Loading video...

Video Failed to Load

Go Home

Just received the first SMT PCB of the Raspberry Pi RTL-SDR/LoRa/GPS/RTC Hat from JLCPCB, and everything works like a charm. I even implemented the GPS PPS indicator as Jeff Geerling suggested. I can think of some fun use cases with it, like using an RTL-SDR to listen in on...

VileerDeng's profile picture

Vileer

1,138 subscribers

63,794 views • 4 months ago •via X (Twitter)

EducationHealth & WellnessScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Just made a U.FL connector version for the RTL-SDR/GPS/LoRa/RTC hat for the Raspberry Pi. So it can be easily installed on other devices, like the Waveshare PocketTerm35, and turned into a portable SDR station. Jeff Geerling To join the waiting list:
0:23

Just made a U.FL connector version for the RTL-SDR/GPS/LoRa/RTC hat for the Raspberry Pi. So it can be easily installed on other devices, like the Waveshare PocketTerm35, and turned into a portable SDR station. Jeff Geerling To join the waiting list:

Vileer

20,003 views • 4 months ago

Haven't seen any demos of using vibe coding to prototype gadgets like this So I made one. Everything here is generated with code Parts of the code and logic can even be reused on a Raspberry Pi to build the real thing Some tips ↓
0:40

Haven't seen any demos of using vibe coding to prototype gadgets like this So I made one. Everything here is generated with code Parts of the code and logic can even be reused on a Raspberry Pi to build the real thing Some tips ↓

Dilum Sanjaya

116,619 views • 2 months ago

#49ers GM John Lynch on using AI to scout players and project in the draft process: “Yeah. A lot. Just like the rest of the world. I think if you aren’t using it, you’re already behind. … Just like you at home planning a travel itinerary—you can just ask the thing, and it can spit out pretty good things. Some pretty good ideas.”
0:55

#49ers GM John Lynch on using AI to scout players and project in the draft process: “Yeah. A lot. Just like the rest of the world. I think if you aren’t using it, you’re already behind. … Just like you at home planning a travel itinerary—you can just ask the thing, and it can spit out pretty good things. Some pretty good ideas.”

Ari Meirov

806,122 views • 4 months ago

“The last time that we were touring and stuff, I think touring is quite an isolating experience. And on top of that, when we started touring, it was 2021, so it was kind of like still depths of the pandemic. We were like one of the first tours to go back out. So I think that tour started in that way where it was like, I was really bubbled up even kind of within the tour. And then staying on that tour, we put an album out and it ended up being like two and a half years type time. It's just a long time to be in like a bubble where you're not really spending that much time like in the world around you, other than shows and hotels and stuff. And when I finished, I kind of went off and was spending a lot of time in Italy, and was kind of like having this quiet time, and I was just feeling a lot of my life as I was kind of like, craving like quiet and kind of like private spaces. I was just feeling myself kind of like shut out a lot of the world, “I think, and I think the last couple of years for me was about, you know, there's a lot of things that I think I got used to saying no to, like invites from friends and like maybe a weekend trip or like a birthday party or something. I think I just got used to, you know, if I have a week at home, I don't want to go somewhere else for three days. I just want to stay home, you know. And I think when I got to Italy and was kind of out in this quiet place, I felt like, okay, if I'm going to spend all this time out here and I'm going to keep saying no to everything, I'm just going to like shut myself off from the world. And while parts of that sound appealing, I also like, I love people and I love like being in the world, you know, so it was kind of, I think for me, like there's been both the thing of like learning to set boundaries and stuff is a large part of that has also been learning when not to set boundaries and when to open up. Being intentional with the things that you don't do and being intentional with the things that you do do. So, you know, as a message in the album sleeve, where I said like thanks to all the people who helped me know when to say no and when to say yes. I think both of those have like greatly enriched my life.” - Harry about feeling isolated during his last tour and opening himself up to the world again on Q
3:02

