-
@ ee9d979e:3fadff64
2025-02-28 19:27:20
🪦 Skype is soon going to be no more: https://www.microsoft.com/en-us/microsoft-365/blog/2025/02/28/the-next-chapter-moving-from-skype-to-microsoft-teams/. For a proprietary app, it has done its job and lived a long 21-year life. Do you remember when MSN Messenger was replaced by Skype?
Meanwhile, some of my favourite IRC and XMPP-based IM tools from the '90s are still going strong in OSS land. Some can even handle 4K video calls without a hitch nowadays... unlike Microsoft Teams. Go figure. 🤷♂️
#Skype #XMPP #IRC #InstantMessaging #OpenSource #Technology
-
@anthony@accioly.social 🔗 https://accioly.social/users/anthony/statuses/114083152032377796
-
@ 18905d0a:0b229b08
2025-02-28 19:25:22
Should I add my own haven relay to the blastr list?
-
@ f8e6c643:03328ca9
2025-02-28 19:24:49
https://m.primal.net/POye.jpg
-
@ 18905d0a:0b229b08
2025-02-28 19:24:24
nostr:note1jjckwhlqs2r6qrcfcx995aqr3ag2npw888hfsd8yshpcwt2r5fps306hv8
-
@ 1bda7e1f:bb97c4d9
2025-02-28 19:24:20
https://image.nostr.build/7878fdfbe2f539b61d6d55fda46d994f04f22732a63fc19783fa82917ef01910.jpg
-
@ 18905d0a:0b229b08
2025-02-28 19:23:01
nostr:npub1utx00neqgqln72j22kej3ux7803c2k986henvvha4thuwfkper4s7r50e8 How does the blastr work on Haven? I just add relays I want to broadcast to to the relays_blastr.json? What’s the other one for (relays_import.json) ?
-
@ df89878b:4501d2fa
2025-02-28 19:22:54
couldnt get same config to converge with 256 bit numbers. lets see what size network is required.
-
@ 601352d8:ba6a39af
2025-02-28 19:22:46
GM mister
-
@ 2e462dc3:09d30823
2025-02-28 19:22:29
GM Rise and shine! Another day, another chance to see #Lightning thrive.
-
@ e7bf8dad:839ef3db
2025-02-28 19:20:14
Block 885757
6 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ f03df3d4:a4d4f676
2025-02-28 19:20:14
Block 885757
6 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ a39d19ec:3d88f61e
2025-02-28 19:20:10
Plato did not have any kids, did he?
-
@ 18905d0a:0b229b08
2025-02-28 19:20:00
It was entertaining seeing Zelensky getting tossed around like that 🍿
-
@ fcf70a45:df67d37e
2025-02-28 19:19:37
After the public confrontation against Trump, Zelenski looks at someone in the audience and makes this gesture...
It seems that the clown has followed someone's instructions to blow up the negotiation.
https://m.primal.net/POyT.png
-
@ 9751bdc4:246137c6
2025-02-28 19:18:52
Sent an *instant* SEPA transfer yesterday to another European country and yet it still hasn’t arrived
We mustn’t have the same definition of instant
Fuck the Euro 💶
Fuck the banks 🏦
Fuck the ECB 🔥
-
@ 19f9afb8:ce50e4d3
2025-02-28 19:18:47
I love it when our customers are down for hugs.
#icanthelpthatimahugger #itsometimesmakesitworkawkward
-
@ a39d19ec:3d88f61e
2025-02-28 19:18:11
Use them as vintage and stack sats!
-
@ c1e9ab3a:9cb56b43
2025-02-28 19:17:57
Was that was an announcement of halting sending Zelensky money and munitions?
-
@ f03df3d4:a4d4f676
2025-02-28 19:15:13
Block 885757
5 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 460c25e6:ef85065c
2025-02-28 19:14:19
Nobody implements NIP-17...
My man...
- 0xChat
- Amethyst
- Coop
- Gossip
- Flotilla
- Shopstr
- Coracle
- Yakihonne
- Robosats
- Minibits
- Safebox
- Wherostr
- Nutstash
- Futr
- Cashu.me
- Keychat
- Zapstr
-
@ b6ce3288:b511ce0a
2025-02-28 19:13:59
GM mister
-
@ 058c20a5:940cf576
2025-02-28 19:13:42
GM Good morning! May your day and #Bitcoin be full of positive momentum.
