Anthony Chen’s quiet character study follows a traumatized Liberian woman (Cynthia Erivo) on a Greek island who befriends an American tour guide (Alia Shawkat).

‘Drift’ Review: Cynthia Erivo Keeps a Breakdown at Bay - The New York Times

Drift is the Revenue Acceleration platform that uses Conversational Marketing and Conversational Sales to help companies grow revenue and increase customer lifetime value, faster

Drift

Now this is an interesting race… The Facebook Messenger team just announced that they are going to compete with Intercom. On Twitter. You read that right. So now they are competing full-steam with , and others. As any good developer knows, every customer has asked the famous question Intercom Drift chatbot Will this chatbot work on my website as well? Up until now, the answer has been: “Yes but..” “Yes but we need some more work…” “Yes but the carousel may not look the same…” “Yes but…” Heck, raised $10 million in order to properly answer this question (and more). From today onwards, developers will be able to tell their clients smooch.io Yes, out of the box. We will need to use Facebook’s plugin on your website. Which may be reasonable for a lot of use cases. And totally not reasonable for a whole lot of other use cases. Given the regulatory framework it does not seem feasible for banks, financial institutions and any organization which is concerned about the private data flowing through their website chat. For startups like us, it may be too limiting. Intercom offers a wide set of features that are very, very useful — CRM-like and auto-messaging capabilities for example. Sure, may potentially build new features and catch up; but its broad audience and ambitions usually limit the depth and complexity of the features that are added to the product. Facebook Still, is a very interesting move. Although still in Beta, you can have a glance of the final user experience on the homepage. Chatfuel The race is now officially open. As the Intercom CEO posted today on ProductHunt, it will be an interesting year From https://www.producthunt.com/posts/messenger-customer-chat

Customer Chat: Facebook vs. Intercom vs. Drift

Despite the woo-woo-seeming presentation of ideas in this and his most popular book , I think Napoleon Hill was onto something, even if it was expressed in a delulu way. Think and Grow Rich The following is the summary of the main concepts of a lesser-known Napoleon Hill book: . First things first - Hill didn’t think he was talking to the devil. He wrote it as a parable about human nature and how we sabotage ourselves. Outwitting the Devil actual What you’ll learn: What happens to the mind when it has no purpose? Mental habits and behaviors for success. How to use human nature to your advantage. The key concept of the book - drifting. Drifting refers to a state of mind that lacks aim or purpose - essentially going through the motions. 9 thought habits that lead to drifting: Fear Superstition Avarice Greed Lust Revenge Anger Vanity Laziness Success - drifter vs non-drifter In Hill’s terms, the difference between a successful and an unsuccessful person is a set of habits. If you’re not successful, it’s because you’re unwilling to train your mind. Drifting ≠ Success. Non-Drifting = Success. Successful people carefully choose the people they surround themselves with because we download the thought habits of the people we interact the most with. Drifter always complains that the world has run out of opportunity and there's nothing to do while the non-drifters will create opportunity no matter the circumstances. Non-drifters provide a service that's useful to as many people as possible; the attitude is to give before you get. How to protect yourself against drifting Value time as your greatest asset. Recognize that the emotion of fear is self-sabotage. Resist it. Decide what you want from life and create a plan to achieve it. Form the habit of being decisive. Don’t leave things unfinished. If temporarily defeated or failed, analyze it to extract the knowledge and advantage it has given you. Your thoughts guide your actions. Be wary of what/how you think. Human nature Hypnotic rhythm - the unconscious part of your mind that tends to solidify your habits and behaviors. These automated behaviours can be good or bad: brushing your teeth, compulsively checking social media, exercising, eating poorly etc. But hypnotic rhythm refers to thinking patterns, too. Do we ruminate or gossip? Are we grateful and forgiving? Do we look for opportunities or focus on what work? These mental habits usually form early on and then become solidified beliefs that produce the illusion that . doesn’t that’s what the world is How to stop drifting You need to develop: DEFINITENESS OF PURPOSE Definite aims, goals, and purpose. Have a plan for how you're gonna achieve them. Plans can change, but the definiteness of purpose doesn’t. If a plan fails don't be discouraged, devise a new plan. MASTERY OVER SELF Master your desire for 1) food 2) sex 3) desire to express loosely organised opinions. The person who talks less has the opportunity to learn more by listening to and observing others. Expressing loosely organized opinions is a destructive habit; don't give opinions unless you’re asked. CONTROLLING ENVIRONMENTAL INFLUENCE (ASSOCIATIONS) Your environment feeds your thought habits. We unconsciously adopt the habits of those we associate with closely. Association through romantic relationships has very strong powers because the minds are connected spiritually. TIME Time converts thought habits into outcomes and either penalizes or rewards you depending on them. Time allows human beings to cultivate wisdom, but wisdom only comes to non-drifters who proactively use their minds for positive thought habits. HARMONY Like attracts like - nature seeks things of a similar nature. Successful people 1) make sure to have harmony among their business associates 2) require their associates to have definiteness of purpose. CAUTION - THINKING THROUGH YOUR PLAN BEFORE YOU ACT Think first act second. A drifter reacts before thinking. Has a plan if things go wrong to avoid unnecessary stress. How the mind works, according to Hill We receive thoughts from the collective consciousness and the Infinite Intelligence (God, Universe, Consciousness - whatever it means to you). All negative desires are nothing but frustrations of positive desires. Thoughts tend to invoke similar thoughts so if you're thinking positively, more positive thoughts will come up and vice versa. The thoughts that we keep having become a part of the hypnotic rhythm law. That's why it's so important to choose your thoughts. Wrap up Would I recommend you read the book? Yes, if you’re not bothered by spiritual-speak. offers a deep dive into the psychology of success through a dialogue with the sabotaging part of ourselves. Outwitting the Devil Key takeaways: Have purpose and vision for your life and guide your actions accordingly. Master self-discipline, both through thought habits and curating the people/information you surround yourself with. Value your time and use it to your advantage by creating positive mental habits that compound over time and bring you ‘luck.’ Also published . here

