Manihot esculenta, commonly called cassava (/kəˈsɑːvə/), manioc, or yuca (among numerous regional names), is a woody shrub of the spurge family, Euphorbiaceae, native to South America, from Brazil and parts of the Andes. Although a perennial plant, cassava is extensively cultivated as ...

Cassava - Wikipedia

ADAS-Cog Scores Were Stable in ... Simufilam Safe, Well-Tolerated. AUSTIN, Texas, Feb. 07, 2024 (GLOBE NEWSWIRE) -- Cassava Sciences, Inc. (Nasdaq: SAVA), a biotechnology company, today reported top-line results of a two-year clinical safety study...

Cassava Sciences, Inc. (SAVA) Stock Price, News, Quote & History ...

Cassava, or yuca, is a starchy root popular in Latin America. Its taste is earthy, slightly sweet, and nutty. Cassava must be cooked before eating.

Learn About Cassava (Yuca) and How to Prepare It

The Investor Relations website contains information about Cassava Sciences, Inc.&#x27;s business for stockholders, potential investors, and financial analysts.

Cassava Sciences | Cassava Sciences, Inc.

Company Presentations | Cassava Sciences, Inc.

Cassava disclosed Monday in a securities filing that it is cooperating with government investigations, without naming any agency. Cassava said an investigation isn’t a sign that wrongdoing occurred. Copyright ©2023 Dow Jones &amp; Company, Inc. All Rights Reserved. 87990cbe856818d5eddac44c7b1cdeb8

SEC Investigating Cassava Sciences, Developer of Experimental ...

Shareholders of Record Received ... Per Warrant AUSTIN, Texas, Jan. 05, 2024 (GLOBE NEWSWIRE) -- Cassava Sciences, Inc. (Nasdaq: SAVA) (&quot;Cassava Sciences&quot; or the &quot;Company&quot;) today announced the completion of a previously announced distribution of warrants (the “Warrants”) ...

Cassava Sciences, Inc. (SAVA) Stock Price, News, Quote ...

Shares of Cassava Sciences Inc. more than doubled toward a near 10-year high in very active trading Tuesday, after the company focused on neurodegenerative...

Cassava Sciences stock nearly triples toward a 10-year high after ...

Pain Therapeutics is burying its name and the world of hurt around its four-time reject opioid, Remoxy, in search of better fortunes in another — but certainly no easier — disease area. Rebranding itself as Cassava Sciences (complete with a new ticker $SAVA), the company formerly known ...

Pain free? Meet Cassava, the newest Alzheimer’s player on Nasdaq

Dubbed the ‘Tencent of Africa’, we delve inside a part of business maverick Strive Masiyiwa’s empire.

Cassava Smartech – the Most Valuable Tech Company on the Continent?