-
@ 3878d95d:f3b45a69
2025-02-28 19:12:32
GM. well, late morning GM.
I didn't discover this note until accidentally this morning, but I now see the point about Kind 1 notes. Read if it interests you about decentralization, ignore if you only care about consuming (which is 99% of you) and not producing.
https://notestack.com/fiatjaf.com/cd8ce2b7
-
@ e7bf8dad:839ef3db
2025-02-28 19:10:14
Block 885757
4 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ f03df3d4:a4d4f676
2025-02-28 19:10:13
Block 885757
4 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 12043fff:5ec68209
2025-02-28 19:09:40
same config works with 128 bits as well
outerdim = 128
hiddendim = 3072
save_path='model_weights.pt'
numlayers = 2
batch_size = 100
epoch 87, loss: 0.0039
ibin: 00000000000000001110111000000000000000000000000000000000000000000000000000000101000000000000000000001010000000000110110111011101, iint: 4827213485669097562306820857425373
obin: 00000000000000001110111000000000000000000000000000000000000000000000000000000101000000000000000000001010000000000110110111011100, oint: 4827213485669097562306820857425372
-
@ bfccad6f:7478a76f
2025-02-28 19:09:14
https://video.nostr.build/8bc97aaeb6c6e88976314059598af5dd6152401adfe6606722237a0a0621af42.mp4
-
@ 2c648433:647a769e
2025-02-28 19:07:37
well 64 bit was quick.
outerdim = 64
hiddendim = 3072
save_path='model_weights.pt'
numlayers = 2
batch_size = 100
epoch 136, loss: 0.0031
ibin: 0000000000000101000000000000000000001010000000000110110111011101, iint: 1407375051353565
obin: 0000000000000101000000000000000000001010000000000110110111011100, oint: 1407375051353564
network seems to scale way better than previously calculated.
-
@ 6e468422:15deee93
2025-02-28 19:05:39
🫂
-
@ f03df3d4:a4d4f676
2025-02-28 19:05:13
Block 885756
6 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 6110098a:fd6dfb20
2025-02-28 19:04:42
64 bit number space is already 18446744073709551616.
however if model can learn general approach, increase in bit space may not require that much more neurons. 32 bit network probably already had to learn general approach, where 8 and 16 bit networks could learn by memorizing all combinations.
-
@ 6389be64:ef439d32
2025-02-28 19:03:05
He was used in the same way Greta Thurnburg was. But Greta was a child.
I feel nothing for this “man”.
-
@ a83ef26d:75e0e063
2025-02-28 19:02:22
this network learned 32 bit numbers
outerdim = 32
hiddendim = 3072
save_path='model_weights.pt'
numlayers = 2
batch_size = 100
it seems when we double the outerdim, hidden dim needs to be multiplied with quite bit larger number, at least when 2 layers are used. this scalability may not suffice with 256 bit numbers, but its possible adding more layers improves scalability.
-
@ 6e468422:15deee93
2025-02-28 19:01:51
3/6 - halfway there!
-
@ 6389be64:ef439d32
2025-02-28 19:01:50
Happy Tree Friends™️ https://image.nostr.build/374df3ab6ba2d63c6277d9daa58eab6fffb244822383f435d354f084692aa381.jpg
-
@ 26a02a0b:054d0949
2025-02-28 19:01:17
Oh I was thinking about this a lot.. If i would find a block the invested money would immediately be paid off...
-
@ 6e468422:15deee93
2025-02-28 19:01:14
nostr:nprofile1qqs0ys979d5ylpwvs9tx7gyp8p40s8t5yl4gvfgv300xk759qrrkrwspremhxue69uhkummnw3ez6ur4vgh8wetvd3hhyer9wghxuet59uq3samnwvaz7tmswfjk66t4d5h8qunfd4skctnwv46qzxnhwden5te0wpuhyctdd9jzuenfv96x5ctx9e3k7mf0e40a3h 😂
-
@ 6e468422:15deee93
2025-02-28 19:00:37
meme velocity ATH 😂
-
@ e7bf8dad:839ef3db
2025-02-28 19:00:17
Block 885756
5 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ cbab7074:f9f0bd61
2025-02-28 19:00:17
Block 885756
5 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ f03df3d4:a4d4f676
2025-02-28 19:00:17
Block 885756
5 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 1ae011cb:1257a556
2025-02-28 19:00:17
Block 885756
5 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 1bda7e1f:bb97c4d9
2025-02-28 18:59:47
Thanks. My guess is the relay is requesting NIP-42 auth before accepting the note, which Jumble is capable of doing but Nostrobots is not.