A Key to Success: Defeat Drifting

DRIFTING INTO DANGER.

Dick Sands the Boy Captain by Jules Verne, is part of the HackerNoon Books Series. You can jump to any chapter in this book . DRIFTING DOWN THE STREAM here DRIFTING DOWN THE STREAM. "Off we go!" It was the voice of Hercules addressing Dick Sands, who, frightfully debilitated by recent sufferings, was leaning against Cousin Benedict for support. Dingo was lying at his feet. Mrs. Weldon gradually recovered her consciousness. Looking around her in amazement she caught sight of Dick. "Dick, is it you?" she muttered feebly. The lad with some difficulty arose, and took her hand in his, while Jack overwhelmed him with kisses. "And who would have thought it was you, Hercules, that carried us away?" said the child; "I did not know you a bit; you were so dreadfully ugly." "I was a sort of a devil, you know, Master Jack," Hercules answered; "and the devil is not particularly handsome;" and he began rubbing his chest vigorously to get rid of the white pattern with which he had adorned it. Mrs. Weldon held out her hand to him with a grateful smile. "Yes, Mrs. Weldon, he has saved you, and although he does not own it, he has saved me too," said Dick. "Saved!" repeated Hercules, "you must not talk about safety, for you are not saved yet." And pointing to Benedict, he continued,- "That's where your thanks are due; unless he had come and informed me all about you and where you were, I should have known nothing, and should have been powerless to aid you." It was now five days since he had fallen in with the entomologist as he was chasing the manticora, and unceremoniously had carried him off. As the canoe drifted rapidly along the stream, Hercules briefly related his adventures since his escape from the encampment on the Coanza. He described how he had followed the kitanda which was conveying Mrs. Weldon; how in the course of his march he had found Dingo badly wounded; how he and the dog together had reached the neighbourhood of KazonndÃ©, and how he had contrived to send a note to Dick, intending to inform him of Mrs. Weldon's destination. Then he went on to say that since his unexpected with Cousin Benedict he had watched very closely for a chance to get into the guardeddÃ©pÃ´t, but until now had entirely failed. A celebrated had been passing on his way through the forest, and he had resolved upon impersonating him as a means of gaining the admittance he wanted. His strength made the undertaking sufficiently easy; and having stripped the magician of his paraphernalia, and bound him securely to a tree, he painted his own body with a pattern like that which he observed on his victim's chest, and having attired himself with the magical garments was quite equipped to impose upon the credulous natives. The result of his stratagem they had all that day witnessed. rencontre mganga He had hardly finished his account of himself when Mrs. Weldon, smiling at his success, turned to Dick. "And how, all this time, my dear boy, has it fared with you?" she asked. Dick said,- "I remember very little to tell you. I recollect being fastened to a stake in the river-bed and the water rising and rising till it was above my head. My last thoughts were about yourself and Jack. Then everything became a blank, and I knew nothing more until I found myself amongst the papyrus on the river-bank, with Hercules tending me like a nurse." "You see I am the right sort of " interposed Hercules; "I am a doctor as well as a conjurer." mganga "But tell me, Hercules, how did you save him?" "Oh, it was not a difficult matter by any means," answered Hercules modestly; "it was dark, you know, so that at the proper moment it was quite possible to wade in amongst the poor wretches at the bottom of the trench, and to wrench the stake from its socket. Anybody could have done it. Cousin Benedict could have done it. Dingo, too, might have done it. Perhaps, after all, it was Dingo that did it." "No, no, Hercules, that won't do," cried Jack; "besides, look, Dingo is shaking his head; he is telling you he didn't do it." "Dingo must not tell tales, Master Jack," said Hercules, laughing. But, nevertheless, although the brave fellow's modesty prompted him to conceal it, it was clear that he had accomplished a daring feat, of which few would have ventured to incur the risk. Inquiry was next made after Tom, Bat, ActÃ¦on, and Austin. His countenance fell, and large tears gathered in his eyes as Hercules told how he had seen them pass through the forest in a slave-caravan. They were gone; he feared they were gone for ever. Mrs. Weldon tried to console him with the