“The last time that we were touring and stuff, I think touring is quite an isolating experience. And on top of that, when we started touring, it was 2021, so it was kind of like still depths of the pandemic. We were like one of the first tours to go back out. So I think that tour started in that way where it was like, I was really bubbled up even kind of within the tour. And then staying on that tour, we put an album out and it ended up being like two and a half years type time. It's just a long time to be in like a bubble where you're not really spending that much time like in the world around you, other than shows and hotels and stuff. And when I finished, I kind of went off and was spending a lot of time in Italy, and was kind of like having this quiet time, and I was just feeling a lot of my life as I was kind of like, craving like quiet and kind of like private spaces. I was just feeling myself kind of like shut out a lot of the world, “I think, and I think the last couple of years for me was about, you know, there's a lot of things that I think I got used to saying no to, like invites from friends and like maybe a weekend trip or like a birthday party or something. I think I just got used to, you know, if I have a week at home, I don't want to go somewhere else for three days. I just want to stay home, you know. And I think when I got to Italy and was kind of out in this quiet place, I felt like, okay, if I'm going to spend all this time out here and I'm going to keep saying no to everything, I'm just going to like shut myself off from the world. And while parts of that sound appealing, I also like, I love people and I love like being in the world, you know, so it was kind of, I think for me, like there's been both the thing of like learning to set boundaries and stuff is a large part of that has also been learning when not to set boundaries and when to open up. Being intentional with the things that you don't do and being intentional with the things that you do do. So, you know, as a message in the album sleeve, where I said like thanks to all the people who helped me know when to say no and when to say yes. I think both of those have like greatly enriched my life.” - Harry about feeling isolated during his last tour and opening himself up to the world again on Q

HSNews

90,121 views • 5 months ago

I transformed my physical door into a mixed-reality portal that can bring you either to your next room or to magical virtual worlds! You can select where you want to go (like when you choose the floor in an elevator) and then when you open the door, you can exactly go to that place, physical or virtual that it may be. What I love the most is that this demo connects the physicality of the door with mixed reality, so you use a real tool to be able to access virtual realities. I've made this demo using #unity3d and #MetaQuest SDK. I've used Meta Scene Setup, so the demo is fully dynamic: it finds my door and adapts all the interfaces to it. Theoretically, it could work with all sliding doors that work like mine with no modifications to the APK. As usual, I'm thinking of making a blog post where I explain how I did it... let me know if you could be interested! And let me know also if you could be interested in some collaborations with me ;) The full video of the demo is on Youtube: #metaverse #VirtualReality #SpatialComputing
2:07

I transformed my physical door into a mixed-reality portal that can bring you either to your next room or to magical virtual worlds! You can select where you want to go (like when you choose the floor in an elevator) and then when you open the door, you can exactly go to that place, physical or virtual that it may be. What I love the most is that this demo connects the physicality of the door with mixed reality, so you use a real tool to be able to access virtual realities. I've made this demo using #unity3d and #MetaQuest SDK. I've used Meta Scene Setup, so the demo is fully dynamic: it finds my door and adapts all the interfaces to it. Theoretically, it could work with all sliding doors that work like mine with no modifications to the APK. As usual, I'm thinking of making a blog post where I explain how I did it... let me know if you could be interested! And let me know also if you could be interested in some collaborations with me ;) The full video of the demo is on Youtube: #metaverse #VirtualReality #SpatialComputing

TonyVT SkarredGhost

20,071 views • 2 years ago

carlos on his relationship with lando: “another good friend, lando. we had a lot of fun together when I was driving in mclaren. we, again, formed a relationship of respect and also friendship. we tend to have a lot of fun, like our sense of humor matches. ever since, we remain good friends and we play a lot of golf together, we play padel together, we do a lot of plans together. and yeah, he is a guy that i really respect, a guy that i enjoy spending my time with. and even if we are rivals on track, we know we can also be good friends out of it.” “i was happy for him to win that championship. i think he deserved it. we are extremely competitive. when i put a helmet on, i want to destroy lando and i want to beat him and i want to beat alex, beat charles, beat all of them. and when i'm driving, i don't see lando in front of me or i don't see... i just see a ferrari or a mclaren and i just know i need to overtake him. but it is true that when you take off the helmet you can see things in relative, you know, not maybe with the pressure of having the helmet on and racing to death like we do.”
1:26

