tweet_id
stringlengths
10
19
full_text
stringlengths
16
359
expanded_url
stringlengths
43
52
embeddings
listlengths
1.54k
1.54k
1689678041214066712
Using git, add and commit all modified files in one command: git commit -a -m 'commit message'
https://twitter.com/i/web/status/1689678041214066712
[ 0.04088904336094856, 0.017970282584428787, 0.009182291105389595, 0.01666908711194992, -0.015939630568027496, -0.039804715663194656, -0.006762267090380192, 0.013228808529675007, 0.002597459824755788, -0.016028348356485367, 0.04250568151473999, 0.01866030879318714, -0.029986614361405373, -0....
1689702491426848768
You need to unlearn the way you think of training data to get the most out of LLMs. With tabular data/models, you need to get a relatively large (usually >1000 rows) dataset to train an ML model, since it has to learn the concept from scratch. However, LLMs already know how to…
https://twitter.com/i/web/status/1689702491426848768
[ -0.013307303190231323, 0.003027531784027815, 0.0019416509894654155, -0.009511232376098633, 0.0376599095761776, 0.007291343063116074, 0.023221604526042938, -0.01325917523354292, -0.03135518357157707, 0.07700428366661072, -0.048681147396564484, -0.022355306893587112, 0.04389244318008423, 0.0...
1689267196759834624
Here's the code: 3/4 https://t.co/SLwcwAINp6
https://twitter.com/i/web/status/1689267196759834624
[ -0.014508859254419804, -0.005442762281745672, 0.054249171167612076, -0.03221432492136955, 0.03739716857671738, 0.003281950717791915, -0.025510765612125397, 0.0030414294451475143, 0.02406763844192028, 0.005803544074296951, 0.01137432549148798, -0.029265999794006348, 0.0005620243609882891, 0...
1689267192095703040
GZIP vs Bag of Words for text classification, more experiments. To summarize: A Bag-of-Words distance in KNN is more effective than GZIP KNN. Moreover, Bag-of-Words in a linear classifier can achieve almost the same accuracy as BERT. #nlproc #machinelearning 1/4 https://t.co/4GurWk4sBY
https://twitter.com/i/web/status/1689267192095703040
[ -0.04232686758041382, 0.023798704147338867, 0.03206440061330795, 0.020153440535068512, 0.00281085935421288, -0.0013676999369636178, 0.00001501023507444188, 0.01191096194088459, -0.005888728424906731, 0.034455880522727966, -0.01518473494797945, 0.02338077686727047, -0.03141428902745247, 0.0...
1689224057223254016
👀 Not enough people know it, but Hugging Face PRO accounts have access to Llama 2 (70b!) programmatically via the Inference API! Not bad for $9/month 🔥 https://t.co/rU2bXZ5Aut
https://twitter.com/i/web/status/1689224057223254016
[ -0.023350829258561134, -0.04100196808576584, 0.01781526580452919, 0.03518291190266609, 0.044135306030511856, 0.0018865307793021202, -0.021097810938954353, 0.05690738931298256, 0.01781526580452919, -0.040733397006988525, -0.0028386551421135664, -0.004692547023296356, 0.012921290472149849, -...
1689363385693896704
Train LLM with a simple English prompt 🤯 Meet gpt-llm-trainer, the easiest way to train a task-specific LLM. Simply describe your task. The AI will generate a dataset from scratch, parse it into the right format, and fine-tune a LLaMA 2 model for you. https://t.co/FvAZhTsjOw
https://twitter.com/i/web/status/1689363385693896704
[ -0.02371973544359207, 0.020792845636606216, 0.022766178473830223, -0.03838067874312401, 0.019534679129719734, -0.03708278015255928, 0.018355976790189743, 0.052074819803237915, 0.018567878752946854, 0.045876696705818176, -0.016872664913535118, -0.013839824125170708, 0.012720718048512936, -0...
1689293981668999169
Getting good results by filtering some public datasets. You'll find lots of duplicates. Filter by instruction similarity score > .95 (cosine) using e5-large-v2. After filtering sort the dataset by instruction length ascending order, this gave best loss + benchmark scores https://t.co/yW7k2CkGyv
https://twitter.com/i/web/status/1689293981668999169
[ 0.008709435351192951, -0.0020905316341668367, 0.06288959830999374, -0.03630712255835533, 0.02464556321501732, -0.001483576139435172, -0.0022575068287551403, 0.03505146875977516, -0.040341246873140335, 0.04787517711520195, 0.02272200770676136, -0.05273749679327011, -0.04143660515546799, -0....
1689175705060290560
The open-source space for text embedding models is 🔥 In the last few months: - Microsoft open-sourced E5 - Alibaba open-sourced General Text Embeddings - BAAI open-sourced FlagEmbedding - Jina open-sourced Jina Embeddings Check all of them at https://t.co/l7YojFx6zT
https://twitter.com/i/web/status/1689175705060290560
[ 0.001082787406630814, -0.004594812635332346, 0.02829555794596672, 0.01268155500292778, 0.01996123045682907, -0.010147816501557827, -0.0030835727229714394, 0.010212124325335026, -0.01996123045682907, -0.03524083271622658, 0.012353583239018917, -0.0287585761398077, -0.026752164587378502, -0....
1688913629306028032
supervision-0.13.0 is out! We added ByteTrack support! Now you can easily plug in any object detector and use it for tracking. GitHub repository: https://t.co/xXMRaS4ejS https://t.co/NvInCeyFpH
https://twitter.com/i/web/status/1688913629306028032
[ 0.009680374525487423, -0.012513797730207443, 0.005155661143362522, -0.0258922278881073, -0.0429278165102005, -0.01901022344827652, 0.08571542054414749, 0.006876162253320217, -0.014476746320724487, 0.004621108062565327, 0.020540855824947357, -0.013635482639074326, -0.016112536191940308, -0....
1688958387973734400
Built a notebook that makes it dumb simple to fine-tune LLaMA 2. Just load in a dataset, and run it! https://t.co/KgMM5Lmgv7
https://twitter.com/i/web/status/1688958387973734400
[ -0.03207424283027649, -0.011215835809707642, 0.009287321008741856, -0.07934460043907166, -0.008265063166618347, -0.03146523982286453, -0.013811355456709862, 0.0063220481388270855, -0.018009137362241745, 0.017878636717796326, 0.0029761476907879114, -0.016211124137043953, -0.025331692770123482...
1688974632928653312
Excited to share https://t.co/2FdhSpUlgD 🚀 - @Google's new browser-based code environment. It has AI assistance for code-generation, code-completion and explaining code built-in. Also supports modern JavaScript frameworks out of the box. Join the waitlist today 🙏 https://t.co/BHkVMGzbJe
https://twitter.com/i/web/status/1688974632928653312
[ -0.02858818508684635, 0.004449113737791777, 0.022239381447434425, 0.02034587785601616, -0.021917609497904778, -0.04074125736951828, -0.021323569118976593, 0.027622871100902557, 0.006825274787843227, 0.02683081664144993, 0.010321448557078838, -0.014380724169313908, -0.003981926012784243, -0...
1688950454216994816
Cycling illustration. Federico Bahamontes famously stopped at top of Col de Romeyere in 1954 Tour de France. He had broken two spokes and didn't want to start descending with battered wheel. He had ice-cream while waiting for team car. (📷 https://t.co/kywHOtjcxN) https://t.co/xcbyVi611x
https://twitter.com/i/web/status/1688950454216994816
[ 0.03891263157129288, -0.010307982563972473, -0.019041134044528008, -0.005089566111564636, -0.04755988344550133, 0.015963055193424225, 0.004842604044824839, 0.011131188832223415, 0.004237725865095854, -0.04581325501203537, 0.04143236204981804, -0.009484775364398956, -0.026170821860432625, -...
