$##clsid->Next(1, &pMoniker, 0) == S_OK; pMoniker = NULL) \\\r\n {\r\n\r\n#define EndEnumSysDev }}}\r\n\r\n#define PauseGraph \\\r\n CComQIPtr<IMediaControl> _pMC(m_pGraph); \\\r\n OAFilterState _fs = -1; \\\r\n if (_pMC) \\\r\n _pMC->GetState(1000, &_fs); \\\r\n if (_fs == State_Running) \\\r\n _pMC->Pause(); \\\r\n \\\r\n HRESULT _hr = E_FAIL; \\\r\n CComQIPtr<IMediaSeeking> _pMS((IUnknown*)(INonDelegatingUnknown*)m_pGraph); \\\r\n REFERENCE_TIME _rtNow = 0; \\\r\n if (_pMS) \\\r\n _hr = _pMS->GetCurrentPosition(&_rtNow);\r\n\r\n#define ResumeGraph \\\r\n if (SUCCEEDED(_hr) && _pMS && _fs !", "= State_Stopped) \\\r\n _hr = _pMS->SetPositions(&_rtNow, AM_SEEKING_AbsolutePositioning, NULL, AM_SEEKING_NoPositioning); \\\r\n \\\r\n if (_fs == State_Running && _pMS) \\\r\n _pMC->Run();\r\n\r\n#define CallQueue(call) \\\r\n if (!", "m_pOutputQueue) \\\r\n return NOERROR; \\\r\n m_pOutputQueue->##call; \\\r\n return NOERROR;\r\n\r\n#define QI(i) (riid == __uuidof(i)) ? ", "GetInterface((i*)this, ppv) :\r\n#define QI2(i) (riid == IID_##i) ? ", "GetInterface((i*)this, ppv) :\r\n\r\n#define SAFE_DELETE(p) { if (p) { delete (p); (p) = NULL; } }\r\n#define SAFE_DELETE_ARRAY(p) { if (p) { delete [] (p); (p) = NULL; } }\r\n#define SAFE_RELEASE(p) { if (p) { (p)->Release(); (p) = NULL; } }\r\n\r\n#define ResStr(id) CString(MAKEINTRESOURCE(id))\r\n\r\ntemplate <typename T> __inline void INITDDSTRUCT(T& dd)\r\n{\r\n ZeroMemory(&dd, sizeof(dd));\r\n dd.dwSize = sizeof(dd);\r\n}\r\n\r\n#define countof(array) (sizeof(array)/sizeof(array[0]))\r\n\r\ntemplate <class T>\r\nstatic CUnknown* WINAPI CreateInstance(LPUNKNOWN lpunk, HRESULT* phr)\r\n{\r\n *phr = S_OK;\r\n CUnknown* punk = DEBUG_NEW T(lpunk, phr);\r\n if (punk == NULL) {\r\n *phr = E_OUTOFMEMORY;\r\n }\r\n return punk;\r\n}\r\n\r\ninline int GCD(int a, int b)\r\n{\r\n if (a == 0 || b == 0) {\r\n return 1;\r\n }\r\n while (a !", "= b) {\r\n if (a < b) {\r\n b -= a;\r\n } else if (a > b) {\r\n a -= b;\r\n }\r\n }\r\n return a;\r\n}\r\n" ]

{ "pile_set_name": "Github" }

[ 0.0006593072903342545, 0.0006754709756933153, 0.0006009606295265257, 0.0005785456742160022, 0.0006360611296258867, 0.0006093313568271697, 0.03265753760933876, 0.015155462548136711, 0.0036808601580560207, 0.00427913898602128, 0.0007785885245539248, 0.04037880524992943, 0.004206845536828041 ]