carlos on his relationship with lando: “another good friend, lando. we had a lot of fun together when I was driving in mclaren. we, again, formed a relationship of respect and also friendship. we tend to have a lot of fun, like our sense of humor matches. ever since, we remain good friends and we play a lot of golf together, we play padel together, we do a lot of plans together. and yeah, he is a guy that i really respect, a guy that i enjoy spending my time with. and even if we are rivals on track, we know we can also be good friends out of it.” “i was happy for him to win that championship. i think he deserved it. we are extremely competitive. when i put a helmet on, i want to destroy lando and i want to beat him and i want to beat alex, beat charles, beat all of them. and when i'm driving, i don't see lando in front of me or i don't see... i just see a ferrari or a mclaren and i just know i need to overtake him. but it is true that when you take off the helmet you can see things in relative, you know, not maybe with the pressure of having the helmet on and racing to death like we do.”

yoyo

13,378 views • 2 months ago

Yunjin talked about the significance of slowness in a world that moves so fast & virtue of patience ❤️‍🩹 “What I've been thinking about lately is the importance of being deliberately slow about things and allowing yourself a pocket of time in your day to not be rushed by anyone or anything, not even yourself. When we had our show in Shizuoka, we got on, we took the train there, the Shinkansen, instead of taking a flight like we would normally and it was really really nice because we were allotted an amount of time in our day to basically just sit with ourselves. You know, that's what I love about being on a plane or just in any vehicle, because you are not driving it or you could be driving it, but you can’t travel in a split second. There's time. There's a time of arrival that you can't control and you just kind of have to wait and sit with yourself. But I feel this need to be on airplane mode when I do have those opportunities to be off my phone and just, like, disconnected from the world, but also not disconnected from the world. You know what I mean? Like, when you're on a plane, you're still part of the world. You're just, like, not on the stratosphere, like, physically. And it kind of like, forces you to be, like. To, like, separate yourself for a second, like a couple hours or a couple days, if you're like a layover or something. I found myself looking out the window of the train and just, like, enjoying myself. Because everything is so fast, and I think everything. Ever since then, I've been deliberately, like, making an effort to do one thing very slowly, and it's very nice. It’s good. It's a good way to kind of put your feet back on the ground and remind yourself that not everything has to be you know, it's okay to take a breather and just wait. It's a virtue. Patience is a virtue, you know, but it's also virtue to enjoy it and gift yourself that time. It's kind of rebellious in a way, because no one can take that right from you. To do something with less urgency when everybody around you is like, of course, if it's an urgent manner, you should or urgent matter, you should do things accordingly but, you know, if it's like you don't need to wash your hair, like, in three minutes. Like, you can take your time to really think about. You know, sometimes we just live on. Autopilot so much, I think it's important to remind ourselves we're not AI.”
4:39

Yunjin talked about the significance of slowness in a world that moves so fast & virtue of patience ❤️‍🩹 “What I've been thinking about lately is the importance of being deliberately slow about things and allowing yourself a pocket of time in your day to not be rushed by anyone or anything, not even yourself. When we had our show in Shizuoka, we got on, we took the train there, the Shinkansen, instead of taking a flight like we would normally and it was really really nice because we were allotted an amount of time in our day to basically just sit with ourselves. You know, that's what I love about being on a plane or just in any vehicle, because you are not driving it or you could be driving it, but you can’t travel in a split second. There's time. There's a time of arrival that you can't control and you just kind of have to wait and sit with yourself. But I feel this need to be on airplane mode when I do have those opportunities to be off my phone and just, like, disconnected from the world, but also not disconnected from the world. You know what I mean? Like, when you're on a plane, you're still part of the world. You're just, like, not on the stratosphere, like, physically. And it kind of like, forces you to be, like. To, like, separate yourself for a second, like a couple hours or a couple days, if you're like a layover or something. I found myself looking out the window of the train and just, like, enjoying myself. Because everything is so fast, and I think everything. Ever since then, I've been deliberately, like, making an effort to do one thing very slowly, and it's very nice. It’s good. It's a good way to kind of put your feet back on the ground and remind yourself that not everything has to be you know, it's okay to take a breather and just wait. It's a virtue. Patience is a virtue, you know, but it's also virtue to enjoy it and gift yourself that time. It's kind of rebellious in a way, because no one can take that right from you. To do something with less urgency when everybody around you is like, of course, if it's an urgent manner, you should or urgent matter, you should do things accordingly but, you know, if it's like you don't need to wash your hair, like, in three minutes. Like, you can take your time to really think about. You know, sometimes we just live on. Autopilot so much, I think it's important to remind ourselves we're not AI.”