1688962643896242187
One of my favorite qualities of DeBERTa-v3 is how easy it is to extend positional embeddings (2048+). Even though it's pretrained with a 512 context length, you can easily get it to perform well on much longer contexts. Here I increased it to 2048 using the… https://t.co/rBzIJoOaHJ
https://twitter.com/i/web/status/1688962643896242187
[ -0.0015844163717702031, -0.06320638209581375, 0.028143184259533882, 0.018703093752264977, 0.002453674329444766, -0.0182399433106184, 0.018512384966015816, 0.07410403341054916, -0.03296539559960365, 0.04211942479014397, 0.020065300166606903, 0.0005631872336380184, -0.012743441388010979, -0....
1688857235198296064
...and we've managed to reduce the total of the entire visual without losing clarity. As always, a combination of small changes can add up to turn 'okay' visuals into truly #PowerfulCharts! 🚀What would you change? #dataviz #datavisualization #informationdesign (6/6) https://t.co/R2TsfBZ4Y4
https://twitter.com/i/web/status/1688857235198296064
[ 0.00286466465331614, 0.02048908732831478, 0.02922496758401394, 0.0036423448473215103, 0.01417425274848938, -0.011128189973533154, 0.0005608097417280078, 0.002615016885101795, -0.008922667242586613, 0.025503598153591156, 0.019540784880518913, 0.003857868257910013, -0.037845902144908905, 0.0...
1688551056614236160
LLM sizes, and when to use them: 100M-500M param, encoder-only: you have a straightforward classification/regression task, or you need local embeddings for reasonably sized datasets. 1-3B: you want to fine tune a resource-constrained model, usually a very specific task. (cont)
https://twitter.com/i/web/status/1688551056614236160
[ 0.04361845925450325, 0.0038204325828701258, 0.06596210598945618, -0.039394035935401917, -0.0003803334548138082, 0.0016573233297094703, 0.027815038338303566, -0.0074945371598005295, -0.023005302995443344, 0.03789258375763893, 0.004348485730588436, -0.00904688611626625, -0.034329816699028015, ...
1688678363060121600
Breaking 🚨 OpenAI just launched GPTBot, a web crawler designed to automatically scrape data from the entire internet. This data will be used to train future AI models like GPT-4 and GPT-5! GPTBot ensures that sources violating privacy and those behind paywalls are excluded. https://t.co/oR3kY4buaU
https://twitter.com/i/web/status/1688678363060121600
[ -0.015767186880111694, 0.03104049153625965, 0.041214462369680405, 0.003506563138216734, -0.007050167303532362, -0.061982207000255585, -0.03188009187579155, 0.04254794493317604, -0.00043060348252765834, -0.0029787265229970217, 0.017446385696530342, -0.06109321862459183, -0.04207875579595566, ...
1688741726880718848
Did you know SDXL can be implemented with 520 lines of code in single file? If you thought diffuser's unet code is now too big to understand in an hour, and wanted very limited but fully diffusers-compatible refactor of SDXL unet, this is for you https://t.co/PIrEtRW2be
https://twitter.com/i/web/status/1688741726880718848
[ 0.017021045088768005, 0.005037505179643631, 0.0015970810782164335, -0.01944020763039589, 0.0068337684497237206, 0.00964767299592495, -0.0002510151534806937, 0.005341710988432169, -0.03563554957509041, 0.012812863104045391, 0.004983182530850172, -0.029725264757871628, -0.01603599824011326, ...
1688820607020933121
You’re unable to view this Post because this account owner limits who can view their Posts. {learnmore}
https://twitter.com/i/web/status/1688820607020933121
[ 0.031293176114559174, 0.00044964285916648805, -0.018083997070789337, 0.03166009858250618, -0.023679552599787712, -0.035093437880277634, -0.03624662011861801, 0.0005192597163841128, -0.02261810004711151, -0.004425992723554373, 0.03920820355415344, 0.03787155821919441, -0.003475927747786045, ...
1688570892811456512
Finally releasing updated versions of my storysummarizer model, which can do long form narrative summarization and analysis LLaMA 8k 7b: https://t.co/77I4XZGjXB LLaMA 8k 13b: https://t.co/9zq7kLCSnO Also includes Flash Attention 2 inference code to make 8k actually usable 🥳
https://twitter.com/i/web/status/1688570892811456512
[ 0.009685053490102291, 0.01996091566979885, 0.018033752217888832, -0.048558901995420456, -0.034942153841257095, -0.0010435873409733176, 0.0009407232282683253, 0.008095494471490383, 0.006794307380914688, 0.0037664095871150494, -0.005412235856056213, -0.012547665275633335, -0.01748514361679554,...
1688662318522007552
I just finished writing a survey on the history of open-source LLM research, spanning from the early days (e.g., OPT and BLOOM) to recent models like LLaMA-2. Here are three takeaways that seem to have the biggest impact on LLM quality… Base models make all the difference.… https://t.co/YubNNqXSW2
https://twitter.com/i/web/status/1688662318522007552
[ 0.0038739212322980165, 0.023316210135817528, 0.044510118663311005, -0.014754334464669228, 0.0573892742395401, -0.04613818600773811, -0.05206898972392082, -0.01962399110198021, -0.03607907146215439, 0.01575733907520771, -0.0037685332354158163, -0.037474554032087326, -0.028316693380475044, 0...
1688564745236860928
Interestingly, DeBERTa-1.5B (and encoder-only model) beats Llama 2 on BoolQ, which is a nice example that encoders still outperform large decoders on classification task. For fairness: The DeBERTa-1.5B model was likely finetuned on the training data 1/3 https://t.co/jo46ggr9Qr
https://twitter.com/i/web/status/1688564745236860928
[ 0.044627703726291656, -0.03672623634338379, 0.015361079014837742, -0.0029727977234870195, 0.027837082743644714, 0.04106684401631355, -0.02436719462275505, 0.04766872897744179, -0.00872020609676838, 0.01072156522423029, 0.011585788801312447, -0.007108722813427448, -0.03446495905518532, -0.0...
1688431435894657024
💡Tip: disallow OpenAI’s web crawlers from accessing your website. https://t.co/Y6hzpCF3dZ
https://twitter.com/i/web/status/1688431435894657024
[ 0.004801201168447733, 0.01729775406420231, 0.013530658558011055, 0.03233927860856056, 0.00571779441088438, -0.03317193686962128, -0.05441809073090553, -0.004287506453692913, -0.009924720972776413, -0.019715143367648125, 0.014195440337061882, -0.051597803831100464, -0.001954055391252041, 0....
1688537406452973568
Hey @KBC_BE waarom krijg ik soms nog deze heel vervelende tussenstap? Ik heb nooit mijn kaart bij, en vind bovendien nooit die vermaledijde kaartlezer https://t.co/iepbpjwj00
https://twitter.com/i/web/status/1688537406452973568
[ -0.011604132130742073, -0.012560375034809113, 0.0011088859755545855, 0.027420900762081146, -0.03669903799891472, -0.014963903464376926, 0.007133052218705416, 0.0408082939684391, -0.04049816355109215, 0.03388199955224991, -0.0008015799685381353, -0.02403528429567814, -0.032719001173973083, ...
1679172761992806402
OpenAI Cookbook notebook: https://t.co/GvAylhQZdI
https://twitter.com/i/web/status/1679172761992806402
[ -0.04133747145533562, -0.05477849021553993, 0.033399663865566254, -0.05772029981017113, 0.029950646683573723, -0.04151499643921852, 0.003591668326407671, 0.034363359212875366, -0.024054352194070816, -0.015013364143669605, 0.020960381254553795, -0.05406840145587921, 0.04689140245318413, -0....