[ "Conventionally, an electronic scanning type radar has been used as a radar mounted on a vehicle, which utilizes the system of an FMCW (Frequency Modulated Continuous Wave) radar, a multifrequency CW (Continuous Wave) radar, a pulse radar, and the like.", "\nAccording to each of the radars mentioned above, an estimation method of a direction of an incoming wave of an array antenna is used as a technology for detecting a direction of an incoming wave (or a reception wave) from a target.", "\nAn example of this estimation method of a direction of an incoming wave is a null operation method (refer to Non-Patent Document 1 and 2, for instance) known as a super-resolution (high precision) algorithm such as a beam scanning method like the Beamformer Method and the Capon Method, a linear prediction method like the Maximum Entropy Method (MEM: Maximum Entropy Method), the Minimum Norm Method, MUSIC (Multiple Signal Classification) Method, and the ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) Method.", "\nFurther, the estimation of the direction of the incoming wave, used by a radar mounted on a vehicle, is conducted by a method (refer to, for example, Patent Document 1) using solely the digital beam forming (DBF: Digital Beam Forming) of the Beamformer Method, or by a method (refer to, for example, Patent Documents 2 and 3) combining the DBF and the Maximum Entropy Method in recent years in order to enhance the detection accuracy of the arrival direction of the reception wave (or the resolution performance of the target).", "\nMoreover, in order to apply a super-resolution algorithm such as MUSIC to a vehicle-mounted radar, a microprocessor is used which is utilized for vehicle-mounted radar with a low arithmetic processing capacity, compared to normal personal computers. ", "Therefore, a method is developed by simplifying an estimation algorithm (refer to, for example, Patent Documents 4, 5, 6, and 7).", "\nAccording to the super-resolution algorithm such as MUSIC, mentioned above, a correlation matrix is generated from data of reception waves received by each array antenna. ", "An eigen value calculation is conducted by this correlation matrix. ", "Thus, the arrival direction of the reception wave is detected. ", "Here, the eigen value calculation (computation) refers to a calculation obtaining the eigen value and the eigenvector.", "\nAt this time, the super-resolution algorithm generates a correlation matrix by an ensemble average of the reception data because the accuracy of detecting the arrival direction enhances as the noise element of the correlation matrix is removed.", "\nFor example, the FMCW radar uses a correlation matrix generated by obtaining highest possible number of samples of a data set of the received beat signal (a temporal sequence data of a certain period of time, which can be converted to a data in a frequency domain) and obtaining an average. ", "The number of samples is called a snapshot number (refer to, for example, Non-Patent Documents 1 and 2).", "\nHowever, according to the vehicle-mounted radar, the distance from the target and the relative velocity change constantly. ", "Therefore, even if the snapshot number is increased nebulously, the accuracy of detecting the arrival direction of the reception wave does not necessarily increase.", "\nOn the contrary, in order to increase the snapshot number within a control cycle (or a detection cycle such as 100 ms) for detecting the target, the frequency resolution operation corresponding to the reception signal must be conducted simultaneously for a number of times multiplying the number of array antennas and the snapshot number. ", "Thus, there is a limit on the number.", "\nPatent Document 4 describes a method in order to enhance the accuracy of detecting the target without increasing the snapshot number. ", "According to this method described in Patent Document 4, for example, a correlation matrix is stored for each beat frequency at the previous control cycle (or the control cycle immediately prior to the previous control cycle). ", "Further, this method obtains a weighted sum (weighted average) of a correlation matrix of a beat frequency, for which a target exists in the present control cycle, and a correlation matrix in the previous control cycle (or the control cycle immediately prior to the previous control cycle) having the same beat frequency. ", "Moreover, Patent Document 4 describes a method which stores the weighted averaged correlation matrix for each beat frequency, and obtains a further weighted sum of the correlation matrix of a beat frequency for which a target exists in the present control cycle and a correlation matrix with the same frequency obtained as described above by weighted summation.", "\nIn addition, Patent Document 5, mentioned above, discloses a method for increasing the snapshot number by averaging each correlation matrix generated for a frequency indicating a peak value, from among the same peak waveform for which a target exists at a beat frequency, and for a frequency in the vicinity (for example, ±two frequency resolutions).", "\nAccording to this Patent Document 5, an average of the correlation matrix in the frequency domain is further averaged by using a past correlation matrix.", "\nNext, Patent Document 6 describes estimating an arrival direction of the reception wave by combining a method estimating the form of a roadside and a method averaging the present and past correlation matrices.", "\nIn addition, Patent Document 7 describes a method which determines a weighted coefficient (or a forgetting coefficient: a constant representing the degree of forgetting) in real time in order to average the present and past correlation matrices.", "\nFurther, an electronic scanning type radar has been used conventionally as a radar mounted on a vehicle, which utilizes the system of an FMCW (Frequency Modulated Continuous Wave) radar, a multifrequency CW (Continuous Wave) radar, a pulse radar, and the like.", "\nAccording to each of the radars mentioned above, an estimation method of a direction of an incoming wave of an array antenna is used as a technology for detecting a direction of an incoming wave (or a reception wave) which is a reflection wave from a target corresponding to a transmission wave.", "\nAn example of this estimation method of a direction of an incoming wave is a null operation method (refer to Non-Patent Document 1 and 2, for instance) known as a super-resolution (high precision) algorithm such as a beam scanning method like the Beamformer Method and the Capon Method, a linear prediction method like the Maximum Entropy Method (MEM: Maximum Entropy Method), the Minimum Norm Method, MUSIC (Multiple Signal Classification) Method, and the ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) Method.", "\nFurther, the estimation of the direction of the incoming wave, used by a radar mounted on a vehicle, is conducted by a method (refer to, for example, Patent Document 1) using solely the digital beam forming (DBF: Digital Beam Forming) of the Beamformer Method, or by a method (refer to, for example, Patent Documents 2 and 3) combining the DBF and the Maximum Entropy Method in recent years in order to enhance the detection accuracy of the arrival direction of the reception wave (or the resolution performance of the target).", "\nMoreover, in order to apply a super-resolution algorithm such as MUSIC to a vehicle-mounted radar, methods have been developed with an inclination to simplify the processing (refer to, for example, Patent Documents 4 and 8). ", "These methods can be applied to a vehicle-mounted device which