nat ☆彡

16,861 views • 21 days ago

I’m testing the new 4-step Turbo LoRA for MiniMax. With just 16GB of VRAM, I can generate 15 seconds of 480p video in around 1 minute and 30 seconds! I also noticed something pretty funny if you use a song as an audio reference and tell the character to dance to it, the character actually follows the rhythm of the music surprisingly well.
0:30

I’m testing the new 4-step Turbo LoRA for MiniMax. With just 16GB of VRAM, I can generate 15 seconds of 480p video in around 1 minute and 30 seconds! I also noticed something pretty funny if you use a song as an audio reference and tell the character to dance to it, the character actually follows the rhythm of the music surprisingly well.

Alex

49,867 views • 26 days ago

WE BLOOMING BEAUTY #PlantneryxWilliamEst #WilliamEst 🦈: i said i’m not going anywhere anyway. there are times when we feel like, maybe we’re not suitable for each other. i feel like i see everything and i realize that maybe i’m the one causing the problem. i think, well, if i step back, maybe it would be better. i’ve thought, like, if this problem didn’t exist, things might be better. it probably wouldn’t have happened. i feel like okay should we separate? but we’re talking about this from the perspective of we actually don’t want to separate. in terms of work, everything’s great. if it’s just the two of us, everything’s happy. but because of external factors, it causes a lot of problems. some of these problems make sense, some don’t. i don’t like having problems, and i feel like it’s okay to just leave it, but when i keep everything inside, it makes me sad and all kinds of things start building up. when we clear things up, we like to deep talk. once we start talking, everything gets more serious, because he doesn’t want to separate and i don’t want to separate, but why do we have to separate? that’s basically what it was like at that time. 🎸: how did we sort it out? i’ve been thinking that even if he tries to push me away, i won’t go. 🦈: i’m not pushing away 🎸: even if you try, or if you say you don’t want to do this anymore, i’ll find a way to make it work, because since we started the thamepo project, we’ve never had any problems. we just argue, which comes from outside issues. we had to talk about it. i feel like it doesn’t make sense for us to separate, so i said, i wouldn’t let him go anywhere. let’s live our lives together.
2:06

WE BLOOMING BEAUTY #PlantneryxWilliamEst #WilliamEst 🦈: i said i’m not going anywhere anyway. there are times when we feel like, maybe we’re not suitable for each other. i feel like i see everything and i realize that maybe i’m the one causing the problem. i think, well, if i step back, maybe it would be better. i’ve thought, like, if this problem didn’t exist, things might be better. it probably wouldn’t have happened. i feel like okay should we separate? but we’re talking about this from the perspective of we actually don’t want to separate. in terms of work, everything’s great. if it’s just the two of us, everything’s happy. but because of external factors, it causes a lot of problems. some of these problems make sense, some don’t. i don’t like having problems, and i feel like it’s okay to just leave it, but when i keep everything inside, it makes me sad and all kinds of things start building up. when we clear things up, we like to deep talk. once we start talking, everything gets more serious, because he doesn’t want to separate and i don’t want to separate, but why do we have to separate? that’s basically what it was like at that time. 🎸: how did we sort it out? i’ve been thinking that even if he tries to push me away, i won’t go. 🦈: i’m not pushing away 🎸: even if you try, or if you say you don’t want to do this anymore, i’ll find a way to make it work, because since we started the thamepo project, we’ve never had any problems. we just argue, which comes from outside issues. we had to talk about it. i feel like it doesn’t make sense for us to separate, so i said, i wouldn’t let him go anywhere. let’s live our lives together.