1688236146407481344
9 FREE #GenerativeAI Courses from Google👇 1. Introduction to Generative #AI: https://t.co/uPdyhIc9tj 2. Introduction to #LLMs Large Language Models: https://t.co/uSaUlrWpGW 3. Introduction to Responsible AI: https://t.co/U0HI2YT3h6 4. Introduction to Image Generation:… https://t.co/fgFjNAbplN
https://twitter.com/i/web/status/1688236146407481344
[ -0.02792654186487198, -0.008586679585278034, 0.05267961323261261, 0.003222293220460415, -0.013523867353796959, -0.09237240999937057, -0.03092913329601288, 0.015220453962683678, -0.007408833596855402, 0.017051301896572113, -0.0035274347756057978, -0.08690427243709564, -0.01674616150557995, ...
1688252452632678400
YYYY-MM-DD is obviously the best date format: it's the one where alphanumerical order matches chronological order.
https://twitter.com/i/web/status/1688252452632678400
[ -0.0007758856518194079, 0.017879676073789597, 0.015952734276652336, 0.021768592298030853, 0.07964688539505005, 0.005576449912041426, 0.0024276534095406532, -0.007555943913757801, -0.0625496655702591, 0.034357938915491104, -0.00596475787460804, -0.005053840111941099, -0.011345594190061092, ...
1688201830285611008
Start using pipelines. It's the simplest way to 10x your Machine Learning setup. The idea of pipelines has been around for a long time, yet many people ignore them or think they are only helpful in making your code more readable. They are much more than that. A pipeline is an… https://t.co/84ZWcZrAHn
https://twitter.com/i/web/status/1688201830285611008
[ -0.014604615978896618, 0.0036165607161819935, 0.01596318557858467, 0.014541719108819962, 0.021020080894231796, -0.021296825259923935, -0.006497229915112257, 0.008497345261275768, -0.030794229358434677, 0.05544973909854889, -0.008736353367567062, -0.043323252350091934, 0.006604154594242573, ...
1687947209705443328
Hey @WorldCoin - there are a number of vulnerabilities in your OpenId Connect implementation. DM me for details before publication.
https://twitter.com/i/web/status/1687947209705443328
[ 0.005762350745499134, -0.006640560459345579, 0.08188944309949875, 0.019997265189886093, -0.0030359418597072363, -0.028088310733437538, 0.005535599775612354, 0.03938985988497734, -0.01306516770273447, -0.009818671271204948, 0.032105039805173874, -0.03504200279712677, -0.02672060765326023, 0...
1688017355350380547
I just cleaned the clogged drain hoses of our dishwasher. A draining job. I more than ever respect the people who do that work for a living.
https://twitter.com/i/web/status/1688017355350380547
[ -0.012245571240782738, -0.005443596746772528, -0.06218947842717171, 0.010873744264245033, 0.0062640029937028885, 0.004075132310390472, 0.010712352581322193, 0.051618341356515884, -0.016704007983207703, -0.027732418850064278, 0.046184834092855453, 0.012353165075182915, -0.006321162451058626, ...
1687857495216177152
Popular Airline Passenger Routes https://t.co/mLOuNBTHOz
https://twitter.com/i/web/status/1687857495216177152
[ -0.0225764699280262, -0.033214107155799866, 0.0036856119986623526, -0.007424881681799889, -0.030692195519804955, -0.018485071137547493, -0.030531222000718117, 0.027030058205127716, 0.001862927689217031, 0.07592561841011047, -0.04069935157895088, -0.01640583574771881, 0.0274190753698349, -0...
1687881599793410048
Agree that this looks to be the most compelling LK-99 video so far. I found this to be an approachable/fun explainer of what's happening: https://t.co/TMEhnwLm2v https://t.co/N6nIGlkEjE
https://twitter.com/i/web/status/1687881599793410048
[ -0.022801106795668602, -0.009320670738816261, -0.057833168655633926, -0.025594979524612427, 0.016235506162047386, 0.01571553573012352, 0.005028971470892429, 0.02612271159887314, 0.01230856217443943, 0.031244812533259392, 0.029972048476338387, -0.01811360940337181, -0.0013241793494671583, -...
1687531009644888086
New blog post by @pdet: DuckDB ADBC - Zero-Copy data transfer via Arrow Database Connectivity https://t.co/sANk4wkq6r
https://twitter.com/i/web/status/1687531009644888086
[ 0.02494785375893116, -0.011893289163708687, 0.015970801934599876, 0.02330380119383335, 0.04947816953063011, -0.02032884769141674, 0.016388339921832085, 0.04856480658054352, -0.006811077240854502, -0.01193243358284235, 0.039170216768980026, -0.044441625475883484, -0.0038198139518499374, 0.0...
1687567376504557568
When was the last time that you did jumping jacks? If there’s not a medical condition prohibiting you, I’d suggest you get busy. They can help improve balance, agility, cardiovascular health and even bone density. Let me know how it goes. https://t.co/aOFlzcfbUp
https://twitter.com/i/web/status/1687567376504557568
[ -0.008424516767263412, 0.028523271903395653, 0.053084976971149445, 0.03679303824901581, -0.011339981108903885, 0.03867478296160698, 0.003942995797842741, -0.01889171451330185, 0.03298003226518631, -0.002076726872473955, 0.006468493491411209, 0.02312563918530941, -0.023992231115698814, 0.03...
1687269740874727424
You weirdos can't get enough high resolution #SAR it seems, so here's a new mystery city to melt your brains. Why do the boats on the left side look like lines in this image? That's the effect of rocking in the waves & spreading their radar return as the satellite passes over.… https://t.co/3Ep2UXrYIn
https://twitter.com/i/web/status/1687269740874727424
[ -0.000035956461942987517, 0.002865453250706196, 0.00629145884886384, 0.0557306706905365, -0.00503021664917469, -0.015326672233641148, 0.017686890438199043, -0.05788436904549599, -0.02876516804099083, 0.026301689445972443, -0.015828218311071396, -0.02965025044977665, -0.05460956692695618, -...
1687151082886316032
🚀Azure ChatGPT: Private ChatGPT for Enterprises 🤖 Exciting new release! Access the GitHub repository here & deploy directly to Azure or run locally: 🧑‍💻 https://t.co/qajoMopDEQ ✅ Private: Fully isolated from OpenAI ✅ Controlled: Isolate network traffic & leverage enterprise… https://t.co/lSUIdr3L4o
https://twitter.com/i/web/status/1687151082886316032
[ -0.023046603426337242, -0.0042118593119084835, 0.043480437248945236, 0.002178335329517722, 0.020976107567548752, -0.03339909017086029, -0.02286173775792122, 0.060833174735307693, -0.0026050673332065344, -0.03579001873731613, 0.033596280962228775, -0.02341633476316929, 0.02016269788146019, ...
1687449917785329664
The most detailed and practical write up on applying LLMs! 🙏 This reads like a survey paper but written for the industry and applications @eugeneyan is known as the best NLP writer for a reason. It’s the most comprehensive overview of patterns on building Large Language Models… https://t.co/8etXVExSAd
https://twitter.com/i/web/status/1687449917785329664
[ -0.014801464043557644, 0.030678465962409973, 0.009225307032465935, -0.01724703423678875, 0.00027168559608981013, -0.029193198308348656, -0.02348260022699833, 0.04471169039607048, -0.05254775658249855, 0.035569608211517334, 0.007240681443363428, -0.025979386642575264, -0.02024317905306816, ...
1687108654217187328
ML Recommender Systems: A cornerstone in powering platforms like Netflix, Amazon, YouTube, and Tinder. 🎬🛍️ Check out this rich dataset: https://t.co/3O6jPkgfMb Dive into 45,000+ movies and analyze 26 million ratings from over 270,000 users. https://t.co/01z3C6sBpg
https://twitter.com/i/web/status/1687108654217187328
[ -0.01722465641796589, 0.02053610235452652, -0.028930233791470528, 0.007238976191729307, 0.02749270759522915, -0.02769806794822216, -0.014169910922646523, 0.025066880509257317, -0.016929449513554573, 0.03203631937503815, 0.01597965508699417, -0.05406128987669945, -0.040815502405166626, -0.0...