hope that they might still be spared to meet again some day; but he shook his head mournfully. She then communicated to Dick the terms of the compact that had been entered into for her own release, and observed that under the circumstances it might really have been more prudent for her to remain in KazonndÃ©. "Then I have made a mistake; I have been an idiot, in bringing you away," said Hercules, ever ready to depreciate his own actions. "No," said Dick; "you have made no mistake; you could not have done better; those rascals, ten chances to one, will only get Mr. Weldon into some trap. We must get to Mossamedes before Negoro arrives; once there, we shall find that the Portuguese authorities will lend us their protection, and when old Alvez arrives to claim his 100,000 dollars-" "He shall receive a good thrashing for his pains," said Hercules, finishing Dick's sentence, and chuckling heartily at the prospect. It was agreed on all hands that it was most important that Negoro's arrival at Mossamedes should be forestalled. The plan which Dick had so long contemplated of reaching the coast by descending some river seemed now in a fair way of being accomplished, and from the northerly direction in which they were proceeding it was quite probable that they would ultimately reach the Zaire, and in that case not actually arrive at S. Paul de Loanda; but that would be immaterial, as they would be sure of finding help anywhere in the colonies of Lower Guinea. On finding himself on the river-bank, Dick's first thought had been to embark upon one of the floating islands that are continually to be seen upon the surface of the African streams, but it happened that Hercules during one of his rambles found a native boat that had run adrift. It was just the discovery that suited their need. It was one of the long, narrow canoes, thirty feet in length by three or four in breadth, that with a large number of paddles can be driven with immense velocity, but by the aid of a single scull can be safely guided down the current of a stream. Dick was somewhat afraid that, to elude observation, it would be necessary to proceed only by night, but as the loss of twelve hours out of the twenty-four would double the length of the voyage, he devised the plan of covering the canoe with a roof of long grass, supported by a horizontal pole from stem to stern, and this not only afforded a shelter from the sun, but so effectually concealed the craft, rudder-scull and all, that the very birds mistook it for one of the natural islets, and red-beaked gulls, black and grey and white kingfishers would frequently alight upon it in search of food. arringhas Though comparatively free from fatigue, the voyage must necessarily be long, and by no means free from danger, and the daily supply of provisions was not easy to procure. If fishing failed, Dick had the one gun which Hercules had carried away with him from the ant-hill, and as he was by no means a bad shot, he hoped to find plenty of game, either along the banks or by firing through a loophole in the thatch. The rate of the current, as far as he could tell, was about two miles an hour, enough to carry them about fifty miles a day; it was a speed, however, that made it necessary for them to keep a sharp look-out for any rocks or submerged trunks of trees, as well as to be on their guard against rapids and cataracts. Dick's strength and spirits all revived at the delight of having Mrs. Weldon and Jack restored to him, and he assumed his post at the bow of the canoe, directing Hercules how to use the scull at the stern. A litter of soft grass was made for Mrs. Weldon, who spent most of her time lying thoughtfully in the shade. Cousin Benedict was very taciturn; he had not recovered the loss of the manticora, and frowned ever and again at Hercules, as if he had not yet forgiven him for stopping him in the chase. Jack, who had been told that he must not be noisy, amused himself by playing with Dingo. The first two days passed without any special incident. The stock of provisions was quite enough for that time, so that there was no need to disembark, and Dick merely lay to for a few hours in the night to take a little necessary repose. The stream nowhere exceeded 150 feet in breadth. The floating islands moved at the same pace as the canoe, and except from some unforeseen circumstance, there could be no apprehension of a collision. The banks were destitute of human inhabitants, but were richly clothed with wild plants, of which the blossoms were of the most gorgeous colours; the asclepiae, the gladiolus, the clematis, lilies, aloes, umbelliferae, arborescent