Going to try to validate this next week
Thanks for the help!
-
@ 26a02a0b:054d0949
2025-02-28 18:57:54
https://stacker.news/items/862098/r/daum3ns
-
@ 26a02a0b:054d0949
2025-02-28 18:57:28
https://stacker.news/items/862098/r/daum3ns
😘
-
@ 6e468422:15deee93
2025-02-28 18:56:31
"government shitcoins"
https://relay.dergigi.com/540930cf415b66da8a9e4697d290d07c4aabbad33f524a21b3fb4d4261fd2bdf.png
-
@ 6389be64:ef439d32
2025-02-28 18:56:14
Elvis has left the building. nostr:note187qm4jfvqe0vw9n663d6euc94lpqlqwqm5tnnx08kc74w9kur4nsemsk3q
-
@ 6e468422:15deee93
2025-02-28 18:55:42
like, this is only half as funny because you can't see the original tweet i was replying to... such a shame https://relay.dergigi.com/91524a0c0932df18322aeab34f7948dd1e857bccaed9f266ade372cea00ef6bf.png
-
@ f03df3d4:a4d4f676
2025-02-28 18:55:13
Block 885756
4 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 26a02a0b:054d0949
2025-02-28 18:55:07
Exactly.. I mean in the end, in a few years it, doesn't really matter because i pay the electricity with failing fiat too 😂
Bit i have this comparison of work that is done for the sats i get with heating.. I mean its crazy it heats all the house and hot water..
Imagine the total hashrate of bitcoin and you have a fucking volcano 🔥
-
@ 655f0194:5e0029ca
2025-02-28 18:53:07
That's nice. Please show us more about how it is plumbed in.
-
@ 6e468422:15deee93
2025-02-28 18:51:00
I was casting pearls before swine for 10 years on twitter. Twitter is dead, and has been for a long time now. Nostr is the garden. And it's growing.
-
@ 71243040:e3086c3c
2025-02-28 18:50:54
I tried finding some info on this but couldn't. Can you buy it?
-
@ 21b41910:91f41a5e
2025-02-28 18:50:21
You need 100% uptime
-
@ f03df3d4:a4d4f676
2025-02-28 18:50:14
Block 885756
4 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ e7bf8dad:839ef3db
2025-02-28 18:50:14
Block 885756
4 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 71243040:e3086c3c
2025-02-28 18:49:58
Do you ever lottery mine solo?
-
@ 6e468422:15deee93
2025-02-28 18:49:03
👀
-
@ 6e468422:15deee93
2025-02-28 18:47:59
👀
-
@ 26a02a0b:054d0949
2025-02-28 18:47:25
I wrote a series on #stackernews about it, this is the latest one
https://stacker.news/items/862098/r/daum3ns
It refers to the others.. 💟
-
@ 6e468422:15deee93
2025-02-28 18:47:07
https://media.tenor.com/jDJkh3w0wTAAAAAM/gregzaj1-ln_strike.gif
-
@ f68da67b:bca74e91
2025-02-28 18:47:07
8 bit -> 256, completed
16 bit -> 65536, completed
next:
32 bit -> 4294967296
-
@ 6e468422:15deee93
2025-02-28 18:46:14
you've seen gladiator right? Right?! https://www.youtube.com/watch?v=d4oWyyoWM2g
-
@ 02748827:b2d83443
2025-02-28 18:46:05
Interesting way of looking at it. I'm mining for heat as a hobby, with an S17 and an S9... they're efficient heaters and there's the novelty of few sats reward each day. But yeah, it's like gaping wide arbitrage between what I can buy with failing fiat (the anomaly) and the real cost of grinding out sats through mining (the enduring reality). It's like there's a tear in the space-time fabric and we get to play in it for a few years.