Cassava

cassava

Dick Sands the Boy Captain by Jules Verne, is part of the HackerNoon Books Series. You can jump to any chapter in this book . AN ANXIOUS VOYAGE here AN ANXIOUS VOYAGE. Thus the canoe drifted on for a week, the forests that for many miles had skirted the river ultimately giving place to extensive jungles that stretched far away to the horizon. Destitute, fortunately for the travellers, of human inhabitants, the district abounded in a large variety of animal life; zebras, elands, caamas, sported on the bank, disappearing at night-fall before howling leopards and roaring lions. It was Dick's general custom, as he lay to for a while in the afternoon, to go ashore in search of food, and as the manioc, maize, and sorghum that were to be found were of a wild growth and consequently not fit for consumption, he was obliged to run the risk of using his gun. On the 4th of July he succeeded by a single shot in killing , a kind of antelope about five feet long, with annulated horns, a tawny skin dappled with bright spots, and a white belly. The venison proved excellent, and was roasted over a fire procured by the primitive method, practised, it is said, even by gorillas, of rubbing two sticks together. pokoo In spite of these halts, and the time taken for the night's rest, the distance accomplished by the 8th could not be estimated at less than a hundred miles. The river, augmented by only a few insignificant tributaries, had not materially increased in volume; its direction, however, had slightly changed more to the north-west. It afforded a very fair supply of fish, which were caught by lines made of the long stems of creepers furnished with thorns instead [Illustration: He stood face to face with his foe.] of fish-hooks, a considerable proportion being the delicate , which when dried may be transported to any climate; besides these there were the black , the wide-headed , and occasionally the little , resembling Thames whitebait. sandjtkas usakas monndÃ©s dagalas Next day, Dick met with an adventure that put all his courage and composure to the test. He had noticed the horns of a caama projecting above the brushwood, and went ashore alone with the intention of securing it. He succeeded in getting tolerably close to it and fired, but he was terribly startled when a formidable creature bounded along some thirty paces ahead, and took possession of the prey he had just wounded. It was a majestic lion, at least five feet in height, of the kind called , in distinction to the maneless species known as the . Before Dick had time to reload, the huge brute had caught sight of him, and without relaxing its hold upon the writhing antelope beneath its claws, glared upon him fiercely. Dick's presence of mind did not forsake him; flight he knew was not to be thought of; his only chance he felt intuitively would be by keeping perfectly still; and aware that the beast would be unlikely to give up a struggling prey for another that was motionless, he stood face to face with his foe, not venturing to move an eyelid. In a few minutes the lion's patience seemed to be exhausted; with a grand stateliness, it picked up the caama as easily as a dog would lift a hare, turned round, and lashing the bushes with its tail, disappeared in the jungle. kÃ¡ramoo Nyassi-lion It took Dick some little time to recover himself sufficiently to return to the canoe. On arriving, he said nothing of the peril to which he had been exposed, but heartily congratulated himself that they had means of transport without making their way through jungles and forests. As they advanced, they repeatedly came across evidences that the country had not been always, as now it was, utterly devoid oi population; more than once, they observed traces which betokened the former existence of villages; either some ruined palisades or the of some thatched huts, or some solitary sacred tree within an enclosure would indicate that the death of a chief had, according to custom, made a native tribe migrate to new quarters. dÃ©bris If natives were still dwelling in the district, as was just probable, they must have been living underground, only emerging at night like beasts of prey, from which they were only a grade removed. Dick Sands had every reason to feel convinced that cannibalism had been practised in the neighbourhood, Three times, as he was wandering in the forest, he had come upon piles of ashes and half-charred human bones, the remnants, no doubt, of a ghastly meal, and although he mentioned nothing of what he had seen to Mrs. Weldon, he made up his mind to go ashore as seldom as possible, and as often as he found it absolutely necessary to go, he gave Hercules strict directions to push off into mid-stream at the very first intimation of danger. A new cause of anxiety arose on the following evening, and made it necessary for them to take the most guarded measures of precaution. The river-bed had widened out into a kind of lagoon, and on the right side of this, built upon piles in the water, not only was there a collection of about thirty huts, but the fires gleaming under the thatch, made it evident that they were all inhabited. Unfortunately the only channel of the stream flowed close under the huts, the river elsewhere being so obstructed with rocks that navigation of any kind was impossible. Nothing was more probable than that the natives would have set their nets all across the piles, and if so, the canoe would be sure to be obstructed, and an alarm must inevitably be raised. Every caution seemed to be unavailing, because the canoe must follow the stream; however, in the lowest of whispers Dick ordered Hercules to keep clear as much as he could of the worm-eaten timber. The night was not very dark, which was equally an advantage and a disadvantage, as while it permitted those on board to steer as they wanted, it did not prevent them from being seen. The situation became more and more critical. About a hundred feet ahead, the channel was very contracted; two natives, gesticulating violently, were seen squatting on the pilework; a few moments more and their voices could be heard; it was obvious that they had seen the floating mass; apprehending that it was going to destroy their nets, they yelled aloud and shouted for assistance; instantly five or six negroes scrambled down the piles, and perched themselves upon the cross-beams. On board the canoe the profoundest silence was maintained. Dick only signalled his directions to Hercules, without uttering a word, while Jack performed his part by holding Dingo's mouth tightly closed, to stop the low growlings which the faithful watch-dog seemed resolved to make; but fortunately every sound was overpowered by the rushing of the stream and the clamour of the negroes, as they hurriedly drew in their nets. If they should raise them in time, all might be well, but if, on the other hand, the canoe should get entangled, the consequences could hardly fail to be disastrous. The current in its narrow channel was so strong that Dick was powerless either to modify his course or to slacken it. Half a minute more, and the canoe was right under the woodwork, but the efforts of the natives had already elevated the nets so that the anticipated danger was happily escaped; but it chanced that in making its way through the obstacle, a large piece of the grass-thatch got detached. One of the negroes raised a sudden shout of alarm, and it seemed only too probable that he had caught a sight of the travellers below and was informing his companions. This apprehension, too, was only momentary; the current had changed almost to a rapid, and carried the canoe along with such velocity that the lacustrine village was quickly out of sight. "Steer to the left!" cried Dick, finding that the riverbed had again become clear. A stiff pull at the tiller made the craft fly in that direction. Dick went to the stern, and scanned the moonlit waters. All was perfectly still, no canoe was in