ferns and fragrant shrubs, combining on either hand to make a border of surpassing beauty. Here and there the forest extended to the very [Illustration: Hercules could leave the boat without much fear of detection.] shore, and copal-trees, acacias with their stiff foliage, bauhinias clothed with lichen, fig-trees with their masses of pendant roots, and other trees of splendid growth rose to the height of a hundred feet, forming a shade which the rays of the sun utterly failed to penetrate. Occasionally a wreath of creepers would form an arch from shore to shore, and on the 27th, to Jack's great delight, a group of monkeys was seen crossing one of these natural bridges, holding on most carefully by their tails, lest the aerial pathway should snap beneath their weight. These monkeys, belonging to a smaller kind of chimpanzee, which are known in Central Africa by the name of , were hideous creatures with low foreheads, bright yellow faces, and long, upright ears; they herd in troops of about ten, bark like dogs, and are much dreaded by the natives on account of their alleged propensity to carry off young children; there is no telling what predatory designs they might have formed against Master Jack if they had spied him out, but Dick's artifice effectually screened him from their observation. sokos Twenty miles further on the canoe came to a sudden standstill. "What's the matter now, captain?" cried Hercules from the stern. "We have drifted on to a grass barrier, and there is no hope for it, we shall have to cut our way through," answered Dick. "All right, I dare say we shall manage it," promptly replied Hercules, leaving his rudder to come in front. The obstruction was formed by the interlacing of masses of the tough, glossy grass known by the name of , which, when compressed, affords a surface so compact and resisting that travellers have been known by means of it to cross rivers dry-footed. Splendid specimens of lotus plants had taken root amongst the vegetation. tikatika As it was nearly dark, Hercules could leave the boat without much fear of detection, and so effectually did he wield his hatchet that, in two hours after the stoppage, the barrier was hewn asunder, and the light craft resumed the channel. It must be owned that it was with a sense of reluctance that Benedict felt the boat was again beginning to move forward; the whole voyage appeared to him to be perfectly uninteresting and unnecessary; not a single insect had he observed since he left KazonndÃ©, and his most ardent wish was that he could return there and regain possession of his invaluable tin box. But an unlooked for gratification was in store for him. Hercules, who had been his pupil long enough to have an eye for the kind of creature Benedict was ever trying to secure, on coming back from his exertions on the grass-barrier, brought a horrible-looking animal, and submitted it to the sullen entomologist. "Is this of any use to you?" The amateur lifted it up carefully, and having almost poked it into his near-sighted eyes, uttered a cry of delight,- "Bravo, Hercules! you are making amends for your past mischief; it is splendid! it is unique!" "Is it really very curious?" said Mrs. Weldon. "Yes, indeed," answered the enraptured naturalist; "it is really unique; it belongs to neither of the ten orders; it can be classed neither with the coleoptera, neuroptera, nor to the hymenoptera: if it had eight legs I should know how to classify it; I should place it amongst the second section of the arachnida; but it is a hexapod, a genuine hexapod; a spider with six legs; a grand discovery; it must be entered on the catalogue as 'Hexapodes Benedictus.' " Once again mounted on his hobby, the worthy enthusiast continued to discourse with an unwonted vivacity to his indulgent ii* not over attentive audience. Meanwhile the canoe was steadily threading its way over the dark waters, the silence of the night broken only by the rattle of the scales of some crocodiles, or by the snorting of hippopotamuses in the neighbourhood. Once the travellers were startled by a loud noise, such as might [Illustration: It was caused by a troop of a hundred or more elephants.] proceed from some ponderous machinery in motion: it was caused by a troop of a hundred or more elephants that, after feasting through the day on the roots of the forest, had come to quench their thirst at the river-side. But no danger was to be apprehended; lighted by the pale moon that rose over the tall trees, the canoe throughout the night pursued in safety its solitary voyage. 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