1687117410984431616
The video for my North Bay Python talk is out, and I've put together an accompanying edited transcript with annotated slides and links https://t.co/0Dbkxz2BpI If you haven't been completely immersed in this world for the last year, my hope is this can help catch you up!
https://twitter.com/i/web/status/1687117410984431616
[ 0.0019518692279234529, 0.019731992855668068, -0.0059013147838413715, 0.006612316705286503, -0.032557111233472824, -0.011443743482232094, -0.011227056384086609, 0.05311521887779236, -0.0394098125398159, 0.00391728151589632, 0.009324280545115471, -0.004001924302428961, 0.0019298619590699673, ...
1687001492807634944
Distributed Llama 2 on CPUs w/ PySpark by Jonathan Apple Interesting showcase of pandasUDF - applyInPandas capabilities https://t.co/i5vUmIKxaB
https://twitter.com/i/web/status/1687001492807634944
[ -0.009643275290727615, -0.06287290155887604, 0.04721749573945999, 0.0011763538932427764, 0.023332329466938972, 0.007457047700881958, -0.009115565568208694, 0.046187207102775574, -0.006037255749106407, -0.003753652796149254, 0.019789131358265877, -0.026109090074896812, -0.027365542948246002, ...
1686819901795921921
Into the snow storm https://t.co/I7KAZN0zIb I took this picture while going up ski touring, in Stubai (the Austrian Alps) on a very cold and windy day. https://t.co/VLGY9EPMrh
https://twitter.com/i/web/status/1686819901795921921
[ -0.016278034076094627, -0.04271894320845604, -0.02737499587237835, -0.026478271931409836, -0.07836372405290604, -0.012703591957688332, -0.016128581017255783, 0.02428627945482731, -0.008892514742910862, -0.020313292741775513, -0.012510547414422035, 0.028994081541895866, -0.014272859320044518,...
1686742068159291393
Hallo! Bonjour! Guess what's quacking? 🦆 We're teaming up with @Datarootsio & @duckdblabs for our 1st MotherDuck User Meetup in Belgium 🇧🇪 on Sept 25! Expect "duck-tacular" topics & speakers 🗣️. Waddle over, let's chat about data. Register here 👇https://t.co/6sPh717Ijq
https://twitter.com/i/web/status/1686742068159291393
[ 0.0023424706887453794, 0.03242357075214386, 0.06072187423706055, -0.007910948246717453, -0.03202110528945923, -0.011885288171470165, 0.02022385597229004, 0.07209150493144989, -0.04849700629711151, -0.010803664103150368, -0.0028486959636211395, -0.012155693955719471, -0.027116063982248306, ...
1686753828471783424
Recently, I've been working on a new course for data analytics. This Python code stands out for its ability to turn raw sales data into insightful visualizations. I want to share why this code is crucial and how it could help transform your data analysis process. https://t.co/Z0vIa8Gmpj
https://twitter.com/i/web/status/1686753828471783424
[ -0.02860371582210064, 0.010742542333900928, 0.015038267709314823, 0.005406799726188183, 0.01030328031629324, -0.04382285475730896, -0.010613348335027695, 0.052608098834753036, 0.0003147103125229478, 0.034934256225824356, 0.06723294407129288, -0.031290967017412186, -0.05010171979665756, -0....
1686500942881882114
📚 Happy August! 🌟 Our DS/ML #bookclub is thrilled to announce our August read: 'Causal Inference in Python' by @MatheusFacure, presented by @OReillyMedia. 🚀 Join us for insightful book discussions and stay motivated throughout the journey! 🔗https://t.co/sdSI4wcfa2 https://t.co/4y3kTpGY1m
https://twitter.com/i/web/status/1686500942881882114
[ 0.003995863255113363, 0.040469247847795486, -0.012278521433472633, -0.0172540545463562, 0.025198280811309814, 0.010336995124816895, -0.013667870312929153, 0.04348544031381607, -0.015223466791212559, 0.01044386811554432, 0.0374768041074276, -0.04008925333619118, -0.01654156669974327, 0.0390...
1686269877794091008
Books I Keep Referencing Regularly As a Product Manager 1. Working Backwards https://t.co/EboYHJN0xP
https://twitter.com/i/web/status/1686269877794091008
[ -0.02516612596809864, 0.015365058556199074, -0.0055715302005410194, -0.011716045439243317, -0.004082521889358759, -0.0004035872407257557, -0.02263292670249939, 0.03661075606942177, -0.0248947124928236, -0.027819953858852386, 0.049940209835767746, -0.050965551286935806, -0.03534415736794472, ...
1686452034369986560
Run Llama 2 on your own Mac using LLM and Homebrew Details of my new llm-llama-cpp plugin and how to use it to run Llama 2 directly on your own Mac https://t.co/eXoNQyCfmz
https://twitter.com/i/web/status/1686452034369986560
[ -0.006549564190208912, -0.013852509669959545, 0.03783922642469406, -0.036429669708013535, -0.0028783450834453106, -0.0064827315509319305, 0.021532148122787476, 0.029454810544848442, -0.008597061969339848, 0.03528744727373123, -0.010638484731316566, 0.028944455087184906, -0.016477195546030998...
1686080535670489104
Nature is amazing! A camera recorded from start to finish how a bird built its nest and had its chicks. https://t.co/lXLuUnILqS
https://twitter.com/i/web/status/1686080535670489104
[ 0.036585669964551926, 0.014283792115747929, -0.00869886763393879, 0.03996436297893524, -0.0076272678561508656, -0.006694345735013485, 0.07201149314641953, 0.029702220112085342, -0.034316401928663254, 0.0007087528356350958, 0.03693866729736328, -0.0025623836554586887, -0.029349222779273987, ...
1686344250454085632
Quickly record what you're doing at your shell? $ script session.log Script started, output file is session.log ... $ exit All that happened is in the log file. #unix #tips
https://twitter.com/i/web/status/1686344250454085632
[ -0.01536781806498766, 0.025578230619430542, 0.016892071813344955, -0.0174453966319561, 0.009688407182693481, 0.0071775647811591625, 0.041447173804044724, 0.02499358542263508, 0.02186155691742897, 0.014856253750622272, 0.01048707403242588, 0.029754268005490303, 0.00041401496855542064, 0.000...
1686112044779507713
🐍 Build a Recommendation Engine With Collaborative Filtering — https://t.co/rfze1eSKOU #python https://t.co/57OSjhEzPv
https://twitter.com/i/web/status/1686112044779507713
[ -0.005608284845948219, -0.006097701843827963, -0.02397218719124794, -0.0014974920777603984, 0.03772510588169098, -0.011370474472641945, -0.004297017585486174, 0.055898163467645645, -0.04572814702987671, 0.04521102458238602, -0.021386587992310524, -0.057769645005464554, -0.04979122802615166, ...
1686099371669020672
Download free 592-page PDF comprehensive eBook on "Game Theory" here: https://t.co/0ZTO7u3zWz ———— #GameTheory #Gamification #Mathematics #Economics #ExperimentalEconomics #Strategy #Auctions #NashEquilibrium #InformationTheory #Statistics #Probability https://t.co/Ib3Annrm0u
https://twitter.com/i/web/status/1686099371669020672
[ -0.0326886810362339, -0.02736726962029934, -0.022920088842511177, -0.00986995454877615, 0.023680290207266808, -0.0153687484562397, 0.01345557440072298, 0.04046808183193207, -0.02398437075316906, -0.015166028402745724, 0.043635591864585876, -0.033195484429597855, 0.0009637143230065703, -0.0...
1685759122367455233
https://t.co/Unyw4jPNvx
https://twitter.com/i/web/status/1685759122367455233
[ -0.04654184356331825, 0.013254187069833279, 0.016367122530937195, 0.020835913717746735, 0.002078749006614089, -0.05965767800807953, -0.04441121220588684, 0.06840156763792038, 0.010362615808844566, -0.005509895738214254, -0.011552449315786362, -0.007471045013517141, -0.017100391909480095, -...