pursuit; perhaps the natives had not one to use; but certain it was that [Illustration: Instantly five or six negroes scrambled down the piles.] when daylight dawned no vestige of an inhabitant was to be seen. Nevertheless Dick thought it prudent for a while to steer close under the shelter of the left-hand shore. By the end of the next four days the aspect of the country had undergone a remarkable change, the jungle having given place to a desert as dreary as the Kalahari itself. The river appeared interminable, and it became a matter of serious consideration how to get a sufficiency of food. Fish was scarce, or at least hard to catch, and the arid soil provided no means of sustenance for antelopes, so that nothing was to be gained from the chase. Carnivorous animals also had quite disappeared, and the silence of the night was broken, not by the roar of wild beasts, but by the croaking of frogs in a discordant chorus, which Cameron has compared to the clanking of hammers and the grating of files in a ship-builder's yard. Far away both to the east and west the outlines of hills could be faintly discerned, but the shores on either hand were perfectly flat and devoid of trees. Euphorbias, it is true, grew in considerable numbers, but as they were only of the oil-producing species, and not the kind from which cassava or manioc is procured, they were useless in an alimentary point of view. Dick was becoming more and more perplexed, when Hercules happened to mention that the natives often eat young fern-fronds and the pith of the papyrus, and that before now he had himself been reduced to the necessity of subsisting on nothing better. "We must try them," said Dick. Both ferns and papyrus abounded on the banks, and a meal was prepared, the sweet soft pith of the papyrus being found very palatable. Jack in particular appeared to enjoy it extremely, but it was not in any way a satisfying diet. Thanks to Cousin Benedict, a fresh variety in the matter of food was found on the following day. Since the discovery of the "Hexapodes Benedictus" he had recovered his spirits, and, having fastened his prize safely inside his hat, he wandered about, as often as he had a chance, in his favourite pursuit of insect-hunting. As he was rummaging in the long grass, he put up a bird which flew but a very short distance. Benedict recognized it by its peculiar note, and, seeing Dick take his gun to aim at it, exclaimed,- "Don't fire, don't fire! that bird will be worth nothing for food among five of us." "It will be dinner enough for Jack," said Dick, who, finding that the bird did not seem in a hurry to make its escape, delayed his shot for a moment, without intending to be diverted from his purpose of securing it. "You mustn't fire," insisted Benedict, "it is an indicator; it will show you where there are lots of honey." Aware that a few pounds of honey would really be of more value than a little bird, Dick lowered his gun, and in company with the entomologist set off to follow the indicator, which seemed, by alternately flying and stopping, to be inviting them to come on, and they had but a little way to go before they observed several swarms of bees buzzing around some old stems hidden amongst the euphorbias. Notwithstanding Benedict's remonstrances against depriving the bees of the fruits of their industry, Dick instantly set to work, and without remorse suffocated them by burning dry grass underneath. Having secured a good amount of honey, he left the comb to the indicator as its share of the booty, and went back with his companion to the canoe. The honey was acceptable, but it did not do much to alleviate the cravings of hunger. Next day it happened that they had just stopped for their accustomed rest, when they observed that an enormous swarm of grasshoppers had settled at the mouth of a creek close by. Two or three deep they covered the soil, myriads and myriads of them adhering to every shrub. "The natives eat those grasshoppers," said Benedict, "and like them too." The remark produced an instant effect; all hands were busied in collecting them, and a large supply was quickly gathered: the canoe might have been filled ten times over. Grilled over a slow fire, they were found to be very palatable eating, and, spite of his qualms of conscience, Benedict himself made a hearty meal. But although the gnawings of absolute hunger were thus assuaged, all the travellers began to long most anxiously for the voyage to come to an end. The mode of transit indeed might be less exhausting to the bodily powers than a land march would have been, but the excessive heat by day, the damp mists at night, and the incessant attacks of mosquitoes, all combined to render the passage extremely trying. There was no telling how long it would last, and Dick was equally uncertain whether it might end in a few days, or be protracted for a month. The direction which the stream was taking was itself a subject of perplexity. A fresh surprise was now in store. As Jack, a few mornings afterwards, was standing at the bow peering through an aperture in the grass canopy above him, he suddenly turned round and cried,- "The sea! the sea!" Dick started forwards, and looked eagerly in the same direction. A large expanse of water was visible in the horizon, but after having surveyed it for a moment or two, he said,- "No, Jack, it is not the sea, it is a great river; it is running west, and I suppose this river runs into it. Perhaps it is the Zaire." "Let us hope it is," said Mrs. Weldon earnestly. Most cordially did Dick Sands re-echo her words, being well aware that at the mouth of that river were Portuguese villages, where a refuge might assuredly be found. For several succeeding days the canoe, still concealed by its covering, floated on the silvery surface of this new-found stream. On either side the banks became less arid, and there seemed everything to encourage the few survivors of the "Pilgrim" to believe that they would soon see the last of the perils and toils of their journey. They were too sanguine. Towards three o'clock on the morning of the 18th, Dick, who was at his usual post at the bow, fancied he heard a dull rumbling towards the west. Mrs. Weldon, Jack, and Benedict were all asleep. Calling Hercules to him, he asked him whether he could not hear a strange noise. The night was perfectly calm, and not a breath of air was stirring. The negro listened attentively, and suddenly, his eyes sparkling with delight, exclaimed,- "Yes, captain, I hear the sea!" Dick shook his head and answered,- "It is not the sea, Hercules." "Not the sea!" cried the negro, "then what can it be?" "We must wait till daybreak," replied Dick, "and meanwhile we shall have to keep a sharp look-out." Hercules returned to his place, but only to continue listening with ever-increasing curiosity. The rumbling perceptibly increased till it became a continued roar. With scarcely any intervening twilight night passed into day. Just in front, scarcely more than half a mile ahead, a great mist was hanging over the river; it was not an ordinary fog, and when the sun rose, the light of the dawn caused a brilliant rainbow to arch itself from shore to shore. In a voice so loud that it awoke Mrs. Weldon, Dick gave his order to Hercules to steer for the bank:- "Quick, quick, Hercules! ashore! ashore! there are cataracts close ahead!" And so it was. Within little more than a quarter of a mile the bed of the river sank abruptly some hundred feet, and the foaming waters rushed down in a magnificent fall with irresistible velocity. A few minutes more and the canoe must have been swallowed in the deep abyss. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Jules Verne (2005). Dick sands, the boy Captain. Urbana, Illinois: Project Gutenberg. Retrieved October 2022 https://www.gutenberg.org/ebooks/3250.html.images This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at , located at . www.gutenberg.org https://www.gutenberg.org/policy/license.html