DRIFTING DOWN THE STREAM

Check out and write about your own experience with Linux. You can also take inspiration from . HackerNoon’s Linux Writing Contest HackerNoon’s Linux Writing Prompt I am a Developer / Student / Freelancer / Artist / Gamer / Weeb. These words describe my daily schedule and my . I have used ubuntu for a year or so, And let me tell you it is great except for a few caveats. personality These 10 questions summarize my full byte-to-byte experience with this epic piece of 1’s and 0’s 1. Why do you prefer to use Linux over macOS or Windows? It is the same as using iPhone, it is reliable and quick, and easy in context for everybody but it does have a learning curve similar to apple os when you first buy an apple product. Will, you feel good if you are in the passenger seat of your most loved car,  definitely not right? You wanna roll that piece of iron around the road, and take over all of the controls. You want to be the admin and the all-mighty in here besides that sweet engine called the kernel. Linux gives you that power, it lets you take control and lets you roar with the engine. Windows makes you the passenger in your own beloved car, it takes control. For some it is good but for many, they wanna tune that thing up. I use Linux for just that sweet terminal control, which just lets me do anything. Besides that, it got an enthusiastic community, and you can rice your desktop like a real tuner 2. Which is the best Linux distro that you have used? I haven’t hoped around too many distros, Just use ubuntu cause it is easy to set up. But it dones’t really matter because with Linux you can do whatever the heck you wanna do… Still, I think is getting good, and even better than any other. But is also looking very promising as canonical wants to set a standard for the desktop environments. Pop_OS Ubuntu 22.04 3. What is the best thing that you love about Linux? Customizability and Control Linux doesn't limit you, this may be outrageous to some. But without getting deep in settings to just delete some folder for which I don’t have permission even though I know the password of the device. Is simply wonderful, to just use a terminal to manage everything. Try before you install And one more notable mention, There is that little option that lets you play around with the distro before you install. It is such a powerful feature that you can carry a pen drive around and just plug it into any PC and just start working from where you left off. Like yeah, people do it, you would often hear people using Kali Linux from just a pen drive. It is really helpful for a newcomer to test the best distro that fits their needs or whether the distro is having drivers correct. 4. What is your favorite customization for your Linux Desktop? Desktop customization, terminal customization. That body kit is wideeee I use this configuration for my desktop. I use Tweaks to save my customization settings, It is different for every distro. Desktop Appearance I use for discovering new Gnome shell themes and icons. Currently, I am using . And icons. Pling Tokyo Night Dark WhiteSur Dark Terminal I am using , which is an open-source collection for the Ubuntu terminal. I have installed all the themes to test from, It’s GitHub website also lets you see how the terminal will look with every theme. Gogh 5. What would be your word of advice for beginners in Linux? Google Everything, And take things easy. Just drift in second gear Wanting to try another distro, just get a pen drive and boot it with the distro you want to try, and just jump into it. Like seriously as I mentioned before, Every distro out there is made to be liked, not to be lost in. Don’t fear, Just google it whether it is just a small driver issue, or your device is not booting up. There is a solution for it if there isn’t, There is a community just waiting there to help you. I personally just search on google, and there is already a solution for it. The same thing works for me when I am working on software, I just google it and there is already a solution. If you are lost between choosing one distro from hundreds of Linux distros, then just choose Pop_OS or Ubuntu 22.04. I think made it less difficult but it’s still a difficult choice, so choose between whether you like Cyan or Purple. It’s easier now right? Lol And don’t worry it will run. It will drift even with a potato PC Linux can run on a 15-year-old device. Like seriously. 6. What are some of the myths about Linux that you would like to bust for our readers? If you are thinking that your drivers wouldn’t be compatible with Linux or that Linux is a just hacker’s thing, then you are wrong ( a little right with the drives one ). Linux and the OSS ( Open Source Community ) are working every ounce of their power to make it the best possible. So don’t worry, If your super exquisite GPU manufacturer dones’t provide Linux drivers then don’t sweat because there would be already an open source driver for it. 7. What are some of the tools and plugins that you recommend using with Linux for a fantastic experience? I use extensions or you may call plugins. These plugins are little applications running in the background which makes your experience better. I use for finding extensions. But some are not available there, so you can just search their names on google, and you can install them from their GitHub repos or websites. Gnome Shell Extensions It is the same for the tools, you can just search their names and you will find a guide explaining them in detail from installation to use. Here’s my list of favorite extensions and tools. Safe Eyes Clipboard Indicator Vitals Oh My Zsh TimeShift Wine 8. What are some disadvantages of Linux from your experience? Compatibility Some engine swaps aren’t possible Yes, it is a disadvantage, I really wanted to use Fusion 360. As I am getting into the 3d printer hobby. It is disappointing and sad when you suddenly get to know that the program you have been using for years suddenly doesn’t support the os you switched to. But that doesn’t mean there aren’t any solutions. There is and just straight-up better alternatives for the same program. Like I found for the Fusion 360, . Wine FreeCAD 9. What excites you about the future of Linux? What are your expectations for Linux in the upcoming years? It’s fun, It is just start of the touge I just get excited as I feel how Linux is affecting the world, how it is getting better and better, and showing how strong the OSS community is. I expect that there would be more polished distros, and more compatible software to run. As seeing gaming has taken off with the release of SteamOS. It feels great to experience and sees it grow. 10. Will you still use Linux if macOS or Windows were more privacy-oriented and even a bit open-source? You own the car because the car’s history reflects your personality Yup, I will still use Linux. As even if Microsoft did then their bit open source and bit our open source definitions don’t align up. And Apple and Microsoft aren’t gone change their habit of putting the user into a passenger seat. I haven’t used MacOS personally so I can’t be sure. But both represent cooperation, not a single open source team working and just living on the donations of supporters. That’s it, folks, I am not a good writer 😅. So let me know how it is. I hope I have convinced you to drift with Linux. Have a great day, and stay safe. Check out and write about your own experience with Linux. You can also take inspiration from . HackerNoon’s Linux Writing Contest HackerNoon’s Linux Writing Prompt

