Loading video...

Video Failed to Load

Go Home

Real Training Still Existsโ€ผ๏ธ ๐๐„๐‹๐ˆ๐„๐•๐„ ๐“๐‡๐„ ๐‡๐˜๐๐„ BIRMINGHAMโ€™S PREMIER SPORTS PERFORMANCE CENTER โŒCome see us TodayโŒ XFP_Bham #XFP #XFPTrained

10,796 views โ€ข 3 years ago โ€ขvia X (Twitter)

1 Comments

_'s profile picture
_2 years ago

@XFP_Bham Hey bro I need to get my son connected with you ASAP!!!!

Related Videos

A team tested Pi0, Pi0 Fast, Gr00t, and ACT on real robot arms in manufacturing tasks. (๐Ÿ”– Bookmark this for later!) The task was precise: place thin rectangular frames from a messy stack into a holder. The team fine-tuned each model on 100 real trajectories and compared training time, inference speed, motion quality, and success rates. โฌ‡๏ธ Hereโ€™s a breakdown of what they found Pi0 (Original) โœ… Strongest overall performance in precise pick-and-place โœ… High success rate even in edge cases โœ… Longest training time (~11 hours, ~$30 per run) โœ… Inference time of 80 ms causes short pauses between actions Despite delays, it handles complex scenarios wellโ€ฆ solid for high-precision tasks, but slow to train. Gr00t โœ… Trains fast (~2 hours, ~$5 per run) โœ… Performs almost as well as Pi0 on large-object tasks โœ… Struggles with fine precision; random movement in some trials โœ… More training didnโ€™t fix jitter or random offsets Best suited for tasks where exact precision isnโ€™t critical. Not ready for manufacturing-grade accuracy without more tuning. Pi0 Fast โœ… Promised faster training, but results were underwhelming โœ… Training at 6 hours still showed low success rates โœ… Inference was slower than expected โœ… Not reliable for generalizing even slightly new tasks Currently too unstable for real-world deployment. Doesnโ€™t live up to the โ€œFastโ€ name yet. ACT (Baseline) โœ… 200MB modelโ€”lightweight, but limited โœ… Struggles with stacked objects or ambiguous scenes โœ… Success rates around 70% in best-case setups โœ… Canโ€™t match newer models on precision or generalization Still a solid baseline, but clearly a generation behind in robustness. ๐Ÿšจ Extra Notes All newer models share a common issue: โ€ขInference takes longer than a frame (80 ms vs 33 ms), so robots โ€œpauseโ€ between chunks. โ€ขThis results in jittery movements, but not a dealbreaker unless tasks are time-sensitive. Language-conditioned tasks also fell short: after training on two labeled tasks, the model couldnโ€™t generalize to a third unseen combination using only text prompts. โœ… The good news? These models adapt well to new robot arms with quick fine-tuning. โŒ The bad news? Thereโ€™s still no plug-and-play solution for improving performance after deployment. Reinforcement learning or DAgger-style data collection during real-world operation may be the next big step, something many teams in robotics are actively working on.

Ilir Aliu

21,844 views โ€ข 1 year ago

SOTUS 10TH ร— JANHAE D1 #SOTUS10thYearConcertD1 #JANHAE #JANIVERSE ๐ŸฆŠ: ok, iโ€™ll go next. first of all, thank you gmmtv and pโ€™tha for making this concert happen. SOTUS was my first series, where i got to work as an actor for the first time. the people on this stage are my first group of actor friends. evern since rehearsal day, iโ€™ve been excited, because some friends here i hadnโ€™t seen for such a long time. once we finished filmingโ€ฆ ๐ŸฆŠ: once we finished filming, or once all the activities from that period ended. we didnโ€™t see or talk to each other. so i thought, โ€œhow is this going to be? weโ€™re going to see each other again. what will everyone be like?โ€ but in my heart, i missed them and wanted to see them, because i was very happy during the time we were filming the series. it was so much fun! it was like going to see your friends. like i said, the cast wasnโ€™t just the people here, but there were so many of us. every time i went to set for this series, i thought, โ€œwow, is filming a series really this fun?โ€ because i got to go see my friends and hang out with them ๐ŸฆŠ: this time, we got to come together for this concert and see my friends for the first time. like prince, we hadnโ€™t seen each other for such a long time, but we can still talk to each other like before. or like pโ€™oat, ice, nโ€™neen, and so on, whom i hadnโ€™t seen in a long time. everyone is still the same as before. i feel very grateful for this concert for allowing us to come back together and see each other. every day when we had rehearsal, i didnโ€™t want to go home for real! it was so much fun! it was like getting to see and hang out with friends. all phis and friends are still very mischievous like before. everyone has been saying that the staff got a bit tired! but it has been so much fun. every day has been so fun, including today. iโ€™m very grateful that today exists ๐ŸฆŠ: most importantly, thank you everyone who is here and at home watching. if there werenโ€™t all of you here, there wouldnโ€™t be us up here either. thank you so much!