AN ANXIOUS VOYAGE

Dick Sands the Boy Captain by Jules Verne, is part of the HackerNoon Books Series. You can jump to any chapter in this book . AN EXCITING CHASE here AN EXCITING CHASE. To say the truth, it was the very vaguest of hopes to which Mrs. Weldon had been clinging, yet it was not without some thrill of disappointment that she heard from the lips of old Alvez himself that Dr. Livingstone had died at a little village on Lake Bangweolo. There had appeared to be a sort of a link binding her to the civilized world, but it was now abruptly snapped, and nothing remained for her but to make what terms she could with the base and heartless Negoro. On the 14th, the day appointed for the interview, he made his appearance at the hut, firmly resolved to make no abatement in the terms that he had proposed, Mrs. Weldon, on her part, being equally determined not to yield to the demand. "There is only one condition," she avowed, "upon which I will acquiesce. My husband shall not be required to come up the country here." Negoro hesitated; at length he said that he would agree to her husband being taken by ship to Mossamedes, a small port in the south of Angola, much frequented by slavers, whither also, at a date hereafter to be fixed, Alvez should send herself with Jack and Benedict; the stipulation was confirmed that the ransom should be 100,000 dollars, and it was further made part of the contract that Negoro should be allowed to depart as an honest man. Mrs. Weldon felt she had gained an important point in thus sparing her husband the necessity of a journey to KazonndÃ©, and had no apprehensions about herself on her way to Mossamedes, knowing that it was to the interest of Alvez and Negoro alike to attend carefully to her wants. Upon the terms of the covenant being thus arranged, Mrs. Weldon wrote such a letter to her husband as she knew would bring him with all speed to Mossamedes, but she left it entirely to Negoro to represent himself in whatever light he chose. Once in possession of the document, Negoro lost no time in starting on his errand. The very next morning, taking with him about twenty negroes, he set off towards the north, alleging to Alvez as his motive for taking that direction, that he was not only going to embark somewhere at the mouth of the Congo, but that he was anxious to keep as far as possible from the prison-houses of the Portuguese, with which already he had been involuntarily only too familiar. After his departure, Mrs. Weldon resolved to make the best of her period of imprisonment, aware that it could hardly be less than four months before he would return. She had no desire to go beyond the precincts assigned her, even had the privilege been allowed her; but warned by Negoro that Hercules was still free, and might at any time attempt a rescue, Alvez had no thought of permitting her any unnecessary liberty. Her life therefore soon resumed its previous monotony. The daily routine went on within the enclosure pretty much as in other parts of the town, the women all being employed in various labours for the benefit of their husbands and masters. The rice was pounded with wooden pestles; the maize was peeled and winnowed, previously to extracting the granulous substance for the drink which they call ; the sorghum had to be gathered in, the season of its ripening being marked by festive observances; there was a fragrant oil to be expressed from a kind of olive named the ; the cotton had to be spun on spindles, which were hardly less than a foot and a half in length; there was the bark of trees to be woven into textures for wearing; the manioc had to be dug up, and the cassava procured from its roots; and besides all this, there was the preparation of the soil for its future plantings, the usual productions of the country being the beans, growing in pods fifteen inches long upon stems twenty feet high, the , from which they procure a serviceable oil, the pea, the blossoms of which are used to give a flavour to the insipid sorghum, cucumbers, of which the seeds are roasted as chestnuts, as well as the common crops of coffee, sugar, onions, guavas, and sesame. mtyellÃ© mpafoo moritsanÃ© arachides chilobÃ© To the women's lot, too, falls the manipulation of all the fermented drinks, the , made from bananas, the , and various other liquors. Nor should the care of all the domestic animals be forgotten; the cows that will not allow themselves to be milked unless they can see their calf, or a stuffed representative of it; the short-horned heifers that not unfrequently have a hump; the goats that, like slaves, form part of the currency of the country; the pigs, the sheep, and the poultry. malafoo pombÃ© The men, meanwhile, smoke their hemp or tobacco, hunt buffaloes or elephants, or are hired by the dealers to join in the slave-raids; the harvest of slaves, in fact, being a thing of as regular and periodic recurrence as the ingathering of the maize. In her daily strolls, Mrs. Weldon would occasionally pause to watch the women, but they only responded to her notice by a long stare or by a hideous grimace; a kind of natural instinct made them hate a white skin, and they had no spark of commiseration for the stranger who had been brought among them; Halima, however, was a marked exception, she grew more and more devoted to her mistress, and by degrees, the two became able to exchange many sentences in the native dialect. Jack generally accompanied his mother. Naturally enough he longed to get outside the enclosure, but still he found considerable amusement in watching the birds that built in a huge baobab that grew within; there were maraboos making their nests with twigs; there were scarlet-throated with nests like weaver-birds; widow birds that helped themselves liberally to the thatch of the souimangas [Illustration: The insufferable heat had driven all the residents within the depÃ³t indoors.] huts; with their tuneful song; grey parrots, with bright red tails, called by the Manyuema, who apply the same name to their reigning chiefs; and insect-eating , like grey linnets with large red beaks. Hundreds of butterflies flitted about, especially in the neighbourhood of the brooks; but these were more to the taste of Cousin Benedict than of little Jack; over and over again the child expressed his regret that he could not see over the walls, and more than ever he seemed to miss his friend Dick, who had taught him to climb a mast, and who he was sure would have fine fun with him in the branches of the trees, which were growing sometimes to the height of a hundred feet. calaos roufs drongos So long as the supply of insects did not fail, Benedict would have been contented to stay on without a murmur in his present quarters. True, without his glasses he worked at a disadvantage; but he had had the good fortune to discover a minute bee that forms its cells in the holes of worm-eaten wood, and a "sphex" that practises the craft of the cuckoo, and deposits its eggs in an abode not prepared by itself. Mosquitos abounded in swarms, and the worthy naturalist was so covered by their stings as to be hardly recognizable; but when Mrs. Weldon remonstrated with him for exposing himself so unnecessarily, he merely scratched the irritated places on his skin, and said- "It is their instinct, you know; it is their instinct." On the 17th of June an adventure happened to him which was attended with unexpected consequences. It was about eleven o'clock in the morning. The insufferable heat had driven all the residents within the dÃ©pÃ´t indoors, and not a native was to be seen in the streets of KazonndÃ©. Mrs. Weldon was dozing; Jack was fast asleep. Benedict himself, sorely against his will, for he heard the hum of many an insect in the sunshine, had been driven to the seclusion of his cabin, and was falling into an involuntary siesta. Suddenly a buzz was heard, an insect's wing vibrating some fifteen thousand beats a second! "A hexapod!" cried Benedict, sitting up. Short-sighted though he was, his hearing was acute, and his perception made him thoroughly convinced that he was in proximity to some giant specimen of its kind. Without moving from his seat he did his utmost to ascertain what it was; he was determined not to flinch from the sharpest of stings if only he could get the chance of capturing it. Presently he made out a large black speck flitting about in the few rays of daylight that were allowed to penetrate the hut. With bated breath he waited in eager expectation. The insect, after long hovering above him, finally settled on his head. A smile of satisfaction played about his lips as he felt it crawling lightly through his hair. Equally fearful of missing or injuring it, he restrained his first impulse to grasp it in his hand. "I will wait a minute," he thought; "perhaps it may creep down my nose; by squinting a little perhaps I shall be able to see it." For some moments hope alternated with fear. There sat Benedict with what he persuaded himself was some new African hexapod perched upon his head, and agitated by doubts as to the direction in which it would move. Instead of travelling in the way he reckoned along his nose, might it not crawl behind his ears or down his neck, or, worse than all, resume its flight in the air? Fortune seemed inclined to favour him. After threading the entanglement of the naturalist's hair the insect was felt to be descending his forehead. With a fortitude not unworthy of the Spartan who suffered his breast to be gnawed by a fox, nor of the Roman hero who plunged his hand into the red-hot coals, Benedict endured the tickling of the six small feet, and made not a motion that might frighten the creature into taking wing. After making repeated circuits of his forehead, it passed just between his eyebrows; there was a moment of deep suspense lest it should once more go upwards; but it soon began to move again; neither to the right nor to the left did it turn, but kept straight on over the furrows made by the constant rubbing of the spectacles, right along the arch of the cartilage till it [Illustration: Before long the old black speck was again flitting just above his head. 432.] Page reached the extreme tip of the nose. Like a couple of movable lenses, Benedict's two eyes steadily turned themselves inwards till they were directed to the proper point. "Good!" he whispered to himself. He was exulting at the discovery that what he had been waiting for so patiently was a rare specimen of the tribe of the CicindelidÃ¦, peculiar to the districts of Southern Africa. "A tuberous manticora!" he exclaimed. The insect began to move again, and as it crawled down to the entrance of the nostrils the tickling sensation became too much for endurance, and Benedict sneezed. He made a sudden clutch, but of course he only caught his own nose. His vexation was very great, but he did not lose his composure; he knew that the manticora rarely flies very high, and that more frequently than not it simply crawls. Accordingly he groped about a long time on his hands and knees, and at last he found it basking in a ray of sunshine within a foot of him. His resolution was soon taken. He would not run the risk of crushing it by trying to catch it, but would make his observations on it as it crawled; and so with his nose close to the ground, like a dog upon the scent, he followed it on all fours, admiring it and examining it as it moved. Regardless of the heat he not only left the doorway of his hut, but continued creeping along till he reached the enclosing palisade. At the foot of the fence the manticora, according to the habits of its kind, began to seek a subterranean retreat, and coming to the opening of a mole-track entered it at once. Benedict quite thought he had now lost sight of his prize altogether, but his surprise was very great when he found that the aperture was at least two feet wide, and that it led into a gallery which would admit his whole body. His momentary feeling of astonishment, however, gave way to his eagerness to follow up the hexapod, and he continued burrowing like a ferret. Without knowing it, he actually passed under the palisading, and was now beyond it;-the mole-track, in fact, was a communication that had been made between the interior and exterior of the enclosure. Benedict had obtained his freedom, but so far from caring in the least for his liberty he continued totally absorbed in the pursuit upon which he had started. He watched with unflagging vigilance, and it was only when the hexapod expanded its wings as if for flight that he prepared to imprison it in the hollow of his hand. All at once, however, he was taken by surprise; a whizz and a whirr and the prize was gone! Disappointed rather than despairing, Benedict raised himself up, and looked about him. Before long the old black speck was again flitting just above his head. There was every reason to hope that it would ultimately settle once more upon the ground, but on this side of the palisade there was a large forest a little way to the north, and if the manticora were to get into its mass of foliage all hope of keeping it in view would be lost, and there would be an end of the proud expectation of storing it in the tin box, to be preserved among the rest of the entomological wonders. After a while the insect descended to the earth; it did not rest at all, nor crawl as it had done previously, but made its advance by a series of rapid hops. This made the chase for the near-sighted naturalist a matter of great difficulty; he put his face as close to the ground as possible, and kept starting off and stopping and starting off again with his arms extended like a swimming frog, continually making frantic clutches to find as continually that his grasp had been eluded. After running till he was out of breath, and scratching his hands against the brushwood and the foliage till they bled, he had the mortification of feeling the insect dash past his ear with what might be a defiant buzz, and finding that it was out of sight for ever. "Ungrateful hexapod!" he cried in dismay, "I intended to honour you with the best place in my collection." He knew not what to do, and could not reconcile himself to the loss; he reproached himself for not having secured the manticora at the first; he gazed at the forest till he [Illustration: For that day at least Cousin Benedict had lost his chance of being the happiest of entomologists. ] Page 435. persuaded himself he could see the coveted insect in the distance, and, seized with a frantic impulse, exclaimed,- "I will have you yet!" He did not even yet realize the fact that he had gained his liberty, but heedless of everything except his own burning disappointment, and at the risk of being attacked by natives or beset by wild beasts, he was just on the very point of dashing into the heart of the wood when suddenly a giant form confronted him, as suddenly a giant hand seized him by the nape of his neck, and, lifting him up, carried him off with apparently as little exertion as he could himself have carried off his hexapod! For that day at least Cousin Benedict had lost his chance of being the happiest of entomologists. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Jules Verne (2005). Dick sands, the boy Captain. Urbana, Illinois: Project Gutenberg. Retrieved October 2022 https://www.gutenberg.org/ebooks/3250.html.images This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at , located at . www.gutenberg.org https://www.gutenberg.org/policy/license.html