Linux is a Drift Car...That Drifts AF

Linux es un coche a la deriva... que a la deriva AF

Linux est une voiture à la dérive... qui dérive AF

Linux एक ड्रिफ्ट कार है...वह ड्रिफ्ट AF

Linuxはドリフトカーです...それはAFをドリフトします

Linux é um carro de drift...que faz drifts AF

Linux là một chiếc Drift Car ... That Drifts AF

Linux 是一辆漂移车……让 AF 漂移

Data drift occurs when a model sees production data that differs from its training data. If a model is asked to make a prediction based upon drifted data, the model is unlikely to achieve its reported performance. This phenomenon happens because during training, a model attempts to learn the most pertinent features to the train dataset. The most important features of the training dataset, however, are not universal to all data. For example, in computer vision tasks, models might identify patterns like straight lines in the sides of a building or circles for the wheels of a car. As useful as those features are for those specific objects, correctly interpreting images of trees or shoes will require a model to hone in on different features. Even in situations where the objects of interest remain the same, data drift can occur. Consider a model that is trained to detect buildings in overhead imagery. Suppose this model was trained on images taken over a city, during the day, in clear skies. If we pass images taken under different weather conditions, at night, or from a rural area, the data has drifted, and the model won’t perform as well. Monitoring drift and taking the appropriate steps to retrain models is essential to ensuring models generate trustworthy results. Trusting your model predictions requires knowing that your model can correctly interpret your data. To automate this process, Modzy developed a statistical method of detecting drift between your data and a model’s training data. The Challenge To train a model, data scientists split up a large dataset into train and test datasets. The data scientist then trains a model on the train dataset. Next, the model’s metrics are calculated based on the performance of the model on the test dataset. Since both the test and train datasets come from the same distribution (e.g. they have the same features: they are all cell phone images, taken during the day, in good weather conditions, etc.), the features learned by the model generalize well to the test data. This leads to metrics that only measure the performance of the model on the available data. It is expected that this model will not perform as well on the data points that largely differ from the train and test data (e.g. they have different features: the production images could be from satellites, or during poor weather conditions, etc.). Therefore, a large difference between production data and train data, i.e., data drift, causes a large decrease in prediction accuracy. There are a couple of different ways to detect drift. One way is by manually labeling the data that you ask the model to make predictions about, and then examining if the model made sufficiently accurate predictions. Not only is this impractical for large datasets, but also defeats the purpose of using a model in the first place. Other drift detection methods compare the prior distribution of model’s predictions with the distribution of predictions made by the model on new data. Modzy’s approach works differently. The Modzy approach centers around how the data is interpreted by the model during inference. The advantage of framing the problem in this way is that our drift detector is applicable in a more general setting. Modzy’s Approach to Drift Detection Modzy’s drift detection solution relies on the observation that a machine learning model will cluster similar samples from a familiar dataset close together in its feature space. Additionally, the features extracted for dissimilar data will be distributed differently in the feature space, indicating a level of drift. This observation led Modzy to the following formulation of drift in neural networks. Modzy compares the way that the training data passes through the network with how the user’s data passes through the network. In other words, we look for patterns in the features identified by the model given the user’s data and compare them with the patterns familiar to the model created during training. Any sizable difference between the two can indicate drift. To holistically sample drift throughout the network, we sample the feature space at multiple locations in the network, and extract the features of the train data samples at these locations. These features are then clustered according to their corresponding classes. The distances between the features learned from the train data and the clusters are calculated and then compared against the distances between the features identified for the production data and the same clusters. In this example, we compare two distance metrics— one estimating the distribution of distances for the training data in the model’s embedding feature space and one estimating the same distributions for the user’s production data. These distributions of distances are compared using statistical methods. As the differences between the pairs of distributions increases, the amount of drift between the two datasets increases, alerting the user to be more wary of the model’s predictions. This can be interpreted as a measure of reliability, i.e., the smaller the amount of drift, the more reliable the model’s predictions on the user’s dataset are. Comparisons of the distributions of distances calculated for the train set, its corresponding test set, and a separate out-of-distribution set at two different layers in an image classification network. Since the train and test sets are sampled from the same probability distribution, their distance distributions are very similar, whereas the out-of-distribution dataset’s distance distribution is obviously different from the others. What this Means for You A model is only as good as the data upon which it was trained, making it crucial to know that the data passing through a model when it is deployed to a production environment is similar to its training data. Suppose a significant amount of drift is detected in your dataset with respect to a given model. There are a few ways that you, as a data scientist working with this model, can respond. First, you could investigate the root cause of the drift. Was the data captured using a new sensor? Was the data formatted differently? Did the wrong data get sent to the model? If you can identify the cause and rectify the issue, then you can continue using the model with extra assurance that your data is familiar to the model; otherwise you should take the model out of production and use another one. In many cases, drift is inevitable. During training, it isn’t possible to account for every possible scenario your model might encounter in the real world, but that doesn’t mean you should give up on AI. Fortunately, when drift is detected and is found to be detrimental to the performance of a model, Modzy’s retraining solution can be utilized to custom tailor that model to your specific data to ensure the best results.