1685976315466272768
[Causal Inference: have you been doing science wrong all this time?] First blog post in my new incarnation as a data scientist 🎉 https://t.co/QiQ4thpIeB
https://twitter.com/i/web/status/1685976315466272768
[ 0.0393679179251194, -0.019308002665638924, -0.013239005580544472, 0.021818848326802254, 0.014313165098428726, 0.010231361724436283, -0.01120481826364994, 0.004608812276273966, -0.04353028163313866, 0.00956672616302967, 0.04675275832414627, -0.023376379162073135, 0.028357788920402527, 0.012...
1685735355503333376
What's up, Python? The GIL removed, a new compiler, optparse deprecated... https://t.co/zMqTquuJ22 https://t.co/DPHHJAIIfN
https://twitter.com/i/web/status/1685735355503333376
[ -0.00021657328761648387, -0.004150740336626768, 0.007934940047562122, -0.022969920188188553, -0.00226373178884387, 0.002538637723773718, -0.002273913472890854, 0.029812023043632507, -0.018367789685726166, 0.004048923496156931, 0.05234752222895622, -0.06076439842581749, -0.057071834802627563,...
1685325080488747009
The proper way to roll a burrito. https://t.co/t2dIESLcNO
https://twitter.com/i/web/status/1685325080488747009
[ 0.027949852868914604, -0.01134950015693903, -0.014891090802848339, 0.041459858417510986, -0.05439555272459984, -0.02142730914056301, 0.0285788606852293, -0.00676867738366127, 0.008293339051306248, 0.02257593348622322, -0.019403541460633278, 0.004372292198240757, 0.04211621731519699, 0.0081...
1685310211425239041
How to Extract Text from Any PDF and Image for Large Language Model by @zoumana_keita_ https://t.co/75uXOSjemJ
https://twitter.com/i/web/status/1685310211425239041
[ -0.006966531276702881, 0.009243943728506565, 0.022125031799077988, -0.017782840877771378, -0.0535835362970829, -0.013239206746220589, -0.011941025964915752, 0.0645061582326889, -0.07301148772239685, 0.011918643489480019, 0.030283426865935326, -0.03381985053420067, -0.043466679751873016, -0...
1684941628228902912
This 4 minute lower back drill will change your life. It's estimated that 80% of people in the western world will experience back problems. The causes of low back pain can be multifactorial but a good number of them are due to having a weak core and a weak set of glutes. Let's… https://t.co/tLiMsp7bzk
https://twitter.com/i/web/status/1684941628228902912
[ -0.02249876782298088, 0.011753290891647339, 0.06700601428747177, 0.04881088808178902, -0.026870502158999443, -0.013244583271443844, -0.014681399799883366, 0.03244071826338768, 0.04379905387759209, 0.00671081617474556, 0.004640711471438408, -0.021354762837290764, 0.0008694949792698026, 0.01...
1685007675396870144
Love a good #dataviz makeover? We have the perfect playlist for you! https://t.co/5O7H8PgTLS https://t.co/yYAllCnrk7
https://twitter.com/i/web/status/1685007675396870144
[ -0.030630718916654587, -0.015613625757396221, -0.026208586990833282, -0.041679564863443375, 0.000958020449616015, -0.041005220264196396, 0.01473179366439581, -0.0035954134073108435, -0.032809361815452576, -0.03786693513393402, 0.023057332262396812, -0.0188167542219162, -0.028063029050827026,...
1684934757878988800
My mind is slightly blown by tips from @bdon @Maxxen_ on using @DuckDB in the command-line for geospatial data processing: https://t.co/KEfOGfOJu7. Being able to pipe things in and out of DuckDB is next level. https://t.co/YdKmIgCYBR
https://twitter.com/i/web/status/1684934757878988800
[ -0.003902533557265997, 0.006263240706175566, 0.07324016839265823, 0.01203994918614626, 0.025327065959572792, -0.01158768031746149, 0.03396129235625267, 0.055697616189718246, -0.017460325732827187, -0.0033628942910581827, 0.024066194891929626, 0.00484133418649435, 0.0015169857069849968, -0....
1684855009123438593
How to Work With Time Series in Python? A look at why Python is a great language for time-series analysis. Plus, tips for getting started today. https://t.co/noNmT9sTI0
https://twitter.com/i/web/status/1684855009123438593
[ -0.013759140856564045, -0.0167702604085207, 0.02571997418999672, 0.005311279091984034, 0.03226497769355774, -0.033143218606710434, 0.01247314177453518, 0.024423521012067795, 0.007878049276769161, 0.0517745167016983, -0.001977353822439909, -0.033665981143713, -0.020732808858156204, -0.01594...
1684772237025710080
Read the chapters from this book “Managing #MachineLearning Projects” from @ManningBooks free online(!) at: https://t.co/dUUNmNNp6G ————— #BigData #DataScience #AI #DeepLearning #DataScientists #ML #NeuralNetworks https://t.co/XWbaVSJAWI
https://twitter.com/i/web/status/1684772237025710080
[ -0.018069935962557793, 0.02977590821683407, 0.038044415414333344, -0.038578614592552185, 0.018964141607284546, -0.033166926354169846, 0.016629913821816444, 0.022494513541460037, -0.05820470303297043, 0.006776225753128529, 0.01868542842566967, -0.04231802374124527, -0.06127054989337921, -0....
1684526828252192768
If you are still thinking which one is better (Random Forest or Xgboost): Random Forest is a Swiss knife of data science, albeit less powerful (but almost as powerful for most tasks unless one is going for that extra 0.001 on Kaggle) than boosted trees… https://t.co/4w8prde1xs
https://twitter.com/i/web/status/1684526828252192768
[ -0.025732150301337242, -0.015752151608467102, 0.03574853017926216, 0.019790230318903923, 0.07867586612701416, -0.012138493359088898, -0.0021069562062621117, 0.03502094745635986, 0.011798955500125885, 0.03596680611371994, 0.009294860064983368, 0.0031346657779067755, -0.065191350877285, -0.0...
1684633251531386880
The Problem With LangChain https://t.co/DiqEwlvhbv
https://twitter.com/i/web/status/1684633251531386880
[ 0.008497640490531921, -0.012152458541095257, 0.01289401575922966, 0.02286720462143421, 0.016132652759552002, 0.006106496788561344, -0.024713531136512756, 0.0015370287001132965, -0.03281012549996376, 0.026590123772621155, 0.035019662231206894, -0.02902667038142681, -0.06210920214653015, 0.0...
1684665379308908546
Thrilled to present my latest blog post: "Demystifying Document Question-Answering Chatbot - A Comprehensive Step-by-Step Tutorial with LangChain" 🔗: https://t.co/5mFOS0q2FJ It's the most exhaustive resource that you'll find on the topic going into depths of @LangChainAI. 1/
https://twitter.com/i/web/status/1684665379308908546
[ 0.0018244140082970262, 0.005818950943648815, -0.029074419289827347, 0.017546910792589188, 0.02819126285612583, -0.0061007472686469555, -0.010917432606220245, 0.04936376214027405, -0.031793609261512756, 0.035093821585178375, 0.03514030575752258, -0.016152454540133476, -0.01325895730406046, ...
1684456898085650432
Made a @Gradio demo for SD-XL https://t.co/0vdKtHegbZ https://t.co/9ImJn0jHzR
https://twitter.com/i/web/status/1684456898085650432
[ -0.013653083704411983, -0.030679453164339066, 0.005616084672510624, -0.03565214201807976, -0.013071483001112938, -0.02294415980577469, 0.0034405335318297148, 0.013275043107569218, -0.005223504267632961, -0.08875231444835663, 0.008447754196822643, -0.017200849950313568, -0.043881796300411224,...