𝑛𝑎𝑛𝑎 ◡̈

137,780 views • 1 year ago

I really like 's interactive landing page My favorite landing pages are the ones that drop you straight into the application and let you use it, you barely see that though I think it makes sense because why would you even need to sell to a customer with a big sales-y landing page if your product is impressive enough to convince them to buy it just by using it? ElevenLabs does a great mix of being able to actually use all the basic features as a demo while still being a traditional landing page This is easier with sites like Nomad List, because people just browse cities and it doesn't cost me much, but it's much harder to do this with AI apps because they cost a lot of money for every time you generate something (like photo, audio or text). One idea I had for Photo AI is to pre-generate a lot of combos of photos then on the landing page make a little demo and let them choose default models (like white man, black woman, etc.) and prompts etc. Then I just pre-gen all those combinations (like 10,000 pics, maybe $100) and I can have a demo on the landing page Unaffiliated with ElevenLabs btw, just like the website!
0:42

I really like 's interactive landing page My favorite landing pages are the ones that drop you straight into the application and let you use it, you barely see that though I think it makes sense because why would you even need to sell to a customer with a big sales-y landing page if your product is impressive enough to convince them to buy it just by using it? ElevenLabs does a great mix of being able to actually use all the basic features as a demo while still being a traditional landing page This is easier with sites like Nomad List, because people just browse cities and it doesn't cost me much, but it's much harder to do this with AI apps because they cost a lot of money for every time you generate something (like photo, audio or text). One idea I had for Photo AI is to pre-generate a lot of combos of photos then on the landing page make a little demo and let them choose default models (like white man, black woman, etc.) and prompts etc. Then I just pre-gen all those combinations (like 10,000 pics, maybe $100) and I can have a demo on the landing page Unaffiliated with ElevenLabs btw, just like the website!

@levelsio

256,002 views • 1 year ago

24h later we tested more! What should we test next on the LibreBoard by 256 Foundation ? - MIPI DSI Port with a Display with Touch! - Full Raspberry Pi HAT via Sense Hat with joystick, Accelerometer, Gyroscope, Brightness, Pressure and a fun 8x8 RGB LED Matrix - Finally powered just by 12v-18v and no longer 5v and 12v, making it super easy to deploy anywhere What feature did you always want a Miner Control-board to have but couldn't because it was not open source? Let me know and I will build and test it!
0:30

24h later we tested more! What should we test next on the LibreBoard by 256 Foundation ? - MIPI DSI Port with a Display with Touch! - Full Raspberry Pi HAT via Sense Hat with joystick, Accelerometer, Gyroscope, Brightness, Pressure and a fun 8x8 RGB LED Matrix - Finally powered just by 12v-18v and no longer 5v and 12v, making it super easy to deploy anywhere What feature did you always want a Miner Control-board to have but couldn't because it was not open source? Let me know and I will build and test it!

Michael Schmid ⚡️

11,849 views • 4 months ago

An honest day punting - Monmore Green Stadium Greyhounds, winter Derby final Always such a good, fun evening at the dogs! Big shout out to Gone To The Dogs who are brining the greyhounds back to the glory days, drop them a follow! Like, comment and follow for more punting content if you enjoy it! As always I show the winners and the losers, the reality of the game! Remember if you’d like to join the Basher Watts Racing Club or HorseRacing Syndicate just head over to and it would be fantastic to have you onboard Comment with any content you’d like to see, thanks for watching. #basherwattsracing #horseracing #basherwattsracingsyndicate
13:37

An honest day punting - Monmore Green Stadium Greyhounds, winter Derby final Always such a good, fun evening at the dogs! Big shout out to Gone To The Dogs who are brining the greyhounds back to the glory days, drop them a follow! Like, comment and follow for more punting content if you enjoy it! As always I show the winners and the losers, the reality of the game! Remember if you’d like to join the Basher Watts Racing Club or HorseRacing Syndicate just head over to and it would be fantastic to have you onboard Comment with any content you’d like to see, thanks for watching. #basherwattsracing #horseracing #basherwattsracingsyndicate

Basher Watts

18,207 views • 1 year ago

Yikes: 49ers GM John Lynch reveals that he is using AI to scout players and help in the draft process. “Yeah. A lot. Just like the rest of the world. I think if you aren’t using it, you’re already behind. … Just like you at home planning a travel itinerary—you can just ask the thing, and it can spit out pretty good things. Some pretty good ideas.” 😵‍💫😵‍💫😵‍💫
0:55