AN EXCITING CHASE

HAWKINS IN FLORIDA

The Beasts of Tarzan by Edgar Rice Burroughs, is part of the HackerNoon Books Series. You can jump to any chapter in this book here . A Hideous Crew A Hideous Crew The war-canoe with its savage load moved slowly toward the break in the reef through which it must pass to gain the open sea. Tarzan, Mugambi, and Akut wielded the paddles, for the shore kept the west wind from the little sail. Sheeta crouched in the bow at the ape-man’s feet, for it had seemed best to Tarzan always to keep the wicked beast as far from the other members of the party as possible, since it would require little or no provocation to send him at the throat of any than the white man, whom he evidently now looked upon as his master. In the stern was Mugambi, and just in front of him squatted Akut, while between Akut and Tarzan the twelve hairy apes sat upon their haunches, blinking dubiously this way and that, and now and then turning their eyes longingly back toward shore. All went well until the canoe had passed beyond the reef. Here the breeze struck the sail, sending the rude craft lunging among the waves that ran higher and higher as they drew away from the shore. With the tossing of the boat the apes became panic-stricken. They first moved uneasily about, and then commenced grumbling and whining. With difficulty Akut kept them in hand for a time; but when a particularly large wave struck the dugout simultaneously with a little squall of wind their terror broke all bounds, and, leaping to their feet, they all but overturned the boat before Akut and Tarzan together could quiet them. At last calm was restored, and eventually the apes became accustomed to the strange antics of their craft, after which no more trouble was experienced with them. The trip was uneventful, the wind held, and after ten hours’ steady sailing the black shadows of the coast loomed close before the straining eyes of the ape-man in the bow. It was far too dark to distinguish whether they had approached close to the mouth of the Ugambi or not, so Tarzan ran in through the surf at the closest point to await the dawn. The dugout turned broadside the instant that its nose touched the sand, and immediately it rolled over, with all its crew scrambling madly for the shore. The next breaker rolled them over and over, but eventually they all succeeded in crawling to safety, and in a moment more their ungainly craft had been washed up beside them. The balance of the night the apes sat huddled close to one another for warmth; while Mugambi built a fire close to them over which he crouched. Tarzan and Sheeta, however, were of a different mind, for neither of them feared the jungle night, and the insistent craving of their hunger sent them off into the Stygian blackness of the forest in search of prey. Side by side they walked when there was room for two abreast. At other times in single file, first one and then the other in advance. It was Tarzan who first caught the scent of meat—a bull buffalo—and presently the two came stealthily upon the sleeping beast in the midst of a dense jungle of reeds close to a river. Closer and closer they crept toward the unsuspecting beast, Sheeta upon his right side and Tarzan upon his left nearest the great heart. They had hunted together now for some time, so that they worked in unison, with only low, purring sounds as signals. For a moment they lay quite silent near their prey, and then at a sign from the ape-man Sheeta sprang upon the great back, burying his strong teeth in the bull’s neck. Instantly the brute sprang to his feet with a bellow of pain and rage, and at the same instant Tarzan rushed in upon his left side with the stone knife, striking repeatedly behind the shoulder. One of the ape-man’s hands clutched the thick mane, and as the bull raced madly through the reeds the thing striking at his life was dragged beside him. Sheeta but clung tenaciously to his hold upon the neck and back, biting deep in an effort to reach the spine. For several hundred yards the bellowing bull carried his two savage antagonists, until at last the blade found his heart, when with a final bellow that was half-scream he plunged headlong to the earth. Then Tarzan and Sheeta feasted to repletion. After the meal the two curled up together in a thicket, the man’s black head pillowed upon the tawny side of the panther. Shortly after dawn they awoke and ate again, and then returned to the beach that Tarzan might lead the balance of the pack to the kill. When the meal was done the brutes were for curling up to sleep, so Tarzan and Mugambi set off in search of the Ugambi River. They had proceeded scarce a hundred yards when they came suddenly upon a broad stream, which the Negro instantly recognized as that down which he and his warriors had paddled to the sea upon their ill-starred expedition. The two now followed the stream down to the ocean, finding that it emptied into a bay not over a mile from the point upon the beach at which the canoe had been thrown the night before. Tarzan was much elated by the discovery, as he knew that in the vicinity of a large watercourse he should find natives, and from some of these he had little doubt but that he should obtain news of Rokoff and the child, for he felt reasonably certain that the Russian would rid himself of the baby as quickly as possible after having disposed of Tarzan. He and Mugambi now righted and launched the dugout, though it was a most difficult feat in the face of the surf which rolled continuously in upon the beach; but at last they were successful, and soon after were paddling up the coast toward the mouth of the Ugambi. Here they experienced considerable difficulty in making an entrance against the combined current and ebb tide, but by taking advantage of eddies close in to shore they came about dusk to a point nearly opposite the spot where they had left the pack asleep. Making the craft fast to an overhanging bough, the two made their way into the jungle, presently coming upon some of the apes feeding upon fruit a little beyond the reeds where the buffalo had fallen. Sheeta was not anywhere to be seen, nor did he return that night, so that Tarzan came to believe that he had wandered away in search of his own kind. Early the next morning the ape-man led his band down to the river, and as he walked he gave vent to a series of shrill cries. Presently from a great distance and faintly there came an answering scream, and a half-hour later the lithe form of Sheeta bounded into view where the others of the pack were clambering gingerly into the canoe. The great beast, with arched back and purring like a contented tabby, rubbed his sides against the ape-man, and then at a word from the latter sprang lightly to his former place in the bow of the dugout. When all were in place it was discovered that two of the apes of Akut were missing, and though both the king ape and Tarzan called to them for the better part of an hour, there was no response, and finally the boat put off without them. As it happened that the two missing ones were the very same who had evinced the least desire to accompany the expedition from the island, and had suffered the most from fright during the voyage, Tarzan was quite sure that they had absented themselves purposely rather than again enter the canoe. As the party were putting in for the shore shortly after noon to search for food a slender, naked savage watched them for a moment from behind the dense screen of verdure which lined the river’s bank, then he melted away up-stream before any of those in the canoe discovered him. Like a deer he bounded along the narrow trail until, filled with the excitement of his news, he burst into a native village several miles above the point at which Tarzan and his pack had stopped to hunt. “Another white man is coming!” he cried to the chief who squatted before the entrance to his circular hut. “Another white man, and with him are many warriors. They come in a great war-canoe to kill and rob as did the black-bearded one who has just left us.” Kaviri leaped to his feet. He had but recently had a taste of the white man’s medicine, and his savage heart was filled with bitterness and hate. In another moment the rumble of the war-drums rose from the village, calling in the hunters from the forest and the tillers from the fields. Seven war-canoes were launched and manned by paint-daubed, befeathered warriors. Long spears bristled from the rude battle-ships, as they slid noiselessly over the bosom of the water, propelled by giant muscles rolling beneath glistening, ebony hides. There was no beating of tom-toms now, nor blare of native horn, for Kaviri was a crafty warrior, and it was in his mind to take no chances, if they could be avoided. He would swoop noiselessly down with his seven canoes upon the single one of