Approaching the Issue of Data Drift Detection

We recently released the first versions of , a new open-source project for infrastructure developers, DevOps, SRE, and cloud practitioners, with the goal of helping manage all kinds of drifts. driftctl (Stephane Jourdan – CTO and founder) Why? Because infrastructure is a living thing and changes are risky. In the team, we all managed cloud infrastructures at some level at our previous jobs, as developers with serverless function needs or DevOps managing large production systems. And we all knew that sometimes an emergency required a manual change, a customer tweaked a setting on the console, the boss activated something that had an impact later…or put more simply, that the whole infrastructure was just not totally under control by Terraform (or similar tool), just because ops life in production is rarely perfect and ideal. And all those tiny changes sometimes got forgotten about, or partially implemented in infrastructure-as-code later on, unfortunately leading to security issues, unstable environments, failed deployments, unexpectedly important bills. And thus this makes us lose the advantages of IaC. Keeping infrastructures secure & in sync with code. Like many of you, we use and love infrastructure-as-code, and especially Terraform, and for years we relied on its features to ensure the coherence of our infrastructures (most of us ended up with terraform plan in crontabs or in CI, expecting that return code 0 would ensure a fully synchronized infrastructure!). While Terraform is excellent at provisioning cloud infrastructures, telling you about changes in your AWS account is not where it shines, mainly because it’s simply not its job. Until now, to be sure that no change happened in your infrastructure, it implied an excellent level of trust in your tools and dependencies, a very stable IaaS provider, every single resource in every region carefully described in Terraform, then the revocation of all admin credentials of your team. We really think this is as bad for developers’ sense of ownership and empowerment as it is deeply against DevOps culture. But that was clearly not the reality we experienced daily out there! That was often either the goal to reach someday, or a naive belief from management when the team knew very well how to work their way around. But most of the time, credentials were passed around in the team, many resources were still not in Terraform; processes were not perfect. People sometimes forgot to push their locally applied changes. Changes happened outside of Terraform; emergencies had to be dealt with quickly (and usually not in Terraform!). Consultants came and went, and projects were still handed over to customers. No one had time or money to implement complex policies (not to mention that not everyone requires the hassle of PCI/DSS compliance). And yet, absolutely no one had a good overview of how good or bad things were, easily, so as to take action at some point. Hope is not a strategy and…that’s why we built driftctl. Driftctl is like your IaC security belt! Driftctl -  a new open-source project to help you monitor IaC changes. We built driftctl to help developers know when things change on their IaaS, whatever reason it does, and give them a good overview of their infrastructure-as-code coverage. This tool is open-source at its core (Apache-2.0 license) and we decided to release driftctl very early, so we can gather feedback sooner than later and prioritize accordingly. Also, by building this tool publicly from the beginning, we hope that many users will contribute ideas, suggestions, and maybe some code early on. The initial releases are focusing on detecting drift on AWS and on Terraform support. Future releases will add support for a lot more, but we didn’t want to wait to release this early. Those releases support what we thought were the most common services on AWS, and this was also backed with hours of interviews with DevOps practitioners around the world those last months. So we started with EC2, S3, IAM, RDS and Lambda, and we’re already working on supporting VPC, CloudFront, Aurora, DynamoDB, API Gateway, SNS/SQS, ECS/EKS/ECR, or KMS. The coming weeks will be exciting! We also included in the first release a view of your cloud / IaC coverage, so you can quickly know with simple data if your IaaS account is more or less under control with Terraform. Also, if that percentage decreases, it means something changed outside of Terraform! The human-readable output can be switched to JSON if you want to process it. Early on, we realized that for our own usage, this wasn’t enough: our own AWS accounts were filled with temporary tests, IAM roles, etc., and even our pretty much ideally locked down and mostly automated production account had a few manual resources (like the initial IAM user!). The output was deceptive as it included a lot of garbage we knew of but couldn’t do anything about. So we also added to the first release two different ways to ignore drifted resources, so you can start fresh with a clean output and start from there. This is a community project, and this is the first step in our mission to build an open-source suite of tools to protect codified infrastructures.eel free to give us a shout to prioritize another service or resource: , and we’d love to chat with you on . Come say Hi! open a GitHub issue or discussion our Discord Getting Started Let me show you how to quickly get started with driftctl, once downloaded from . We’ll start with the most simple case where you have the Terraform state locally available (beware, usually not a good practice!). https://github.com/cloudskiff/driftctl/releases First, export your AWS environment variables as usual: $ export AWS_PROFILE=”my-aws-profile” 
$ export AWS_REGION=”us-east-1”
$ driftctl scan 
Scanning AWS on region: us-east-1
Found unmanaged resources:
  aws_iam_access_key:
    - [...]
Found 18 resource(s)
 - - - - - AKIASBXWQ5ATWVVI7REW % coverage 22 covered by IaC 4 covered by IaC 14 not deleted on cloud provider 0 / drifted from IaC 0 4 Now, you probably follow good practices and don’t store your state locally or on git but on S3. Here’s how it would work: $ driftctl scan -- tfstate+s3: [...] from //acmecorp/states/terraform.tfstate Maybe, also as a good practice, you used tags extensively on your AWS resources, so you’d like driftctl to scan only for the key and value? Environment: Production $ driftctl scan --filter "Attr.Tags.Environment == 'Production'" Alternatively, you can exclude some resources directly in the file (if you want to ignore the “terraform” IAM user, simply add a line containing ). Obviously, at any time, you can check the help by executing driftctl scan help. .driftignore aws_iam_user.terraform This is only a quick introduction, a lot more details are included in the . Github documentation We really hope driftctl will find its place in your DevOps toolbox, and that you’ll join us on GitHub and on Discord to make it even better soon!

