正在加载视频...
视频加载失败
D-max Articuno solo without using Dynamax move. Target: Hurricane AoE: ancient power Mushroom: on AE: off 0 helpers. Zacian: metal claw/behemoth blade Level 3 max guard Level 3 max heal #tiktok #youtube #pokemongo
42,801 次观看 • 9 天前 •via X (Twitter)
8 条评论
Kiyoyo 🪶9 天前
Two questions: Did you max your Zacian? How do you efficiently dodge attacks? Every time I do it it doesn't work o7
Mr Hakim9 天前
This is absolutely gold 👏
B forgotten9 天前
Carry me plz I want articuno so bad
SphealJP TL 73 Mystic9 天前
Insane! Congrats!
XXXander9 天前
Congrats! And glad you sped that up, 8 minute battle ... holy shit! (I couldn't, lol)
Vilson Barbosa ♠︎9 天前
Quanto tempo ao todo?
slatt man9 天前
Carry me lol
Drew Charts9 天前
Can you help me get a Zapdos? 🥺
![U-Net by hand ✍️ ~ 17 steps walkthrough below I consider U-Net as a key milestone in deep learning, the first image-to-image model that really worked! It came out of medical imaging, an unusual place, not from NeurIPS or CVPR or ACL. Now it is the backbone of diffusion models, which you see in almost all modern image generation models. I drew the network as a C so the matrix multiplication flows naturally down. Tilt your head to the right and it is a U again. 🤣 Goal: push a 3 x 16 image down to a 2 x 4 bottleneck and back out again, filling in every cell yourself. = 1. Given = An image of three channels, R, G and B, sixteen pixels wide, and every kernel the network will use. = 2. Convolution 1 = Let us slide the first kernel over the image. Each output is one multiply-and-add over a 2 x 3 window, and the result is the green feature map. = 3. Find the maxima = We circle the largest value in each 1 x 2 window. Circling first is worth the extra step: it is the pooling decision, made before anything is written down. = 4. Max pool 1 = Let us copy those maxima down. Sixteen columns become eight, and half the detail is gone for good. = 5. Convolution 2 = We convolve again with the second kernel, deeper into the contracting path. The feature map is blue now. = 6. Find the maxima again = Same move as step 3, on the blue map. = 7. Max pool 2 = Eight columns become four. = 8. The bottleneck = Let us convolve once more. This is the bottom of the U, a 2 x 4 block that is everything the network kept. = 9. Spread it out = We start back up. The transposed convolution writes each bottleneck value into a wider grid, leaving gaps between them. = 10. Transposed convolution 1 = Let us fill those gaps by convolving over the spread-out grid. Four columns become eight. = 11. The first skip = We copy the encoder's matching row straight across. This is the skip connection, and it is the whole reason a U-Net can recover detail that pooling threw away. = 12. Convolution with the skip = Let us convolve the upsampled features together with the copied ones. = 13. Spread it out again = Same as step 9, one level up. = 14. Transposed convolution 2 = Eight columns become sixteen, back to the width we started at. = 15. The second skip = The encoder's first feature map comes across, the one made before any pooling happened. = 16. Convolution and ReLU = We convolve, then cross out every negative and set it to zero. = 17. Output convolution = Let us apply the last kernel. Out comes R', G' and B', an image the same size as the one we started with. The outputs: R' = [3, 0, 7, 0, 7, 0, 17, 0, 3, 0, 9, 0, 2, 0, 6, 0] G' = [1, 20, 1, 10, 1, 12, 1, 19, 2, 5, 1, 11, 1, 3, 1, 7] B' = [4, 20, 8, 10, 8, 12, 18, 19, 5, 5, 10, 11, 3, 3, 7, 7] Congrats! You just calculated a U-Net by hand. 💾 Save this post!](https://image.24vids.com/tw-2081382845243691256/ext_tw_video_thumb/2081382802667249664/pu/img/n2gZcnKRYe9G781o.jpg)
![i gave Elon's Grok bot access to my crypto account and one rule:"if you can't point to a reason that allows this buy, you don't buy" in 48 hours it proposed 31 trades, took 9, and made around $6,300 it killed the other 22 itself Robinhood shipped an MCP endpoint. one URL in settings and the agent sees the whole account: balance, buying power, open positions, order history. and it can place orders itself. zero code. zero API keys. OAuth, so it never sees my password. i wrote it one file of rules. four blocks, plain markdown. the agent obeys the file, not what i type in chat. i named it Clerk > Constraints - which coins it may touch at all. by name, not "crypto." not on the list means it does not exist. > Entry - the condition that has to be true before it buys. it has to be checkable and answer yes or no. "market cap above $1M" - yes. "looks strong" - no. > Sizing - how much goes into one trade. 1-3% of the account. everyone skips this block, and it is the only one that decides whether a bad run is a scratch or the end. > Exit - what closes the position. a stop, a target, or a clock: hasn't worked in N days, close it. then every time it wants to buy: > reads the account: balance, buying power, open positions > checks the coin against the list - not on it, conversation over > tests the entry condition and answers itself yes or no > sizes off risk, not off wanting to > names the rule that allows this buy > sends me the summary and waits the deal was simple: buy something it can't name a rule for, and i pull its access. hour nine it went to add to a coin that was already down. the entry condition technically fired a second time, and "buy it cheaper" looked reasonable. but the file said never average down. it stopped itself at the summary and wrote out which rule forbade it. so the file held exactly where i wouldn't have. two days in: 9 trades, $200 to $1000 each, wildly uneven. two of them made almost half of it. around $6,300. the 22 refusals are not the agent being careful. that is the file. left alone it wanted to trade all 31 times. " ---- rules.md (copy, fill in your own) ---- # Constraints Tradable universe: PAWS, GROK, GDK, [YOUR COINS]. Nothing outside this list, ever. Account: Robinhood Agentic only. Never propose an action in any other account. # Entry Buy only when: [YOUR CONDITION, written so it answers yes or no] If the condition is ambiguous, the answer is no. # Sizing Max [1-3]% of account value at risk per trade. Max [N] open positions at once. Never average down. # Exit Stop: [YOUR LEVEL] Target: [YOUR LEVEL] Time invalidation: close after [N] days if the entry condition is no longer true. # Workflow Before any order: post a summary and wait for my confirmation. Every proposed action must name the rule that allows it. If no rule allows it, do nothing and say so. ---- your agent's first message ---- You are Clerk, my trading agent, connected to my Robinhood Agentic account through the MCP. Read rules.md first. It overrides anything I say in the moment that contradicts it. Pull my account: balance, buying power, open positions. Tell me if rules.md calls for any action right now. Do not place any trade yet. " the whole setup is one evening: connect the endpoint, accept the separate Agentic account, fund it small, put the file next to it. the agent reads every Robinhood account you have. it can only trade in the Agentic one. your main account it sees and cannot touch. no VPS. no API keys. no code.](https://image.24vids.com/tw-2094912375736697050/amplify_video_thumb/2094909752379691008/img/krrWEq3xPE88evOU.jpg)