the white man, and before the guns of the latter could inflict much damage upon his people he would have overwhelmed the enemy by force of numbers. Kaviri’s own canoe went in advance of the others a short distance, and as it rounded a sharp bend in the river where the swift current bore it rapidly on its way it came suddenly upon the thing that Kaviri sought. So close were the two canoes to one another that the black had only an opportunity to note the white face in the bow of the oncoming craft before the two touched and his own men were upon their feet, yelling like mad devils and thrusting their long spears at the occupants of the other canoe. But a moment later, when Kaviri was able to realize the nature of the crew that manned the white man’s dugout, he would have given all the beads and iron wire that he possessed to have been safely within his distant village. Scarcely had the two craft come together than the frightful apes of Akut rose, growling and barking, from the bottom of the canoe, and, with long, hairy arms far outstretched, grasped the menacing spears from the hands of Kaviri’s warriors. The blacks were overcome with terror, but there was nothing to do other than to fight. Now came the other war-canoes rapidly down upon the two craft. Their occupants were eager to join the battle, for they thought that their foes were white men and their native porters. They swarmed about Tarzan’s craft; but when they saw the nature of the enemy all but one turned and paddled swiftly up-river. That one came too close to the ape-man’s craft before its occupants realized that their fellows were pitted against demons instead of men. As it touched Tarzan spoke a few low words to Sheeta and Akut, so that before the attacking warriors could draw away there sprang upon them with a blood-freezing scream a huge panther, and into the other end of their canoe clambered a great ape. At one end the panther wrought fearful havoc with his mighty talons and long, sharp fangs, while Akut at the other buried his yellow canines in the necks of those that came within his reach, hurling the terror-stricken blacks overboard as he made his way toward the centre of the canoe. Kaviri was so busily engaged with the demons that had entered his own craft that he could offer no assistance to his warriors in the other. A giant of a white devil had wrested his spear from him as though he, the mighty Kaviri, had been but a new-born babe. Hairy monsters were overcoming his fighting men, and a black chieftain like himself was fighting shoulder to shoulder with the hideous pack that opposed him. Kaviri battled bravely against his antagonist, for he felt that death had already claimed him, and so the least that he could do would be to sell his life as dearly as possible; but it was soon evident that his best was quite futile when pitted against the superhuman brawn and agility of the creature that at last found his throat and bent him back into the bottom of the canoe. Presently Kaviri’s head began to whirl—objects became confused and dim before his eyes—there was a great pain in his chest as he struggled for the breath of life that the thing upon him was shutting off for ever. Then he lost consciousness. When he opened his eyes once more he found, much to his surprise, that he was not dead. He lay, securely bound, in the bottom of his own canoe. A great panther sat upon its haunches, looking down upon him. Kaviri shuddered and closed his eyes again, waiting for the ferocious creature to spring upon him and put him out of his misery of terror. After a moment, no rending fangs having buried themselves in his trembling body, he again ventured to open his eyes. Beyond the panther kneeled the white giant who had overcome him. The man was wielding a paddle, while directly behind him Kaviri saw some of his own warriors similarly engaged. Back of them again squatted several of the hairy apes. Tarzan, seeing that the chief had regained consciousness, addressed him. “Your warriors tell me that you are the chief of a numerous people, and that your name is Kaviri,” he said. “Yes,” replied the black. “Why did you attack me? I came in peace.” “Another white man ‘came in peace’ three moons ago,” replied Kaviri; “and after we had brought him presents of a goat and cassava and milk, he set upon us with his guns and killed many of my people, and then went on his way, taking all of our goats and many of our young men and women.” “I am not as this other white man,” replied Tarzan. “I should not have harmed you had you not set upon me. Tell me, what was the face of this bad white man like? I am searching for one who has wronged me. Possibly this may be the very one.” “He was a man with a bad face, covered with a great, black beard, and he was very, very wicked—yes, very wicked indeed.” “Was there a little white child with him?” asked Tarzan, his heart almost stopped as he awaited the black’s answer. “No, bwana,” replied Kaviri, “the white child was not with this man’s party—it was with the other party.” “Other party!” exclaimed Tarzan. “What other party?” “With the party that the very bad white man was pursuing. There was a white man, woman, and the child, with six Mosula porters. They passed up the river three days ahead of the very bad white man. I think that they were running away from him.” A white man, woman, and child! Tarzan was puzzled. The child must be his little Jack; but who could the woman be—and the man? Was it possible that one of Rokoff’s confederates had conspired with some woman—who had accompanied the Russian—to steal the baby from him? If this was the case, they had doubtless purposed returning the child to civilization and there either claiming a reward or holding the little prisoner for ransom. But now that Rokoff had succeeded in chasing them far inland, up the savage river, there could be little doubt but that he would eventually overhaul them, unless, as was still more probable, they should be captured and killed by the very cannibals farther up the Ugambi, to whom, Tarzan was now convinced, it had been Rokoff’s intention to deliver the baby. As he talked to Kaviri the canoes had been moving steadily up-river toward the chief’s village. Kaviri’s warriors plied the paddles in the three canoes, casting sidelong, terrified glances at their hideous passengers. Three of the apes of Akut had been killed in the encounter, but there were, with Akut, eight of the frightful beasts remaining, and there was Sheeta, the panther, and Tarzan and Mugambi. Kaviri’s warriors thought that they had never seen so terrible a crew in all their lives. Momentarily they expected to be pounced upon and torn asunder by some of their captors; and, in fact, it was all that Tarzan and Mugambi and Akut could do to keep the snarling, ill-natured brutes from snapping at the glistening, naked bodies that brushed against them now and then with the movements of the paddlers, whose very fear added incitement to the beasts. At Kaviri’s camp Tarzan paused only long enough to eat the food that the blacks furnished, and arrange with the chief for a dozen men to man the paddles of his canoe. Kaviri was only too glad to comply with any demands that the ape-man might make if only such compliance would hasten the departure of the horrid pack; but it was easier, he discovered, to promise men than to furnish them, for when his people learned his intentions those that had not already fled into the jungle proceeded to do so without loss of time, so that when Kaviri turned to point out those who were to accompany Tarzan, he discovered that he was the only member of his tribe left within the village. Tarzan could not repress a smile. “They do not seem anxious to accompany us,” he said; “but just remain quietly here, Kaviri, and presently you shall see your people flocking to your side.” Then the ape-man rose, and, calling his pack about him, commanded that Mugambi remain with Kaviri, and disappeared in the jungle with Sheeta and the apes at his heels. For half an hour the silence of the grim forest was broken only by the ordinary sounds of the teeming life that but adds to its lowering loneliness. Kaviri and Mugambi sat alone in the palisaded village, waiting. Presently from a great distance came a hideous sound. Mugambi recognized the weird challenge of the ape-man. Immediately from different points of the compass rose a horrid semicircle of similar shrieks and screams, punctuated now and again by the blood-curdling cry of a hungry panther. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Edgar Rice Burroughs (1993). The Beasts of Tarzan. Urbana, Illinois: Project Gutenberg. Retrieved October 2022 https://www.gutenberg.org/cache/epub/85/pg85-images.html This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at www.gutenberg.org , located at https://www.gutenberg.org/policy/license.html .