1647421539279851521
For science I also added: - Choice of Embedding: simple tfidf bigrams or the OpenAI API embeddings ada-002 (ada should work better (?), tfidf is much much simpler) - Choice of Ranker: kNN (much faster/simpler) or SVM Default that seems to be both good & fast is ada+knn https://t.co/JTdj3XW2eK
https://twitter.com/i/web/status/1647421539279851521
[ -0.04651428014039993, -0.030749091878533363, 0.04330543428659439, -0.017857911065220833, 0.01921120658516884, -0.04891393706202507, 0.016044218093156815, 0.005521303508430719, -0.01967160589993, 0.017746299505233765, 0.0042447419837117195, -0.029967807233333588, -0.030051516368985176, -0.0...
1656002284860612608
RE: "how often do you see teams actually fine tuning LLMs?" It's an interesting question, about how prompting (optimization over prefix tokens) and finetuning (optimization over weights) will be used over time. If people have data points please pitch in. I expect that finetuning… https://t.co/ecKIwpJ7e8
https://twitter.com/i/web/status/1656002284860612608
[ 0.013004069216549397, 0.02213693968951702, 0.06731146574020386, -0.017949722707271576, 0.013517595827579498, 0.0036934399977326393, 0.038553982973098755, 0.048729706555604935, 0.008911658078432083, 0.07331576943397522, 0.02398563362658024, -0.029057694599032402, -0.05182666704058647, 0.018...
1679463907344146438
Good / slightly obscure tip is that applications can benefit from custom supervised finetuning of emebeddings returned by APIs. Collect a few examples of +ve (and optionally hard -ve) pairs, use them to train a linear projection that better discriminates your pairs. https://t.co/QeSFcrOgNU
https://twitter.com/i/web/status/1679463907344146438
[ -0.021790791302919388, -0.017164897173643112, 0.05696841701865196, 0.003714846447110176, 0.000012404900189721957, -0.03596596047282219, 0.030759969726204872, 0.012940606102347374, -0.007042962126433849, 0.052625130861997604, -0.00824777688831091, -0.029763394966721535, -0.007905669510364532,...
1684395685276176384
1. Why SVM? SVMs learn better from smaller datasets than XGBoost and can capture more complex features than linear/logistic regression (depending on your kernel) 2. But SVMs don’t scale? It’s not that bad, and *certainly* faster than training an LLM. I’ll release some code.
https://twitter.com/i/web/status/1684395685276176384
[ -0.0061942050233483315, 0.03274698555469513, 0.015668058767914772, -0.009071633219718933, 0.01824227347970009, -0.017648225650191307, 0.009752314537763596, -0.02498721331357956, 0.023700105026364326, 0.07123645395040512, 0.013304236344993114, -0.03064306080341339, -0.045816078782081604, 0....
1684285421415456768
You don't need to fine-tune an LLM for every problem. You'd be amazed how far you can get training an SVM on top of a 100-300M parameter encoder-only model. Don't waste your time training a 7B+ parameter model for every problem.
https://twitter.com/i/web/status/1684285421415456768
[ 0.016362890601158142, 0.014542213641107082, 0.02576778270304203, -0.021535001695156097, -0.003728326875716448, -0.007479846943169832, 0.05158195272088051, 0.007381275296211243, -0.008587327785789967, 0.0836351290345192, -0.02790156938135624, -0.04696648195385933, -0.016954319551587105, 0.0...
1684545431026401281
TIL there is a site module in the #Python Standard Library. Here is how to print sys.path from the command line: https://t.co/ccHdXWbjgV
https://twitter.com/i/web/status/1684545431026401281
[ -0.05394374579191208, -0.03351889178156853, 0.009025203064084053, 0.0018110920209437609, 0.014062258414924145, -0.01854604482650757, -0.007567108608782291, 0.02468963712453842, -0.019652582705020905, 0.03427963703870773, -0.002991111483424902, -0.019548844546079636, -0.018188724294304848, ...
1684265216815530005
Op papier zag het er nochtans veelbelovend uit. https://t.co/F8VmwKWHTh
https://twitter.com/i/web/status/1684265216815530005
[ -0.016082940623164177, 0.04519280046224594, -0.0288767758756876, 0.024111460894346237, -0.00811916496604681, 0.036309633404016495, 0.01754620298743248, 0.03136302903294563, -0.00021305524569470435, 0.014321845024824142, 0.013622586615383625, -0.029576033353805542, -0.01732606627047062, -0....
1684507056420663296
**Instruction-Tuned Llama 2: Comprehensive Guide & Code** 🚀 Dive into the incredible potential of instruction-tuning Llama 2 with this comprehensive step-by-step guide, complete with code examples. 📚💻 The extended guide covers the following key aspects: 📝✅ 1. Define the… https://t.co/CHYBhjfnfY
https://twitter.com/i/web/status/1684507056420663296
[ 0.006931401323527098, -0.00847100280225277, -0.02142665535211563, -0.04857730492949486, 0.027393409982323647, -0.0056856656447052956, 0.002374883508309722, -0.0050436328165233135, -0.03955690190196037, 0.006324504502117634, 0.01581764966249466, -0.0062925624661147594, -0.005823016166687012, ...
1684226326079610882
What are embeddings? https://t.co/qbkpMUl2up #AI #MachineLearning #DeepLearning #LLMs #DataScience https://t.co/MHfRHz57iW
https://twitter.com/i/web/status/1684226326079610882
[ -0.027684101834893227, -0.037940990179777145, 0.005888213869184256, 0.0054341331124305725, 0.021760273724794388, -0.042428381741046906, 0.019587809219956398, 0.007977579720318317, -0.02982095442712307, -0.006831990089267492, -0.01675054430961609, -0.029512297362089157, -0.011901077814400196,...
1684497503515811840
#Python's pathlib keeps amazing me 🐍😍 I wanted to get the relative path of a file in my project ... turns out you can use Path's relative_to() method 💡 https://t.co/bnVtciW0lX
https://twitter.com/i/web/status/1684497503515811840
[ -0.02534797228872776, 0.0070169768296182156, 0.02863932214677334, 0.018708128482103348, 0.06596415489912033, -0.003994191065430641, 0.022445183247327805, 0.015805339440703392, -0.02301659807562828, 0.021473778411746025, -0.011479727923870087, 0.010611177422106266, -0.03250208869576454, -0....
1684241306048925696
Join Poli and Andrew in learning how to build generative AI applications with Gradio, Transformers and Diffusers! 🤗 https://t.co/e8JU2ugq6k https://t.co/rt08XD6pRA
https://twitter.com/i/web/status/1684241306048925696
[ 0.0017492985352873802, -0.00847212690860033, 0.01681843027472496, 0.003507334738969803, -0.030337484553456306, -0.03682439401745796, -0.010541227646172047, 0.007682234048843384, -0.02812858112156391, -0.014959036372601986, -0.001049404381774366, -0.0260874405503273, -0.0013989146100357175, ...
1684472905147486210
This is handy to link these two Makefile commands together (which would not work because state would not carry over) #unix #commandline #tips https://t.co/vkPYSzsxSd
https://twitter.com/i/web/status/1684472905147486210
[ -0.022465018555521965, -0.029345348477363586, 0.04270365089178085, -0.00041514181066304445, -0.021727360785007477, -0.015316448174417019, -0.022478429600596428, 0.0032842515502125025, -0.005723549984395504, 0.030123243108391762, 0.012171345762908459, 0.045439690351486206, 0.02832604013383388...
1684105103442161664
Today, w/@huggingface, @eleutherai, @laionai,@github & @creativecommons, we publish a statement on “Supporting #OpenSource & #OpenScience in the #EU #AI Act”. Open practices help uphold European values, democratic control & scientific research practices 📃https://t.co/AoZ4U4JuxP
https://twitter.com/i/web/status/1684105103442161664
[ -0.006601326167583466, -0.033281419426202774, 0.05132802948355675, 0.006227485369890928, 0.03489181399345398, -0.020807279273867607, -0.003209599293768406, 0.02660979889333248, -0.03471288084983826, -0.028501370921730995, 0.05945667251944542, -0.058689817786216736, -0.020551661029458046, 0...
