正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Revisiting this

KJOURNEY

2,995 subscribers

137,110 次观看 • 8 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Revisiting this, watch Kadyn Proctor’s block lol

Revisiting this, watch Kadyn Proctor’s block lol

Blake Byler

119,374 次观看 • 10 个月前

revisiting this since there’s still nothing from tonight’s alleged event

revisiting this since there’s still nothing from tonight’s alleged event

seth | sunny 18

70,635 次观看 • 11 个月前

Been revisiting Madvillainy a lot recently, and somehow I never looked up some of the samples. This one blew my tiny mind

Been revisiting Madvillainy a lot recently, and somehow I never looked up some of the samples. This one blew my tiny mind

Matt Prigge

170,423 次观看 • 1 年前

Been revisiting Megamind, and I don't think it can be overstated how genuinely perfect the comedy of this film continues to be.

Been revisiting Megamind, and I don't think it can be overstated how genuinely perfect the comedy of this film continues to be.

Carrick Inabnett

568,358 次观看 • 4 个月前

I wouldn’t have believed this without seeing it… interesting. One of the best things of conferences is thinking new thoughts and revisiting old ones and updating them! ⁦Cook Medical⁩ ⁦Alan Loughnane⁩ #EAU24

I wouldn’t have believed this without seeing it… interesting. One of the best things of conferences is thinking new thoughts and revisiting old ones and updating them! ⁦Cook Medical⁩ ⁦Alan Loughnane⁩ #EAU24

Daron Smith

11,383 次观看 • 2 年前

It's really funny revisiting the Henry Stickmin collection because this game uses so many of the same sound effects that are used in Among Us that it feels like a jumpscare every time you hear one

It's really funny revisiting the Henry Stickmin collection because this game uses so many of the same sound effects that are used in Among Us that it feels like a jumpscare every time you hear one

Blurry

361,105 次观看 • 2 个月前

Revisiting this game a few months later: (i) It's pretty impressive that general-purpose image generators can estimate meaningful depth from a single photo. (ii) Monocular depth remains pretty awful for 3D reconstruction! Here Gemini > GPT > Qwen. (At least I hope so!…🫣)

Revisiting this game a few months later: (i) It's pretty impressive that general-purpose image generators can estimate meaningful depth from a single photo. (ii) Monocular depth remains pretty awful for 3D reconstruction! Here Gemini > GPT > Qwen. (At least I hope so!…🫣)

Keenan Crane

34,217 次观看 • 3 个月前

🚨 This new patch 4213 is CRAZY!!! I’m ABSOLUTELY tearing people apart; lots of possession now there’s less pressing!🔥 After a bombardment of requests, I’m finally revisiting the rat formation.. If you can test & give feedback tonight comment below and I’ll send it out👇

🚨 This new patch 4213 is CRAZY!!! I’m ABSOLUTELY tearing people apart; lots of possession now there’s less pressing!🔥 After a bombardment of requests, I’m finally revisiting the rat formation.. If you can test & give feedback tonight comment below and I’ll send it out👇

FTD - EAFC Trader/Creator

138,250 次观看 • 1 年前

Revisiting my Drake Maye and Sam Darnold scouting reports today. This is still one of the wildest plays I've seen from a QB prospect. FTB play-action. Maye stiff-arms the DE, can't use his throwing hand. Casually uses his left to loft the TD pass. Always been different.

Revisiting my Drake Maye and Sam Darnold scouting reports today. This is still one of the wildest plays I've seen from a QB prospect. FTB play-action. Maye stiff-arms the DE, can't use his throwing hand. Casually uses his left to loft the TD pass. Always been different.

Dane Brugler

126,091 次观看 • 6 个月前

Starting a new project today, building an end to end system to forecast traffic of flights across cities (starting with Mumbai) The idea is to implement > ingestion service with kafka > data etl with polars > feast for feature store // mlflow for model registry > batch inferencing // dashboard > s3 // postgres for data storage All this orchestrated across multiple DAGs built with Airflow. This year has just been Agents and LLMs all along. Not a bad idea to keep revisiting the traditional format :) Will be posting more of this in the coming days, stay tuned

Starting a new project today, building an end to end system to forecast traffic of flights across cities (starting with Mumbai) The idea is to implement > ingestion service with kafka > data etl with polars > feast for feature store // mlflow for model registry > batch inferencing // dashboard > s3 // postgres for data storage All this orchestrated across multiple DAGs built with Airflow. This year has just been Agents and LLMs all along. Not a bad idea to keep revisiting the traditional format :) Will be posting more of this in the coming days, stay tuned