Yikes: 49ers GM John Lynch reveals that he is using AI to scout players and help in the draft process. “Yeah. A lot. Just like the rest of the world. I think if you aren’t using it, you’re already behind. … Just like you at home planning a travel itinerary—you can just ask the thing, and it can spit out pretty good things. Some pretty good ideas.” 😵‍💫😵‍💫😵‍💫

Dov Kleiman

33,806 views • 4 months ago

Hunter Dickinson on his one season playing for Kansas men's basketball "I had expectations coming in of what it was going to be like. And, you know, it was even better than I expected to be able to play for Coach Self, to be able to play with these guys and just be able to play and Allen Fieldhouse...you can kind of almost dream of, you know, when you're growing up playing basketball to be able to play on the stage for this program. You know, obviously it didn't go the way we wanted to, But, you know, I had a lot of fun this year and, you know, I wouldn't trade it for the world."
0:27

Hunter Dickinson on his one season playing for Kansas men's basketball "I had expectations coming in of what it was going to be like. And, you know, it was even better than I expected to be able to play for Coach Self, to be able to play with these guys and just be able to play and Allen Fieldhouse...you can kind of almost dream of, you know, when you're growing up playing basketball to be able to play on the stage for this program. You know, obviously it didn't go the way we wanted to, But, you know, I had a lot of fun this year and, you know, I wouldn't trade it for the world."

Landon Reinhardt

199,998 views • 2 years ago

As I promised yesterday, I'll briefly explain LoRA training and share a workflow I made so you can do it quickly. First, let me answer a very common question: 'Why train LoRAs when we have such advanced models?' Even though we have incredibly advanced models now (like NBP), we still can't always get them to do specific things we want. Simplest example: the spritesheet LoRA I made the other day. I generated 1000 images with Nano Banana and only 100 were what I wanted. The LoRA I trained using those 100 images gives me nearly 100% consistent results. Second point is cost and speed. With LoRA, we can cut costs by 4-5x. And while doing that, we're generating 4-5x faster. How many images do you need for a good LoRA? This depends on your LoRA's complexity. For example, when I training the spritesheet LoRA, even though I used 100 images, I didn't include buildings in the training data, so this LoRA doesn't work for buildings. So think about your LoRA's use cases and add examples for as many use cases as possible to improve quality. What are paired images and how to train LoRAs for image-editing? When training LoRAs for image editing on fal, we call each edit example paired images - one with _start suffix, one with _end suffix. For example, if you're training a background remove LoRA, the unedited original photo will be your '_start' image. The image with background removed will be the '_end' image. Simply put: images we want to edit or use as reference get _start, target images we want to achieve get '_end'. Important: save both images with the same name. Like image332_start.jpg and image332_end.jpg. This way the system knows which images pair together. What about training LoRAs for models with multiple image inputs? Same logic. We still use _start and _end suffixes, but with one difference. Since there are multiple input images, we can number them: _start, _start1, _start2. Example: start images, 1st image = Woman portrait (image35_start.jpg) 2nd image = Glasses photo (image35_start1.jpg) 3rd image = Hat photo (image35_start2.jpg) Output image = portrait of woman wearing glasses and hat (image35_end.jpg) Can we do more detailed captioning? Yes. Similarly, you can improve training quality by creating a txt file for each set with the caption inside. Example: create image35.txt and write: 'Recreate the image by putting the glasses from the second image and the hat from the third image on the woman in the first image.' What are Steps? How many should I use? What's Learning Rate? Steps determines how many times the model sees and processes your training data (your images). Each step, the model learns a bit more. But as steps increase, so does the risk of overfitting. So there's no real default. But for a simpler LoRA with 20 paired images, 1000 steps is ideal. Here's a metaphor for the Steps and Learning Rate relationship: Imagine you have a balloon. Our goal is to inflate it to the optimal size. Steps = How many times we blow into the balloon Learning rate = How hard we blow each time If we blow too softly, we need to blow many more times. If we blow too hard, we risk popping it quickly and can't reach optimal size. Of course training won't explode, but it won't work as intended because it wasn't trained optimally. Training's done, now what? Once training's complete, you'll have a safetensors file. Every model you train on fal has a LoRA inference endpoint. In that inference, add your safetensors file link to the LoRA url input, and you can use your LoRA. Thanks for the read! The workflow in the video: If I forgot anything, let me know in the replies.
1:06