1684331479050997768
Still using using XGBoost (substitute your favourite classifier) and think it outputs probabilities? Read on. Probabilistic Prediction in scikit-learn. 'Adding confidence measures to predictive models should increase the trustworthiness, but only if the models are… https://t.co/TvOMgdX41V
https://twitter.com/i/web/status/1684331479050997768
[ 0.026645002886652946, -0.05411222577095032, 0.07009408622980118, 0.01243604812771082, 0.01388777606189251, 0.004037219099700451, 0.0037802758160978556, 0.024859247729182243, 0.017844701185822487, 0.042935196310281754, -0.029394295066595078, -0.025334592908620834, -0.04468240961432457, 0.04...
1684277595418812442
👉 Machine Learning Recommender Systems This repository provides a curated list of papers about Recommender Systems including:🔹 comprehensive surveys 🔹 social recommender system 🔹 deep learing-based recommender system 🔸and more. 🔗https://t.co/9p08KsLmhk https://t.co/tgVxeRrajz
https://twitter.com/i/web/status/1684277595418812442
[ -0.0034237978979945183, -0.01741836778819561, -0.014250402338802814, -0.011702965945005417, 0.028217757120728493, -0.01455522421747446, -0.03764544799923897, 0.014108878560364246, -0.005132975522428751, 0.03379163518548012, -0.03618666157126427, -0.02902335673570633, -0.02466876432299614, ...
1684093296593735680
Python package “TSFRESH extracts 100s of features from time series that describe basic characteristics…such as the number of peaks... or more complex features such as the time-reversal symmetry statistic.” https://t.co/DpEqJGEoFo https://t.co/nKIDbkku8A https://t.co/WILnurOfcp
https://twitter.com/i/web/status/1684093296593735680
[ -0.036064740270376205, -0.04655074700713158, 0.0053034513257443905, -0.026189567521214485, 0.016747070476412773, -0.015169079415500164, 0.005459341686218977, 0.02587142400443554, -0.02837839536368847, 0.0492231547832489, 0.028632909059524536, -0.050928402692079544, -0.012623932212591171, -...
1684135678286655488
The bible of #timeseries analysis. One might argue that machine learning has taken over. 

True, BUT until and unless someone created a Machine Learning Time Series Analysis 'bible' similar to what @sirbayes did for general machine learning Hamilton's book (800 pages!) is and… https://t.co/0WVBgbxAIz
https://twitter.com/i/web/status/1684135678286655488
[ -0.022380225360393524, 0.002205711556598544, 0.005641214549541473, -0.03877304866909981, 0.012185814790427685, 0.0034322061110287905, 0.02197139337658882, 0.024674957618117332, -0.0096207270398736, 0.01420359592884779, -0.000005473581495607505, -0.0342099629342556, -0.011440686881542206, 0...
1684068277331324928
"Getting started is often the most challenging part when building ML projects. How should you structure your repository? Which standards should you follow?" Read more from @KhuyenTran16's post. https://t.co/YtZwPTEhDO
https://twitter.com/i/web/status/1684068277331324928
[ -0.02201101742684841, -0.002591449534520507, 0.04292893409729004, 0.007378411013633013, 0.064592145383358, -0.032072484493255615, 0.044817015528678894, 0.017179027199745178, -0.05987195298075676, 0.054953012615442276, -0.02722807414829731, 0.004080484621226788, -0.025886544957756996, -0.01...
1683839667454771201
Upsampling! Simple strat. Works great. https://t.co/WZdp46f0hU https://t.co/EvLtHdgBDH
https://twitter.com/i/web/status/1683839667454771201
[ -0.037711530923843384, -0.022008545696735382, -0.01302235759794712, -0.011765814386308193, -0.009534498676657677, -0.00431793974712491, 0.020394079387187958, 0.010532117448747158, -0.045905716717243195, -0.012298893183469772, 0.0181399155408144, -0.024399787187576294, -0.022435009479522705, ...
1684135545688162305
Cool people to follow in HF 🤗 AK https://t.co/2UC41luJnQ Karpathy https://t.co/UZGtvdd4mq Matei Zaharia https://t.co/skdEjXlAzF Richard Socher https://t.co/AnZwvbpUne TheBloke https://t.co/6H0kVqGe2n Aran Komatsuzaki https://t.co/XUJpjibdJq TheJulien https://t.co/W3LKQRhYDN
https://twitter.com/i/web/status/1684135545688162305
[ -0.015427103266119957, -0.03620553016662598, -0.05828356742858887, 0.00035046396078541875, 0.01574818231165409, -0.03216910362243652, -0.058925725519657135, 0.052198346704244614, 0.027643412351608276, -0.039691537618637085, 0.011069596745073795, -0.0028820703737437725, -0.01412749569863081, ...
1684111218355707905
@sauhaarda @arxiv Many thanks! Instead of summaries can we extract key phrases and build knowledge graph embeddings ? https://t.co/hKZ54zaida
https://twitter.com/i/web/status/1684111218355707905
[ 0.033256515860557556, -0.002179386094212532, 0.006725683342665434, -0.01067158579826355, 0.02063719555735588, -0.014207769185304642, -0.0353996604681015, 0.057940468192100525, -0.012827333994209766, 0.009108352474868298, -0.01780068315565586, -0.06394126266241074, -0.002095866482704878, -0...
1684119850942189568
A short statistics on an "average" cyclist that came to our lab in the last few months. Statistics of just above 50 test. Data only includes males. @HumanCentre For all cyclist we used the same protocol. Lactate profile with 4min stages and separate VO2max test. https://t.co/6171iCsYas
https://twitter.com/i/web/status/1684119850942189568
[ 0.018297284841537476, 0.00000469876385977841, 0.004213190637528896, -0.03608473762869835, 0.04007841646671295, 0.014268199913203716, -0.0201666671782732, 0.025349952280521393, 0.03622635826468468, -0.0036254681181162596, -0.004103435203433037, -0.007413798943161964, -0.025548221543431282, ...
1684076355292643328
Download 424-page PDF eBook >> “Advanced #DataScience and #Analytics with #Python” at https://t.co/REX4XiEMNK ———— #BigData #MachineLearning #AI #Coding #DataScientists #DataAnalytics https://t.co/0ByhrMZmY2
https://twitter.com/i/web/status/1684076355292643328
[ -0.006524583324790001, -0.032320596277713776, 0.005452726501971483, -0.0035426232498139143, 0.006172794383019209, -0.02849489450454712, 0.024361375719308853, 0.052284613251686096, 0.004842592868953943, 0.008442931808531284, 0.036190275102853775, -0.03412351384758949, -0.05325203388929367, ...
1683871228740640769
awesome-python-htmx: Curated List of Python / HTMX https://t.co/dgA7uoFYN7
https://twitter.com/i/web/status/1683871228740640769
[ -0.05327921733260155, -0.05885845422744751, -0.000492817023769021, -0.0035749850794672966, -0.0014348633121699095, -0.03980862349271774, -0.002704799873754382, 0.04098981246352196, 0.02362380363047123, 0.016875941306352615, -0.017956603318452835, -0.03362622112035751, -0.005843121558427811, ...
1683716180358447105
Renewed emphasis on cornering strategies in cycling time trials! 🚴‍♂️🔄 This was written some time ago, but still relevant IMO 👉 https://t.co/I3hDq58mKd #Cycling #TimeTrials #theartofdescending 🚴‍♂️ https://t.co/8pO73E9jTi https://t.co/SxUlbfH6OS
https://twitter.com/i/web/status/1683716180358447105
[ 0.0026675642002373934, 0.01653052493929863, 0.033791683614254, 0.015784673392772675, 0.011621599085628986, 0.029788432642817497, -0.020092351362109184, 0.039758503437042236, -0.003784439992159605, 0.015000767074525356, 0.026805022731423378, -0.03190422058105469, -0.0007867601816542447, 0.0...