Introducing Driftctl: Your IaC Security Belt

When talking about infrastructure drift, you often get knowing glances and heated answers. Recording gaps in your infra between what you expected to be and the reality of what is, is a well known and widespread issue bothering hundreds of DevOps teams around the globe. Interesting to note though, is that depending on their context, the exact definition they will give of drift will vary. Facing impacts and consequences ranging from intensive toil to dangerous security threats, many DevOps teams are keenly aware of the issue and actively looking for solutions. We decided to look more closely into how they deal with it and conducted a study that will be released in the coming weeks. Here is a foretaste of this study, outlining some of the key facts we recorded. How infrastructure drift is becoming a major issue: a combination of growing volume and automation The ever-growing amount of workloads running in the cloud has gone hand in hand with a similarly growing number of people interacting with infras, themselves working with several environments. In this context, as Infrastructure as Code becomes widely adopted by users with heterogenous skillsets, the IaC codebases become larger, mechanically fueling daily occasions to create drift and making it harder to track. The various definitions and core causes of infrastructure drift Drift is a multi-faceted problem, let’s agree on a definition. Indeed, depending on their context and process (or lack thereof) people will tend to stress one specific aspect of drift or another when prompted on a definition. Typically, some people will tend to relate drift to the lack of consistency between, say and environments, which might be true, but incomplete. staging production Still, most of the people we talked to define infrastructure drift as an recorded between their Infrastructure code base and the actual state of their infras. “unwanted delta” Drift can be driven by human input, poor configuration, applications making unwanted changes, etc. It has consequences on toil and efficiency, forces teams to put in place strict controls that decrease flexibility, and can have a security impact. Two of the most common causes of drift are linked with process or workflow issues, like manual changes in a cloud console not being transposed as code, or changes applied to some environment but not propagated to others. Obviously those issues becomes more and more complex as the number of environments and team members grows. Some teams have dozens of environments that they need to keep updated. Much more surprising though is the fact that what we call API-driven drift is a widespread issue. API-driven drift happens when an update from an application or a script calling a cloud provider API (or a Terraform provider), autonomously and directly impacts your actual . In this case, you’ve just lost consistency with your codebase, and you won’t be aware of it until some random reveals a problem that might end up taking a couple of hours to fix. This happens even on the cloud provider’s side. One of the people we talked to told us of the time he suddenly got locked out of Azure because of a default machine authentication parameter updated from to . state apply true false The impacts of infrastructure drift Maintaining a solid and reliable source of knowledge of how your cloud resources are organized certainly is important, but at the end of the day you might simply think of infrastructure drift as a mere annoyance not worth the effort. While drift mostly causes additional work, it also triggers severe security issues too. In one of our interviews, a DevOps lead analyzed the problem quite clearly: “ . Willingly or not, a developer can do a lot with IAM keys and an SDK. Especially a junior one. Being able to catch bad decisions quickly and reverse your situation back to normal is crucial. every drift event causes uncertainty, a resolution time, and a potential security issue” There are also more subtle effects of drift, especially on efficiency. To avoid excessive drift, some teams make significant adjustments to their workflows. In some teams, only the DevOps team lead is allowed access to production. In others where developers are not skilled at IaC, getting a small change to environments done goes through a painful and long ticketing system. In other words, drift causes issues, which leads to rigid/counterproductive processes, which leads to a decrease in speed and flexibility. How teams solve drift so far Whatever the maturity level of a team and the degree of automation of their cloud infrastructures, maintaining reliability and security while moving at the developer's speed implies tracking down inconsistencies, like infrastructure drift. The faster it is detected, the easier it is to remediate drift, which is why many DevOps we interviewed had a in a cron job. terraform plan Most teams will explain that to tackle drift, the first thing they did was restrict access to production to a few team members. That certainly contains the problem but does not solve staging environments drift, as only a smaller part of teams restrict access to staging environments. Following through with that, deploying a full GitOps workflow will prevent some of the drift, but not all of it (not the API-driven events for example) and is still pretty hard to do by the book in practice when there is an urgent issue to solve on the production environment and MTTR is key for the business. Why we think infrastructure drift should be addressed with a better solution We recently caught an excellent Twitter thread about the various impacts of fast and slow-moving broken stuff. Infrastructure drift might not be considered a fast-moving broken thing. This is maybe all the more dangerous as sometimes you just can’t get anyone to pay attention to slow-moving broken things because what harm could that iceberg do?”* Analyzing what solutions are deployed against drift in over 100 teams led us to discover how poorly the topic is addressed. Stay tuned if you’d like to see how we intend to bring an open-source solution to this issue. *borrowed from @yvonnezlam Lead image via Danial RiCaRoS on Unsplash. Previously published at https://driftctl.com/2020/11/24/infrastructure-drift/

DRIFT

drift.com

-- employees

Since n.d.

COMPANY RANKING

DRIFT

EVERGREEN INDEX #4095

Representing this company? Click here to claim and customize this page!

Drift WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

DRIFT

drift.com

-- employees

Since n.d.

COMPANY RANKING

DRIFT

EVERGREEN INDEX #4095

RELATED COMPANIESRANK

STORIES ABOUT DRIFT

Representing this company? Click here to claim and customize this page!

Drift WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!