As I promised yesterday, I'll briefly explain LoRA training and share a workflow I made so you can do it quickly. First, let me answer a very common question: 'Why train LoRAs when we have such advanced models?' Even though we have incredibly advanced models now (like NBP), we still can't always get them to do specific things we want. Simplest example: the spritesheet LoRA I made the other day. I generated 1000 images with Nano Banana and only 100 were what I wanted. The LoRA I trained using those 100 images gives me nearly 100% consistent results. Second point is cost and speed. With LoRA, we can cut costs by 4-5x. And while doing that, we're generating 4-5x faster. How many images do you need for a good LoRA? This depends on your LoRA's complexity. For example, when I training the spritesheet LoRA, even though I used 100 images, I didn't include buildings in the training data, so this LoRA doesn't work for buildings. So think about your LoRA's use cases and add examples for as many use cases as possible to improve quality. What are paired images and how to train LoRAs for image-editing? When training LoRAs for image editing on fal, we call each edit example paired images - one with _start suffix, one with _end suffix. For example, if you're training a background remove LoRA, the unedited original photo will be your '_start' image. The image with background removed will be the '_end' image. Simply put: images we want to edit or use as reference get _start, target images we want to achieve get '_end'. Important: save both images with the same name. Like image332_start.jpg and image332_end.jpg. This way the system knows which images pair together. What about training LoRAs for models with multiple image inputs? Same logic. We still use _start and _end suffixes, but with one difference. Since there are multiple input images, we can number them: _start, _start1, _start2. Example: start images, 1st image = Woman portrait (image35_start.jpg) 2nd image = Glasses photo (image35_start1.jpg) 3rd image = Hat photo (image35_start2.jpg) Output image = portrait of woman wearing glasses and hat (image35_end.jpg) Can we do more detailed captioning? Yes. Similarly, you can improve training quality by creating a txt file for each set with the caption inside. Example: create image35.txt and write: 'Recreate the image by putting the glasses from the second image and the hat from the third image on the woman in the first image.' What are Steps? How many should I use? What's Learning Rate? Steps determines how many times the model sees and processes your training data (your images). Each step, the model learns a bit more. But as steps increase, so does the risk of overfitting. So there's no real default. But for a simpler LoRA with 20 paired images, 1000 steps is ideal. Here's a metaphor for the Steps and Learning Rate relationship: Imagine you have a balloon. Our goal is to inflate it to the optimal size. Steps = How many times we blow into the balloon Learning rate = How hard we blow each time If we blow too softly, we need to blow many more times. If we blow too hard, we risk popping it quickly and can't reach optimal size. Of course training won't explode, but it won't work as intended because it wasn't trained optimally. Training's done, now what? Once training's complete, you'll have a safetensors file. Every model you train on fal has a LoRA inference endpoint. In that inference, add your safetensors file link to the LoRA url input, and you can use your LoRA. Thanks for the read! The workflow in the video: If I forgot anything, let me know in the replies.

ilker

15,192 views • 7 months ago

David Sedaris thinks that if we just lived for the plot, we'd have material to write anything. He explains: "My friend asked me what I did in the afternoon. I said, 'I went to the post office.' He replied, 'Why would you do that? You don't even need to go to the post office. If you have something you want mailed, you have the hotel called FedEx and they come and pick it up.'" "I don't want to live like you. I want to go to the post office. And I went to the post office and it was remarkable. Or sometimes you go to the post office and the person is just grumpy and you can even get something out of that." "I wanna live. I wanna be out there. I don't wanna sit at home and order everything online, and use all the shortcuts."
1:26

David Sedaris thinks that if we just lived for the plot, we'd have material to write anything. He explains: "My friend asked me what I did in the afternoon. I said, 'I went to the post office.' He replied, 'Why would you do that? You don't even need to go to the post office. If you have something you want mailed, you have the hotel called FedEx and they come and pick it up.'" "I don't want to live like you. I want to go to the post office. And I went to the post office and it was remarkable. Or sometimes you go to the post office and the person is just grumpy and you can even get something out of that." "I wanna live. I wanna be out there. I don't wanna sit at home and order everything online, and use all the shortcuts."