has a low processing capacity compared to normal personal computers.", "\nAccording to the super-resolution algorithm such as MUSIC, mentioned above, it is preferred that the direction of the incoming wave is estimated after a number of the incoming waves is estimated, in order to enhance the accuracy with which the direction is estimated.", "\nNon-Patent Documents 1 and 2 introduce the AIC (Akaike Information Criteria) and the MDL (Minimum Description Length) as methods to estimate the number of incoming waves based on the maximum-likelihood approach in statistical processing.", "\nHowever, according to the estimation methods introduced in Non-Patent Documents 1 and 2, it is necessary to collect a large amount of data and then conduct a dispersion assessment. ", "Therefore, it is not suitable to use these methods for a vehicle-mounted radar whose relative distance to the target and relative velocity fluctuate rapidly.", "\nPatent Document 8 describes a method for estimating the number of incoming waves, necessary for computing the MUSIC spectrum, with a light computation load. ", "In other words, Patent Document 8 describes a method which is an application of a threshold approach in which an eigen value is computed, and the signal space and a noise space are estimated individually based on the magnitude of the eigen value.", "\nIn this case, the reception intensity of the radar declines as the measured distance becomes larger. ", "Therefore, the method described above estimates the number of incoming waves by storing and setting a threshold for each relative distance to the target, and then comparing this threshold with the eigen value (equivalent to the reception intensity).", "\nIn addition, there is a method (refer to Patent Documents 9, for example) which normalizes an eigen value as one of the diagonal element value of the original covariance matrix (in other words, the correlation matrix), and then distinguishes based on a threshold value. ", "This method is not designed to be used for a vehicle-mounted device.", "\nFurthermore, an estimation method which estimates the number of incoming waves with a high degree of accuracy constantly performs a spectrum computation assuming that the number of incoming waves is the maximum number that can be received, and in the subsequent calculation of electronic power, unnecessary peak values are removed, and obtain a final estimation value of the number of incoming waves (Patent Document 10).", "\nMoreover, an electronic scanning type radar has been conventionally used as a radar mounted on a vehicle, which utilizes the system of an FMCW (Frequency Modulated Continuous Wave) radar, a multifrequency CW (Continuous Wave) radar, a pulse radar, and the like.", "\nAccording to each of the radars mentioned above, an estimation method of a direction of an incoming wave of an array antenna is used as a technology for detecting a direction of an incoming wave (or a reception wave) which is a reflection wave from a target corresponding to a transmission wave.", "\nAn example of this estimation method of a direction of an incoming wave is a null operation method (refer to Non-Patent Document 1 and 2, for instance) known as a super-resolution (high precision) algorithm such as a beam scanning method like the Beamformer Method and the Capon Method, a linear prediction method like the Maximum Entropy Method (MEM: Maximum Entropy Method), the Minimum Norm Method, MUSIC (Multiple Signal Classification) Method, and the ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) Method.", "\nFurther, the estimation of the direction of the incoming wave, used by a radar mounted on a vehicle, is conducted by a method (refer to, for example, Patent Document 1) using solely the digital beam forming (DBF: Digital Beam Forming) of the Beamformer Method, or by a method (refer to, for example, Patent Documents 2 and 3) combining the DBF and the Maximum Entropy Method in recent years in order to enhance the detection accuracy of the arrival direction of the reception wave (or the resolution performance of the target).", "\nMoreover, in order to apply a super-resolution algorithm such as MUSIC to a vehicle-mounted radar, methods have been developed with an inclination to simplify the processing (refer to, for example, Patent Documents 4 and 8). ", "These methods can be applied to a vehicle-mounted device which has a low processing capacity compared to normal personal computers.", "\nAccording to the super-resolution algorithm such as MUSIC, mentioned above, it is preferred that the direction of the incoming wave is estimated after a number of the incoming waves is estimated, in order to enhance the accuracy with which the direction is estimated.", "\nNon-Patent Documents 1 and 2 introduce the AIC (Akaike Information Criteria) and the MDL (Minimum Description Length) as methods to estimate the number of incoming waves based on the maximum-likelihood approach in statistical processing.", "\nHowever, according to the estimation methods introduced in Non-Patent Documents 1 and 2, it is necessary to collect a large amount of data and then conduct a dispersion assessment. ", "Therefore, it is not suitable to use these methods for a vehicle-mounted radar whose relative distance to the target and relative velocity fluctuate rapidly.", "\nPatent Document 8 describes a method for estimating the number of incoming waves, necessary for computing the MUSIC spectrum, with a light computation load. ", "In other words, Patent Document 8 describes a method which is an application of a threshold approach in which an eigen value is computed, and the signal space and a noise space are estimated individually based on the magnitude of the eigen value.", "\nIn this case, the reception intensity of the radar declines as the measured distance becomes larger. ", "Therefore, the method described above estimates the number of incoming waves by storing and setting a threshold for each relative distance to the target, and then comparing this threshold with the eigen value (which is proportional to the reception intensity).", "\nIn addition, there is a method (refer to Patent Documents 9, for example) which normalizes an eigen value as one of the diagonal element value of the original covariance matrix (in other words, the correlation matrix), and then distinguishes based on a threshold value. ", "This method is not designed to be used for a vehicle-mounted device. ", " [Non-Patent Document 1] “Adaptive Signal Processing Using Array Antennas,” Kikuma Nobuyoshi (Kagaku Gijutsu Shyuppan, 1998). ", " [Non-Patent Document 2] “Adaptive Antenna Technology,” Kikuma Nobuyoshi (Ohm Sha, 2003). ", " [Patent Document 1] Japanese Unexamined Patent Application, First Publication No. ", "2000-284044 [Patent Document 2] Japanese Unexamined Patent Application, First Publication No. ", "2006-275840 [Patent Document 3] Japanese Unexamined Patent Application, First Publication No. ", "2006-308542 [Patent Document 4] Japanese Unexamined Patent Application, First Publication No. ", "2007-040806 [Patent Document 5] Japanese Unexamined Patent Application, First Publication No. ", "2006-145251 [Patent Document 6] Japanese Unexamined Patent Application, First Publication No. ", "2006-242695 [Patent Document 7] Japanese Unexamined Patent Application, First Publication No. ", "2006-284182 [Patent Document 8] Japanese Unexamined Patent Application, First Publication No. ", "2006-047282 [Patent Document 9] Japanese Unexamined Patent Application, First Publication No. ", "2006-153579 [Patent Document 10] Japanese Unexamined Patent Application, First Publication No. ", "2000-121716" ]