ยฒยฒ

30,888 views โ€ข 21 days ago

โšซ๏ธ UNCANNY VALLEY: AI IN SPORTS โ€” FAIR OR UNFAIR ADVANTAGE? Are we witnessing the end of human merit in sports? Swupnil Sahai ๐ŸŽพ๐Ÿง โ€”ex-Tesla engineer and co-founder of SwingVisionโ€”thinks AI isn't just changing the game. It's redefining the rules, the training, even the players themselves. From iPhones that out-call line judges to AI coaching that levels the field, this episode explores what happens when everyone has a superpowerโ€”and why the next era of sports might feel more like cyborg vs. cyborg. But Swupnil says the real win? It's access. AI could democratize elite performance for everyone. Welcome to The Uncanny Valley Weekly Series with Dr Danish, Fridays at 4:20PM ET, ONLY on ๐•. Episode 8: THE AUGMENTED ATHLETEโ€”ARE WE ALL AI JUICING NOW? 00:22 โ€“ โ€œItโ€™s already better than us at some tasks.โ€ 01:21 โ€“ Why user-friendly AI will explode in the next 24 months. 02:41 โ€“ Most people still underestimate whatโ€™s coming. 04:30 โ€“ Universal basic income? Guardrails? Swupnil says we better start now. 06:08 โ€“ โ€œAI might be more dangerous than nukes.โ€ 07:30 โ€“ The kill switch questionโ€”and why Tesla has it right. 08:27 โ€“ AI is now coaching, calling, and editing your tennis matches. 09:25 โ€“ AI > coach? Why SwingVision is training better and cheaper. 10:29 โ€“ The Elon Musk moment that changed everything. 12:48 โ€“ On-device AI: faster, private, and zero marginal cost. 15:21 โ€“ No data? No problem. Building training sets from scratch. 16:32 โ€“ AI wonโ€™t ruin sportsโ€”it could equalize them. 17:43 โ€“ Microchips and implants: when tech becomes the body. 19:29 โ€“ Will SwingVision become more than sports? 20:00 โ€“ Craziest prediction: AI commentators for every high school game. 21:50 โ€“ Subscription vs. surveillance: the real cost of AI access. 23:42 โ€“ Hardware is kingโ€”and itโ€™s catching up fast. 24:54 โ€“ โ€œ10 people will run the next billion-dollar startup.โ€ 25:33 โ€“ His biggest fear? โ€œRegulation is too slowโ€”and bad actors move fast.โ€ 26:06 โ€“ His message to AGI in 2040: โ€œHey... we come in peace.โ€ 28:33 โ€“ Trusted by pros like Andy Roddick + 100+ D1 teams. 29:36 โ€“ Are we all just juicing with AI now? 30:48 โ€“ โ€œItโ€™s still up to you. Itโ€™s on your racket. Literally.โ€