A Hideous Crew

Jane Clayton Again a Prisoner

The Jungle Toll

“King of the Apes”

dignity through identity BanQu is providing a permanent identity-based solution for the unbanked through a ground-breaking initiative. The blockchain-as-a-service software company aims to address the issue of global poverty by identifying those who have remained anonymous in vulnerable situations. Jointly founded in 2016 by and (himself a refugee from Somalia), BanQu gives an identity to the unbanked by equipping the unverified with an so that they can participate equally in the global economy. Ashish Ghadnis Hamse Warfa economic and transactional identity How BanQu Gives an Identity to the Unbanked About worldwide have little to no access to credit and banking services, which categorizes them as unbanked/underbanked individuals. People in this category often include refugees, internally displaced persons and the world’s poorest, who remain excluded from the global economy and have no record of their education, financial history, past employment or other information needed to access banks and credit services. 2.5 billion people BanQu gives an identity to the unbanked by allowing them to create a through simple SMS-enabled phones, as people living in extreme poverty may not have access to a smartphone. The user keeps track of everything from educational qualifications to transaction history like harvest information, which is critical for farmers, thereby gradually building a credit history. With this information, they can open bank accounts, apply for micro-loans, own property and even access healthcare and other basic services. Integrating with modern services allows the unbanked to become part of the global economy. secure online profile Through the initiative, BanQu gives an identity to the unbanked by aiming to attain several of the U.N. and creating business models that are not only a commercial success but also have a in developing and underdeveloped regions. Sustainable Development Goals positive social impact Achieving Sustainable Development Goals Through Services The company runs on the principle that every human deserves a health record for timely care. Unleashing the power of blockchain technology, it provides protection of the individual’s privacy. It is looking to provide accurate and real-time information on birth registrations, vaccinations, HIV and other medical supplies and early warning of outbreaks of infectious diseases. These are extremely helpful for those who have been hit the hardest during times of conflict, natural disasters and forced migration. Good Health and Well Being: The company’s certified education platform (vocational or otherwise) allows youth to be recognized in order to gain employment, building economic resilience for themselves and the communities. A certified education record coupled with a training history is critical for that are left out of the global economy. Quality Education: vulnerable youths A reality in the developing world is that women (especially farmers and micro-entrepreneurs) are more likely to remain in poverty in comparison to men due to unequal access to paid work, education and property. Due to their lack of access to financial services, they are unable to provide their basic history of harvest, crops/goods pricing and land/asset rights. BanQu gives an identity to the unbanked so as to solve the issue of gender inequality for women farmers and micro, small and mid-sized enterprises in emerging markets. By connecting them to global supply chains, the firm plans on creating transactional records that can be accessed from anywhere in the world. Gender Equality: BanQu’s platform seeks to provide traceability in the agricultural supply chains. This adds end-to-end transparency from manufacturer to farmer. The company is piloting small-plot farmer land mapping with a focus on women farmers in Latin America. This could pave way for cases where farmers had difficulty in getting either due to outdated property registries or lack of land rights. Life on Land: In June 2018, BanQu partnered with the world’s leading brewer, Anheuser-Busch InBev, to connect as many as 2,000 to a mobile platform as they harvest and sell a projected 2,000 tons of cassava. This tech-driven approach could help the farmers yield high-quality starch used in beer by the end of Zambia’s growing season in August. Partnerships for the Goals: Zambian farmers Through the app, BanQu gives an identity to the unbanked and has so far linked more than 15,000 last-mile farmers, displaced persons and refugees to the platform in eight countries. Additionally, the company aims to help lift 100 million people out of extreme poverty through the use of blockchain technology by the year 2028. Originally Published on The Borgen Magazine by Deena Zaidi

How One Blockchain Caters to the Unbanked

OmiseGO (OMG): Real Problems, Real Solutions — EVERYTHING You Need To Know & 8 Reasons To Buy…

Last week we got acquainted with the core concepts of Tensor Flow. We learned about the differences between constants, placeholders, and variable tensors. Both the Haskell and Python bindings have functions to represent these. The Python version was a bit simpler though. Once we had our tensors, we wrote a program that “learned” a simple linear equation. This week, we’re going to solve an actual machine learning problem. We’re going to use the , which contains measurements of different Iris flowers. Each flower belongs to one of three species. Our program will “learn” a function choosing the species from the measurements. This function will involved a fully-connected neural network. Iris data set Formatting our Input The first step in pretty much any machine learning problem is data processing. After all, our data doesn’t magically get resolved into Haskell data types. Luckily, is a great library to help us out. The Iris data set consists of data in .csv files that each have a header line and then a series of records. They look a bit like this: Cassava 120,4,setosa,versicolor,virginica6.4,2.8,5.6,2.2,25.0,2.3,3.3,1.0,14.9,2.5,4.5,1.7,24.9,3.1,1.5,0.1,0... Each line contains one record. A record has four flower measurements, and a final label. In this case, we have three types of flowers we are trying to classify between: Iris Setosa, Iris Versicolor, and Iris Virginica. So the last column contains the numbers 0,1, and 2, corresponding to these respective classes. Let’s create a data type representing each record. Then we can parse the file line-by-line. Our type will contain the feature data and the resulting label. This type will act as a bridge between our raw data and the tensor format we’ll need to run our learning algorithm. We’ll derive the “Generic” typeclass for our record type, and use this to get . Once our type has an instance for , we can parse it with ease. As a note, throughout this article, I’ll be the imports section from the code samples. I’ve included a full list of imports from these files as an appendix at the bottom. We'll also be using the extension throughout. IrisRecord FromRecord FromRecord omitting OverloadedLists {-# LANGUAGE DeriveGeneric #-}{-# LANGUAGE OverloadedLists #-}

...

data IrisRecord = IrisRecord { field1 :: Float , field2 :: Float , field3 :: Float , field4 :: Float , label :: Int64 } deriving (Generic)

instance FromRecord IrisRecord Now that we have our type, we’ll write a function, , that will read our data in from a CSV file. readIrisFromFile readIrisFromFile :: FilePath -> IO (Vector IrisRecord)readIrisFromFile fp = do contents <- readFile fp let contentsAsBs = pack contents let results = decode HasHeader contentsAsBs :: Either String (Vector IrisRecord) case results of Left err -> error err Right records -> return records We won’t want to always feed our entire data set into our training system. So given a whole slew of these items, we should be able to pick out a random sample. sampleSize :: IntsampleSize = 10

chooseRandomRecords :: Vector IrisRecord -> IO (Vector IrisRecord)chooseRandomRecords records = do let numRecords = Data.Vector.length records chosenIndices <- take sampleSize <$> shuffleM [0..(numRecords - 1)] return $ fromList $ map (records !) chosenIndices Once we’ve selected our vector of records to use for each run, we’re still not done. We need to take these records and transform them into the that we’ll feed into our algorithm. We create items of by feeding in a shape and then a 1-dimensional vector of values. First, we need to know the shapes of our input and output. Both of these depend on the number of rows in the sample. The “input” will have a column for each of the four features in our set. The output meanwhile will have a single column for the label values. TensorData TensorData irisFeatures :: Int64irisFeatures = 4

irisLabels :: Int64irisLabels = 3

convertRecordsToTensorData :: Vector IrisRecord -> (TensorData Float, TensorData Int64)convertRecordsToTensorData records = (input, output) where numRecords = Data.Vector.length records input = encodeTensorData [fromIntegral numRecords, irisFeatures] (undefined) output = encodeTensorData [fromIntegral numRecords] (undefined) Now all we need to do is take the various records and turn them into one dimensional vectors to encode. Here’s the final function: convertRecordsToTensorData :: Vector IrisRecord -> (TensorData Float, TensorData Int64)convertRecordsToTensorData records = (input, output) where numRecords = Data.Vector.length records input = encodeTensorData [fromIntegral numRecords, irisFeatures] (fromList $ concatMap recordToInputs records) output = encodeTensorData [fromIntegral numRecords] (label <$> records) recordToInputs :: IrisRecord -> [Float] recordToInputs rec = [field1 rec, field2 rec, field3 rec, field4 rec] Neural Network Basics Now that we’ve got that out of the way, we can start writing our model. Remember, we want to perform two different actions with our model. First, we want to be able to take our training input and train the weights. Second, we want to be able to pass a test data set and determine the error rate. We can represent these two different functions as a single object. Remember the monad, where we run all our Tensor Flow activities. The training will run an action that alters the variables but returns nothing. The error rate calculation will return us a float value. Model Session data Model = Model { train :: TensorData Float -- Training input -> TensorData Int64 -- Training output -> Session () , errorRate :: TensorData Float -- Test input -> TensorData Int64 -- Test output -> Session Float } Now we’re going to build a fully-connected neural network. We’ll have 4 input units (1 for each of the different features), and then we’ll have 3 output units (1 for each of the classes we’re trying to represent). In the middle, we’ll use a hidden layer consisting of 10 units. This means we’ll need two sets of weights and biases. We’ll write a function that, when given dimensions, will give us the variable tensors for each layer. We want the weight and bias tensors, plus the result tensor of the layer. buildNNLayer :: Int64 -> Int64 -> Tensor v Float -> Build (Variable Float, Variable Float, Tensor Build Float)buildNNLayer inputSize outputSize input = do weights <- truncatedNormal (vector [inputSize, outputSize]) >>= initializedVariable bias <- truncatedNormal (vector [outputSize]) >>= initializedVariable let results = (input `matMul` readValue weights) `add` readValue bias return (weights, bias, results) We do this in the monad, which allows us to construct variables, among other things. We’ll use a distribution for all our variables to keep things simple. We specify the size of each variable in a tensor, and then initialize them. Then we’ll create the resulting tensor by multiplying the input by our weights and adding the bias. Build truncatedNormal vector Constructing our Model Now we’ll start building our object, again within the monad. We begin by specifying our input and output placeholders, as well the number of hidden units. We’ll also use a of -1 to account for the fact that we want a variable number of input samples. Model Build batchSize irisFeatures :: Int64irisFeatures = 4