{ "pile_set_name": "USPTO Backgrounds" }

[ 0.0006613566656596959, 0.0005532340728677809, 0.0006083153421059251, 0.0005509995971806347, 0.0006797508685849607, 0.0005622247117571533, 0.0005825206753797829, 0.0006386969471350312, 0.0006002688896842301, 0.0006526470533572137, 0.0005988448392599821, 0.0005745668313466012, 0.0005598672432824969, 0.0005577609990723431, 0.0006095857825130224, 0.0006205432582646608, 0.0007122925599105656, 0.0005849500885233283, 0.0005660229362547398, 0.0006338327657431364, 0.0005817998899146914, 0.0005746454116888344, 0.0005678139277733862, 0.0005656667053699493, 0.0005720147746615112, 0.0006503654294647276, 0.0005585706676356494, 0.0006083153421059251, 0.0005509995971806347, 0.0005533519433811307, 0.0006449770298786461, 0.000548154697753489, 0.0005465928697958589, 0.0005329044652171433, 0.0006220301147550344, 0.0005799294449388981, 0.0005808754358440638, 0.0006304336129687726, 0.0006079513113945723, 0.0005699464818462729, 0.0006359562976285815, 0.0005811431910842657, 0.0006512083928100765, 0.0005585706676356494, 0.0006083153421059251, 0.0005509995971806347, 0.0005533519433811307, 0.0006449770298786461, 0.000548154697753489, 0.0005465928697958589, 0.0005329044652171433, 0.0006220301147550344, 0.0005799294449388981, 0.0005808754358440638, 0.0006304336129687726, 0.0006099743186496198, 0.0005699464818462729, 0.0006359562976285815, 0.0006865265895612538, 0.0005966224707663059, 0.0005997035768814385, 0.000626336841378361, 0.0006258066277951002, 0.0005988022894598544, 0.0006460043950937688, 0.0006290187011472881, 0.0006408282788470387, 0.0006158063188195229, 0.000647393346298486, 0.000636583601590246, 0.000728566781617701 ]