Mario Nawfal

1,602,951 views โ€ข 1 year ago

HOW TO LOOK COOL & GET THE GIRL WHEN YOU APPROACH โŒ Home in on the girl you like and ignore her friend whoโ€™s also staring at you โœ… Open both girls at once, talk to both, win them both over โŒ Describe yourself as chasing after her (โ€œcourage to come up to youโ€) โœ… Use a chill opener instead (e.g., โ€œLooks like youโ€™re putting a lot of thought into that decisionโ€) โŒ Ask her name unprompted (โ€œWhatโ€™s your name?โ€; you are seeking information from her) โœ… Offer your name first (โ€œIโ€™m Danielโ€; you are offering value to her) โŒ Shake her hand like itโ€™s a new business partnership โœ… Hold her hand a moment and give it a squeeze without shaking it โŒ Use her foreignness to compliment her looks in another language (โ€œMuy bonita graciasโ€) โœ… Use her foreignness to ask what sheโ€™s doing here (creates an open loop) โŒ Compliment her looks a 3rd time while explaining yourself (โ€œcame up because youโ€™re gorgeousโ€) โœ… Have her comply with you instead (โ€œI love that bracelet. Let me have a look?โ€ [beckon her]) โŒ Get her to show you the lingerie she bought then completely change the topic โœ… Get her to show you the lingerie she bought then flirt over it (โ€œOhโ€ฆ lingerieโ€ฆ what is THAT for??โ€) โŒ Ask her out 40 seconds in after on a non-high point not building any real attraction or connection โœ… Talk to her for at least 2 minutes, build attraction + connection, THEN ask her out on a HIGH point (i.e., she is really engaged, laughing hard, showing a lot of curiosity about you, etc.) Notes from the video: โ€ข During the initial moment, she smile genuinely and eyes him up and down (see her head to toe glance) while quickly finishing what she is doing. She shows initial openness to him. HOWEVERโ€ฆ โ€ข She gives the โ€œOh, goshโ€ eye shut and grimacing nod in response to his โ€œchasing after youโ€ opener. She acknowledges itโ€™s โ€˜funnyโ€™ but still cringe. โ€ข She follows this by PICKING UP HER CUP (which she does not need to do) and holding it in front of herself defensively, creating a barrier between her and him. (If she was into the approach, sheโ€™d leave the cup there and face him without a barrier.) โ€ข After he blasts her with the fourth compliment, she backs away from him while laughing awkwardly and eye codes her friend (whom this guyโ€™s still ignoring) in disbelief. โ€ข When he asks her out on โ€œlike a food dateโ€, she takes two seconds to break eye contact to the side, still with her forced smile, briefly shakes her head โ€œnoโ€, then nods politely and says it โ€œsounds goodโ€ and gives her number (while eye coding her friend again) just to get rid of him. โ€ข After he gets the number, he tells her โ€œNext Friday?โ€ and she doesnโ€™t even need to think about what she has going on or check her schedule because she has no intention of actually meeting this guy, so she just says โ€œNext Friday, perfectโ€ to get this over with. Approach better & avoid these mistakes!

Girls Chase ๐Ÿƒโ€โ™€๏ธ๐Ÿ’จ

65,755 views โ€ข 3 months ago

This was the LAST footage of Dolly Parton before her passing today at age 80. Just watched Dollyโ€™s video from the Night Flight Expedition opening at Dollywood. She couldnโ€™t make the trip. Doctors clipped her wings. Dizzy. Dehydrated. Sitting there in those sparkly butterfly wings, still smiling, still talking to her people like family. She said sheโ€™d keep everybody in the loop. Promised to see us down the road. Then she pushed the new ride anyway. That was her last public message. Today we found out she kept her word the hard way. Dolly Parton was 80. Born in a one-room cabin in the Smokies, raised on hard work and harder times, and somehow turned it into one of the purest American success stories weโ€™ve ever seen. No scandal machine. No political lectures. Just songs that still hit, a theme park that employs thousands, and a heart that never got too big for the mountains she came from. She spent the last stretch of her life dealing with real health problems after years of putting her own needs second so she could care for Carl. She told us straight. No filters. No PR spin. Just Dolly. In a world full of manufactured celebrities and empty noise, she stayed real till the end. Thatโ€™s rare. Thatโ€™s the kind of toughness and grace that doesnโ€™t come from Hollywood or the coasts. It comes from people who know what it means to work, to love hard, and to keep showing up even when the body starts saying no. Rest easy, Dolly. You kept the loop open just like you promised. Weโ€™ll see you down the road. Godspeed.

AmericanPapaBearโ„ข

83,553 views โ€ข 25 days ago