1683718922434912257
👉 Machine Learning Recommender System This repository provides a list of papers about Recommender Systems including: 🔹 general recommender system 🔹 social recommender system 🔹 deep learing-based recommender system 🔸 and more. 🔗https://t.co/9p08KsLmhk https://t.co/ZuMWt2nvfW
https://twitter.com/i/web/status/1683718922434912257
[ -0.010884703136980534, -0.01956518553197384, 0.008183987811207771, -0.004070169758051634, 0.034940171986818314, -0.01080831978470087, -0.014207402244210243, 0.025643160566687584, -0.01678263023495674, 0.022325918078422546, -0.04482642933726311, -0.024159131571650505, -0.03264865651726723, ...
1683744373756817408
🚨 Quick tip for #PowerfulCharts: if one of the categories in your stacked area chart is much smaller than the other, place it at the bottom. Otherwise it will just follow the bumps and wiggles of the bigger categories and its evolution will be hard to spot! #dataviz https://t.co/2I6Qw3OCac
https://twitter.com/i/web/status/1683744373756817408
[ -0.02967987395823002, -0.008507960475981236, 0.03546447679400444, 0.012731530703604221, -0.016259057447314262, -0.009372947737574577, -0.06195471063256264, -0.015015637502074242, -0.029355503618717194, 0.022124752402305603, 0.0111299529671669, 0.018043093383312225, -0.020894847810268402, 0...
1683597683661963266
Fascinating reading on #GameTheory: https://t.co/fnvZT6gLvw "A Nontechnical Intro to the Analysis of Strategy" (3rd Ed.), covers N-person strategies, Nash Equilibria, dominant strategies, auctions, bargaining, #BehavioralEconomics, #ExperimentalEconomics, etc. ——— #Gamification https://t.co/oqRmgnnS62
https://twitter.com/i/web/status/1683597683661963266
[ -0.05737577751278877, -0.0010081510990858078, 0.0030317881610244513, -0.016299936920404434, 0.04360559210181236, 0.011233916506171227, -0.007413211278617382, 0.05247275531291962, -0.019742483273148537, 0.009525682777166367, 0.054246190935373306, -0.043266553431749344, -0.016000017523765564, ...
1683517968418783232
Transforming a busy multi-lane road into a safer and greener road with less car lanes and more space for cycling isn't impossible or rocket science. It about making the choice and then act to it! 👇🏼 This is how it's done in #Utrecht. 🎞 utrechtonderweg https://t.co/OuoBlyF2jK
https://twitter.com/i/web/status/1683517968418783232
[ 0.0053984662517905235, -0.010541463270783424, 0.011785194277763367, 0.025224216282367706, 0.009734719060361385, -0.0290428064763546, -0.04036412388086319, 0.01786939427256584, 0.008652335964143276, 0.04937277361750603, 0.027469655498862267, -0.04197761416435242, -0.041816264390945435, -0.0...
1683466479176384515
Playfair's Wheat and Wages. A design reproduction exercise with @observable Plot. https://t.co/oXKvFqj8zz
https://twitter.com/i/web/status/1683466479176384515
[ 0.0011865956475958228, -0.0005016481154598296, 0.014369857497513294, -0.03868996351957321, -0.002714260946959257, -0.01654290035367012, -0.021632395684719086, -0.003143151057884097, -0.0162651427090168, -0.04153289273381233, 0.06669443845748901, -0.020178252831101418, -0.038036417216062546, ...
1683450687961563137
New blog post! Our writer @lisacmuth takes you on a tour of all the fascinating ways to use color (and lots of gray!) to lead the reader’s eye in your #dataviz. Read her seven principles here: https://t.co/dbExUii1Kq
https://twitter.com/i/web/status/1683450687961563137
[ -0.020979590713977814, 0.012457454577088356, 0.010746446438133717, 0.007331011351197958, 0.011575628072023392, -0.039379507303237915, 0.01039766427129507, -0.013286635279655457, -0.061490997672080994, 0.01075302716344595, 0.020782167091965675, -0.015965022146701813, 0.015267456881701946, 0...
1683114005739585536
Waouh ! Juste hallucinant 🙊🙈 ! Ça se passe quelque part en Chine... https://t.co/PypU4ZyZUI
https://twitter.com/i/web/status/1683114005739585536
[ -0.0018552631372585893, 0.0027066092006862164, -0.029342452064156532, 0.013096692971885204, 0.010472471825778484, 0.0006247781566344202, 0.014195204712450504, 0.0319056436419487, -0.0372273214161396, -0.0581478513777256, 0.02795100398361683, -0.041670188307762146, -0.025070464238524437, -0...
1682972954324262914
18. Charge your phone outside of reach 19. Practice 4-7-8 breathing 20. Give your mind an overnight task https://t.co/I7NEA9u1zw
https://twitter.com/i/web/status/1682972954324262914
[ 0.003410259960219264, -0.00461813947185874, -0.019478490576148033, 0.020667318254709244, -0.04819325730204582, -0.017680006101727486, -0.02199331857264042, 0.022831594571471214, 0.015500488691031933, 0.03237270191311836, 0.029217734932899475, 0.012147383764386177, 0.019676627591252327, 0.0...
1683010844534382592
🚀 Introducing Icon Buddy — 100K+ Open Source SVG Icons, Fully Customizable! → https://t.co/G9u2f3T0ht It's live on PH and link in next tweet. 📺 https://t.co/fkwAhDMjjm
https://twitter.com/i/web/status/1683010844534382592
[ -0.012195154093205929, -0.021163873374462128, -0.06017556041479111, 0.010549928061664104, -0.006664123386144638, -0.04906228929758072, 0.024415913969278336, 0.04519568756222725, 0.008475791662931442, 0.0012307182187214494, -0.014672594144940376, -0.0034152832813560963, 0.006094376090914011, ...
1682806727291334661
GUYS DID YOU KNOW THE RED WEDDING OF #GoT HAPPENED IN FRANCE IN 1572? The Catholic queen mother forced her daughter to marry a protestant prince, invited all the protestant nobilities to the wedding at the predominantly catholic Paris, then bodied them all. Aka St… https://t.co/QM7SNpq4LJ
https://twitter.com/i/web/status/1682806727291334661
[ 0.010467196814715862, 0.034954674541950226, -0.019854063168168068, 0.034378498792648315, -0.07562309503555298, 0.012939952313899994, -0.0050745499320328236, -0.03397037461400032, -0.015940869227051735, -0.010815302841365337, 0.027536408975720406, 0.009825000539422035, -0.01686515100300312, ...
1682850564835532802
An Easy Introduction to #MachineLearning Recommender Systems #KDnuggets https://t.co/ki4QH3UgTz
https://twitter.com/i/web/status/1682850564835532802
[ -0.049007829278707504, -0.028192540630698204, -0.02727348916232586, 0.01012820191681385, 0.05489472672343254, -0.01840587891638279, -0.007731214631348848, 0.05390115827322006, -0.014593055471777916, 0.06483042240142822, -0.022218704223632812, -0.028465773910284042, -0.02142384834587574, -0...
1682487356836700160
🧵 Difference Between Confidence Interval & Credible Interval 1/ Intro
Both Confidence Intervals (CIs) and Credible Intervals (CrIs) provide a range for estimating an unknown parameter. But they're based on different philosophies and interpretations. #DataScience #Stats https://t.co/JLRtbw0Slk
https://twitter.com/i/web/status/1682487356836700160
[ -0.0017142976867035031, -0.02074243128299713, -0.004573358688503504, 0.028134984895586967, 0.012279155664145947, -0.009562477469444275, 0.008799302391707897, 0.013144847005605698, 0.024854468181729317, 0.03813600540161133, 0.026585852727293968, -0.003710514632984996, -0.03394423425197601, ...