[ "3D Systems Corporation and Mr. Kevin P. McAlea are parties to a severance arrangement pursuant to\nwhich Mr. McAlea would become entitled to severance payments equal to nine months of his then\ncurrent salary if his employment is terminated other than for cause." ]

{ "pile_set_name": "Pile-CC" }

[ 0.0006605245871469378 ]

[ "WHY LOOK ANYWHERE ELSE\n\nAre you looking for that special item you cant find anywhere else Well check my store. ", "I will be posting new things every day. ", "I also run sp...read more\n\nWHY LOOK ANYWHERE ELSE\n\nAre you looking for that special item you cant find anywhere else Well check my store. ", "I will be posting new things every day. ", "I also run specials all the time. ", "And yes I do combine shipping. ", "MY ITEMS COME FROM MY HOME,FAMILY,ESTATE SALES,GARAGE SALES,AUCTIONS AND SOME STUFF I SELL FOR FRIENDS." ]

{ "pile_set_name": "Pile-CC" }

[ 0.0014208806678652763, 0.0006585316732525826, 0.0014290654798969626, 0.0006585316732525826, 0.0006852637743577361, 0.0005769702256657183, 0.0008354945457540452 ]

[ "Proto-human\n\nProto-human may refer to:\n\n Archaic Homo sapiens, a loosely defined classification that includes a number of varieties of Homo\n Australopithecina, a collective classification of extinct, close relatives of humans, some of which are probable ancestors of humans and early hominins\n Proto-Human language, a designation of the hypothetical most recent common ancestor of all the world's languages\n\nSee also\n Human evolution" ]

{ "pile_set_name": "Wikipedia (en)" }

[ 0.0006742120604030788 ]

[ "The U.S. military said Monday it had killed the second in command of a Somalia-based ISIS affiliate in an airstrike over the weekend.", "\n\nIn a statement, U.S. Africa Command said Abdulhakim Dhuqub died in the Bari region of Somalia, an area which includes the tip of the Horn of Africa peninsula.", "\n\nThe statement referred to Dhuqub as ISIS-Somalia’s “second in command” and responsible for the “daily operations of the extremist group, attack planning, and resource procurement.\"", "\n\nISIS MEMBER TASKED WITH 'BRINGING IN MIGRANTS,' NEWLY OBTAINED TERROR GROUP DOCUMENTS SHOW\n\nAfrica Command said the strike killed only Dhuqub, and also destroyed a vehicle.", "\n\n“We continue to work with our Somali partners to keep pressure on the al-Shabaab and ISIS Somalia terror networks,\" said Marine Corps Maj. ", "Gen. Gregg Olson, U.S. Africa Command’s director of operations. \"", "When it supports the strategy, we use precision airstrikes to target those who plan and carry out the violent extremist activities that put Somalis at risk.\"", "\n\nNot much is known about Dhuqub; his name does not appear on the State Dept.", "'s Reward for Justice website normally reserved for top terrorists and there is not much information publicly available online. ", "ISIS-Somalia is a relatively small terrorist group consisting of roughly 125 fighters, according to officials.", "\n\nCLICK HERE TO GET THE FOX NEWS APP\n\nThe strike took place days after Somalia's prime minister Hassan Khayre visited the White House and the Pentagon.", "\n\nThe U.S. military has launched 31 airstrikes in Somalia this year, following a record-setting 47 last year. ", "In 2017, President Trump authorized the military to conduct offensive strikes in Somalia against Al Qaeda-linked Shabab fighters as well as the ISIS affiliate." ]

{ "pile_set_name": "OpenWebText2" }

[ 0.0012623491929844022, 0.0006765878642909229, 0.000722177792340517, 0.00070505979238078, 0.0008972410578280687, 0.0006273796898312867, 0.001033305423334241, 0.0006926865316927433, 0.014721461571753025, 0.0008152412483468652, 0.0006392599316313863, 0.0006294975755736232, 0.0008463067933917046 ]