How I Write Clips

395,943 views • 2 months ago

💾 Okay I finally bought a Raspberry Pi 5 to make into a real retro computer as a proof of concept I love the Raspberry Pi much more than my ESP32 simply because I can SSH into it and it's just like a normal computer or server, the ESP32 required specific software to re-flash it everytime and it was TEDIOUS!!!! I need a fast feedback loop to develop with so all this flashing is too slow for me The Raspberry Pi 5 instead just runs Linux which then runs DOSBox-X to run Windows 3.11 install live downloaded from there so it stays fresh I then also got a Checkmate 19" IPS Retro Monitor by Stephen Jones (very indie and Kickstarter-style product) which is like a vintage retro PC monitor but with an OLED and it runs at 1280x1024, and the Raspberry Pi 5 has HDMI out so it works perfectly with it Then I got an 8BitDo retro IBM keyboard, but the spacebar broke, so I gotta send it back and get a new one, but it was great while it worked, so it's not in this video, and I got the 8BitDo retro mouse So altogether it's a proof of concept of a real vintage retro PC with internet and it works now My end goal is to make this into production and SELL it, not for money but more for fun because I think it should exist, just like Palmer Luckey's ModRetro Chromatic Gameboy-remake I'd go with something smaller than this 19" checkmate display though, it's great but too big for the Windows 3.11 era I think, more like 14" with a matching sized retro IBM-style keyboard and Windows 3.11 preinstalled (if legally allowed by Satya Nadella) with WiFi working out of the box We could also go with Windows 98/XP, because for me too Windows 3.11 was a very short time, so I too have vague memories of it, most of my computing memories is Windows 98/XP etc. Inside it could simply run on a Raspberry Pi 5 like here, which is easily powerful enough to run all of these OS'es I see many SF founders with Apple II's in their living room, and I guess cause Apple was more popular there back then, but in the rest of the world we were mostly IBM PCs and clones that ran Windows, so it's cool to have that as one retro object working with internet in your living room Anyway, step by step, we're here now 😊
1:59

💾 Okay I finally bought a Raspberry Pi 5 to make into a real retro computer as a proof of concept I love the Raspberry Pi much more than my ESP32 simply because I can SSH into it and it's just like a normal computer or server, the ESP32 required specific software to re-flash it everytime and it was TEDIOUS!!!! I need a fast feedback loop to develop with so all this flashing is too slow for me The Raspberry Pi 5 instead just runs Linux which then runs DOSBox-X to run Windows 3.11 install live downloaded from there so it stays fresh I then also got a Checkmate 19" IPS Retro Monitor by Stephen Jones (very indie and Kickstarter-style product) which is like a vintage retro PC monitor but with an OLED and it runs at 1280x1024, and the Raspberry Pi 5 has HDMI out so it works perfectly with it Then I got an 8BitDo retro IBM keyboard, but the spacebar broke, so I gotta send it back and get a new one, but it was great while it worked, so it's not in this video, and I got the 8BitDo retro mouse So altogether it's a proof of concept of a real vintage retro PC with internet and it works now My end goal is to make this into production and SELL it, not for money but more for fun because I think it should exist, just like Palmer Luckey's ModRetro Chromatic Gameboy-remake I'd go with something smaller than this 19" checkmate display though, it's great but too big for the Windows 3.11 era I think, more like 14" with a matching sized retro IBM-style keyboard and Windows 3.11 preinstalled (if legally allowed by Satya Nadella) with WiFi working out of the box We could also go with Windows 98/XP, because for me too Windows 3.11 was a very short time, so I too have vague memories of it, most of my computing memories is Windows 98/XP etc. Inside it could simply run on a Raspberry Pi 5 like here, which is easily powerful enough to run all of these OS'es I see many SF founders with Apple II's in their living room, and I guess cause Apple was more popular there back then, but in the rest of the world we were mostly IBM PCs and clones that ran Windows, so it's cool to have that as one retro object working with internet in your living room Anyway, step by step, we're here now 😊

@levelsio

155,824 views • 10 months ago