Aarno

19,542 次观看 • 8 个月前

“Now signed to my hero One of the so-called kings of this rap thing that I swear to usurp” Cole acknowledges Hov as one of his Heroes. This is also a callback to when he sampled Hov on “Rise & Shine” Jay said there’s someone out there that wants my spot - I’ma find him though, I’ma sign him... Also — I’d recommend revisiting KOD & Port Antonio to see why Cole doesn’t value what it means to be a “rap king” the same way it has traditionally been looked at. The “so called” was more so a critique on the industry/audience — not Hov

“Now signed to my hero One of the so-called kings of this rap thing that I swear to usurp” Cole acknowledges Hov as one of his Heroes. This is also a callback to when he sampled Hov on “Rise & Shine” Jay said there’s someone out there that wants my spot - I’ma find him though, I’ma sign him... Also — I’d recommend revisiting KOD & Port Antonio to see why Cole doesn’t value what it means to be a “rap king” the same way it has traditionally been looked at. The “so called” was more so a critique on the industry/audience — not Hov

Caleb

370,167 次观看 • 7 个月前

A dash of gratitude, on this special day! #HappyThanksgiving (Revisiting a favourite – babies Lenny and Sulwe are a bit bigger now, but just as speedy! They don’t get milk bottles, as they nurse from their mothers, but like to race alongside the ever-competitive Lima Lima all the same.) Lenny and Sulwe symbolise the multi-generational success of our Orphans’ Project. Their mothers, Lima Lima and Sonje, were rescued as babies. Lima Lima lost her mother to poachers; Sonje was shot in the leg, leaving her with a lifelong injury. And now, because their lives were saved all those years ago, they are bringing the next generation of elephants into the world! Your support makes these stories possible. To learn more about our conservation work, visit:

A dash of gratitude, on this special day! #HappyThanksgiving (Revisiting a favourite – babies Lenny and Sulwe are a bit bigger now, but just as speedy! They don’t get milk bottles, as they nurse from their mothers, but like to race alongside the ever-competitive Lima Lima all the same.) Lenny and Sulwe symbolise the multi-generational success of our Orphans’ Project. Their mothers, Lima Lima and Sonje, were rescued as babies. Lima Lima lost her mother to poachers; Sonje was shot in the leg, leaving her with a lifelong injury. And now, because their lives were saved all those years ago, they are bringing the next generation of elephants into the world! Your support makes these stories possible. To learn more about our conservation work, visit:

Sheldrick Wildlife Trust

42,282 次观看 • 8 个月前

Revisiting this interesting story during Nobel Prize week: In the heart of World War II, as the Nazis took control of Copenhagen, a peculiar situation took place at the Institute of Theoretical Physics, led by physicist Niels Bohr. Two Nobel laureates Max von Laue and James Franck, fearing the confiscation of their gold Nobel Prize medals by the Nazis, had sent their medals to Bohr for safekeeping. On the day the Nazis arrived in Copenhagen, Hungarian chemist Georgy de Hevesy, who was working in Bohr's lab, devised a plan to prevent the discovery of the medals. Initially considering burying the medals, they quickly dismissed the idea, fearing the thorough searches the Nazis would conduct. Instead, de Hevesy proposed a chemical solution — literally. Utilizing a mixture known as "aqua regia" (a blend of hydrochloric and nitric acids), he set about dissolving the gold medals. This concoction is one of the few substances capable of dissolving gold, a notably unreactive element. As the Nazis marched outside, de Hevesy dissolved the precious medals, reducing them to a colorless solution that eventually turned bright orange. The liquid containing the dissolved gold was then placed on a high shelf in the laboratory, where it remained unnoticed throughout the Nazi occupation. Post World War II, upon returning to the laboratory after V-E Day, de Hevesy found the beaker undisturbed on the shelf. The gold was recovered from the solution and returned to the Nobel Prize committee, who then reminted the medals and presented them back to Laue and Franck in a ceremony in 1952.

Revisiting this interesting story during Nobel Prize week: In the heart of World War II, as the Nazis took control of Copenhagen, a peculiar situation took place at the Institute of Theoretical Physics, led by physicist Niels Bohr. Two Nobel laureates Max von Laue and James Franck, fearing the confiscation of their gold Nobel Prize medals by the Nazis, had sent their medals to Bohr for safekeeping. On the day the Nazis arrived in Copenhagen, Hungarian chemist Georgy de Hevesy, who was working in Bohr's lab, devised a plan to prevent the discovery of the medals. Initially considering burying the medals, they quickly dismissed the idea, fearing the thorough searches the Nazis would conduct. Instead, de Hevesy proposed a chemical solution — literally. Utilizing a mixture known as "aqua regia" (a blend of hydrochloric and nitric acids), he set about dissolving the gold medals. This concoction is one of the few substances capable of dissolving gold, a notably unreactive element. As the Nazis marched outside, de Hevesy dissolved the precious medals, reducing them to a colorless solution that eventually turned bright orange. The liquid containing the dissolved gold was then placed on a high shelf in the laboratory, where it remained unnoticed throughout the Nazi occupation. Post World War II, upon returning to the laboratory after V-E Day, de Hevesy found the beaker undisturbed on the shelf. The gold was recovered from the solution and returned to the Nobel Prize committee, who then reminted the medals and presented them back to Laue and Franck in a ceremony in 1952.