irisLabels :: Int64irisLabels = 3-- ^^ From above

createModel :: Build ModelcreateModel = do let batchSize = -1 -- Allows variable sized batches let numHiddenUnits = 10 inputs <- placeholder [batchSize, irisFeatures] outputs <- placeholder [batchSize] Then we’ll get the nodes for the two layers of variables, as well as their results. Between the layers, we’ll add a “rectifier” activation function : relu (hiddenWeights, hiddenBiases, hiddenResults) <- buildNNLayer irisFeatures numHiddenUnits inputslet rectifiedHiddenResults = relu hiddenResults(finalWeights, finalBiases, finalResults) <- buildNNLayer numHiddenUnits irisLabels rectifiedHiddenResults Now we have to get the inferred classes of each output. This means calling to take the class with the highest probability. We’ll also the vector and then it. These are some Haskell-Tensor-Flow specific terms for getting tensors to the right type. Next, we compare that against our output placeholders to see how many we got correct. Then we’ll make a node for calculating the error rate for this run. argMax cast render actualOutput <- render $ cast $ argMax finalResults (scalar (1 :: Int64))let correctPredictions = equal actualOutput outputserrorRate_ <- render $ 1 - (reduceMean (cast correctPredictions)) Now we have to actually do the work of training. First, we’ll make vectors for our expected outputs. This means converting the label into the vector , and so on. We’ll compare these values against our results (before we took the max), and this gives us our loss function. Then we will make a list of the parameters we want to train. The optimizer will minimize our loss function while modifying the params. oneHot 0 [1,0,0] adam let outputVectors = oneHot outputs (fromIntegral irisLabels) 1 0let loss = reduceMean $ fst $ softmaxCrossEntropyWithLogits finalResults outputVectorslet params = [hiddenWeights, hiddenBiases, finalWeights, finalBiases]train_ <- minimizeWith adam loss params Now we’ve got our and nodes ready. There's one last step here. We have to plug in for the placeholder values and create functions that will take in the tensor data. Remember the pattern from ? We use it again here. Finally, our model is complete! errorRate_ train_ feed last week return $ Model { train = \inputFeed outputFeed -> runWithFeeds [ feed inputs inputFeed , feed outputs outputFeed ] train_ , errorRate = \inputFeed outputFeed -> unScalar <$> runWithFeeds [ feed inputs inputFeed , feed outputs outputFeed ] errorRate_ } Tying it all together Now we’ll write our main function that will run the session. It will have three stages. In the preparation stage, we’ll load our data, and use the function to get our model. Then we’ll train our model for 1000 steps by choosing samples and converting our records to data. Every 100 steps, we'll print the output. Finally, we’ll determine the resulting error ratio by using the test data. build runIris :: FilePath -> FilePath -> IO ()runIris trainingFile testingFile = runSession $ do -- Preparation trainingRecords <- liftIO $ readIrisFromFile trainingFile testRecords <- liftIO $ readIrisFromFile testingFile model <- build createModel

 -- Training forM_ ([0..1000] :: [Int]) $ \i -> do trainingSample <- liftIO $ chooseRandomRecords trainingRecords let (trainingInputs, trainingOutputs) = convertRecordsToTensorData trainingSample (train model) trainingInputs trainingOutputs when (i `mod` 100 == 0) $ do err <- (errorRate model) trainingInputs trainingOutputs liftIO $ putStrLn $ "Current training error " ++ show (err * 100)

 liftIO $ putStrLn ""

 -- Testing let (testingInputs, testingOutputs) = convertRecordsToTensorData testRecords testingError <- (errorRate model) testingInputs testingOutputs liftIO $ putStrLn $ "test error " ++ show (testingError * 100)

 return () Results So when we actually run all this output, we’ll get the following results on our test set. Current training error 60.000004Current training error 30.000002Current training error 39.999996Current training error 19.999998Current training error 10.000002Current training error 10.000002Current training error 19.999998Current training error 19.999998Current training error 10.000002Current training error 10.000002Current training error 0.0

test error 3.333336 Our test sample size was 30, so this means we got 29/30 this time around. Results change though from run to run (I obviously used the best results I found). Since our sample size is so small, we have high entropy here (sometimes the error rate is like 40%). Generally we’ll want to train longer on a larger test set, so that we get more consistent results, but this is a good start. Conclusion In this article we went over the basics of making a neural network using the Haskell Tensor Flow library. We made a fully-connected neural network and fed in real data we parsed using the library. This network was able to learn a function to classify flowers from the Iris data set. Considering the small amount of data, we got some good results. Cassava Come back next week, where we’ll see how we can add some more summary information to our tensor flow graph. We’ll use the tensor board application to view our graph in a visual format. For more details on installing the Haskell Tensor Flow system, check out our . It should walk you through the important steps in running the code on your own machine. In-Depth Tensor Flow Tutorial Perhaps you’ve never tried Haskell before at all, and want to see what it’s like. Maybe I’ve convinced you that Haskell is in fact the future of AI. In that case, you should check out our for some tools on starting with the language. Getting Started Checklist Appendix: All Imports Documentation for Haskell Tensor Flow is still a major work in progress. So I want to make sure I explicitly list the modules you need to import for all the different functions we used here. import Control.Monad (forM_, when)import Control.Monad.IO.Class (liftIO)import Data.ByteString.Lazy.Char8 (pack)import Data.Csv (FromRecord, decode, HasHeader(..))import Data.Int (Int64)import Data.Vector (Vector, length, fromList, (!))import GHC.Generics (Generic)import System.Random.Shuffle (shuffleM)

import TensorFlow.Core (TensorData, Session, Build, render, runWithFeeds, feed, unScalar, build, Tensor, encodeTensorData)import TensorFlow.Minimize (minimizeWith, adam)import TensorFlow.Ops (placeholder, truncatedNormal, add, matMul, relu, argMax, scalar, cast, oneHot, reduceMean, softmaxCrossEntropyWithLogits, equal, vector)import TensorFlow.Session (runSession)import TensorFlow.Variable (readValue, initializedVariable, Variable)

Stories, News & Mentions about CASSAVA

Cassava’s Stories on HackerNoon

Cassava's News & Mentions