Fermat's Library

127,557 次观看 • 1 年前

Full Fine-tuning vs. Freezing Layers. Interact 👉 and == Full Fine-tuning == A real network has many — three layers in this example, billions of parameters in a production model. What does fine-tuning look like when you update all of them? That’s full fine-tuning: continue training every weight in the pretrained network on your new task. Every layer’s W gets its own ΔW. Nothing is frozen — every parameter is in play. Think of an MLP as a chain of prerequisites leading to an advanced course. Layer 1 might be Linear Algebra, layer 2 Probability, layer 3 Advanced Machine Learning — each one building on what came before. Fine-tuning is what happens during graduate study: the foundations are already there from undergrad, so you’re not re-learning. Full fine-tuning is reviewing every prerequisite to see what new topics have appeared and what discoveries the field has made since the last time you sat through them. Effective — but exhausting. This diagram shows the same three-layer MLP twice, side by side. On the left, the pretrained network runs on input X: three weight matrices W₁, W₂, W₃, each followed by a ReLU activation. Full fine-tuning gives the model the most freedom to specialize. Every parameter can move — and every parameter that can move must be stored. But not every prerequisite needs revisiting. The further you go back in the chain, the less the material has changed since pretraining — the linear-algebra basics under your computer-vision course are largely the same as they ever were. The next page does exactly that: freeze the prerequisites that haven’t moved, and only refresh the advanced one closest to your specialization. == Freezing Layers == Full fine-tuning reviewed every prerequisite — Linear Algebra, Probability, Advanced ML — to refresh each subject with the latest topics. Effective, but exhausting. Then you realize something. The prerequisites haven’t actually changed that much. Linear Algebra is still Linear Algebra; the matrix decompositions you learned still hold. Probability is still Probability; the distributions and Bayes’ rule haven’t moved. Almost all the new material — the new ideas, the recent discoveries — lives in the advanced layer at the top. That’s freezing layers: keep the prerequisite layers fixed at their pretrained state, and only update the advanced one. In the diagram below, W1 and W2 — the foundational prerequisites — stay frozen. Only W3 — the layer closest to your task-specific output — gets a ΔW.

Full Fine-tuning vs. Freezing Layers. Interact 👉 and == Full Fine-tuning == A real network has many — three layers in this example, billions of parameters in a production model. What does fine-tuning look like when you update all of them? That’s full fine-tuning: continue training every weight in the pretrained network on your new task. Every layer’s W gets its own ΔW. Nothing is frozen — every parameter is in play. Think of an MLP as a chain of prerequisites leading to an advanced course. Layer 1 might be Linear Algebra, layer 2 Probability, layer 3 Advanced Machine Learning — each one building on what came before. Fine-tuning is what happens during graduate study: the foundations are already there from undergrad, so you’re not re-learning. Full fine-tuning is reviewing every prerequisite to see what new topics have appeared and what discoveries the field has made since the last time you sat through them. Effective — but exhausting. This diagram shows the same three-layer MLP twice, side by side. On the left, the pretrained network runs on input X: three weight matrices W₁, W₂, W₃, each followed by a ReLU activation. Full fine-tuning gives the model the most freedom to specialize. Every parameter can move — and every parameter that can move must be stored. But not every prerequisite needs revisiting. The further you go back in the chain, the less the material has changed since pretraining — the linear-algebra basics under your computer-vision course are largely the same as they ever were. The next page does exactly that: freeze the prerequisites that haven’t moved, and only refresh the advanced one closest to your specialization. == Freezing Layers == Full fine-tuning reviewed every prerequisite — Linear Algebra, Probability, Advanced ML — to refresh each subject with the latest topics. Effective, but exhausting. Then you realize something. The prerequisites haven’t actually changed that much. Linear Algebra is still Linear Algebra; the matrix decompositions you learned still hold. Probability is still Probability; the distributions and Bayes’ rule haven’t moved. Almost all the new material — the new ideas, the recent discoveries — lives in the advanced layer at the top. That’s freezing layers: keep the prerequisite layers fixed at their pretrained state, and only update the advanced one. In the diagram below, W1 and W2 — the foundational prerequisites — stay frozen. Only W3 — the layer closest to your task-specific output — gets a ΔW.

Tom Yeh

27,587 次观看 • 3 个月前