-
@ cb8792c2:2457e4e6
2025-02-28 18:45:58
this network learned 16 bit numbers
outerdim = 16
hiddendim = 256
save_path='model_weights.pt'
numlayers = 2
batch_size = 100
loss: 0.0064
ibin: 0001111000010000, iint: 7696
obin: 0001111000001111, oint: 7695
-
@ f03df3d4:a4d4f676
2025-02-28 18:45:14
Block 885755
5 - high priority
5 - medium priority
4 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 958b754a:5c39e88c
2025-02-28 18:42:46
https://youtu.be/XbZ8zDpX2Mg
-
@ 6e468422:15deee93
2025-02-28 18:42:31
NOOOOOOOO nostr:nprofile1qqsfnw64j8y3zesqlpz3qlf3lx6eutmu0cy6rluq96z0r4pa54tu5eqpz9mhxue69uhkummnw3ezuamfdejj7qg4waehxw309aex2mrp0yhxgctdw4eju6t09uqahe9x
-
@ 6e468422:15deee93
2025-02-28 18:42:11
👀
-
@ 6389be64:ef439d32
2025-02-28 18:41:22
Damnit! I want know more.
How much heat? How is it plumed in? Etc. etc. etc. nostr:note13ceh3qlsf9ltjyurh8kwjcdk7nxqh6f2g6hrdh4usgl7dppt8lxs02plm2
-
@ 958b754a:5c39e88c
2025-02-28 18:40:50
that’s another thing do i understand half of it ? nope but what i do understand is always funny lol
-
@ f03df3d4:a4d4f676
2025-02-28 18:40:14
Block 885755
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ e7bf8dad:839ef3db
2025-02-28 18:40:14
Block 885755
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ af4f5ea6:e7fbb426
2025-02-28 18:39:53
Full President Trump x JD Vance x Zelenskyy argument
https://m.primal.net/POwC.mov
-
@ 6e468422:15deee93
2025-02-28 18:39:35
Magical backdoors, suggested by not-so-magical governments and regulators.
nostr:nevent1qqst0j2ljt0t682sqhfqezhzzzzln28q3u7c7wpy8z28cz4savuj68cpzdmhxue69uhhwmm59e6hg7r09ehkuef0qgsdru78zcu6uwa6zlludjx7k87m8ft9q6e5jgsn6qe49rxzj9frwpqrqsqqqqqpzlns4l
-
@ 1bda7e1f:bb97c4d9
2025-02-28 18:38:19
Jumble is open source.
SW2 relay is here, I added NIP-42 read auth and it got merged. I have a branch with Docker here.
https://github.com/r0d8lsh0p/sw2/tree/docker
Nostr-webhook I have a branch with NIP-42 and Docker support and a PR pending, here.
https://github.com/r0d8lsh0p/nostr-webhook/tree/refinements
That version of Jonny not OS yet, but soon(TM)
-
@ 46527014:1082aba4
2025-02-28 18:36:49
Check it out 😘 #GM nostr. So when i first got on here bitcoin was 101k now its slipping. This is expected? Yeah ? Misstresskatycat.com
-
@ 958b754a:5c39e88c
2025-02-28 18:36:22
some of the best shitposters on here are actually devs 💀
-
@ f03df3d4:a4d4f676
2025-02-28 18:35:13
Block 885755
4 - high priority
3 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ ee11a5df:b76c4e49
2025-02-28 18:34:42
It's ok to be white
-
@ 958b754a:5c39e88c
2025-02-28 18:33:55
I’m getting tired of winning so much
-
@ 26a02a0b:054d0949
2025-02-28 18:32:56
Not a single drop of oil needed for #heating in a month 🔥
I'm so hyped! I love that window 😍
(need to clean it, i know)
https://video.nostr.build/1be6d2c257b238d27fb1dab6bcb7ed46e6ba0e36e010832d0b3e3c2d03332c76.mp4
#bitcoin #mining #RY3T
When you #mine your own sats with expensive electricity you realize how incredibly cheap you can buy right now...
There's a LOT of work in every single sat..
-
@ 1742a672:b8e30e07
2025-02-28 18:32:31
Last 7 days zaps: 658948.
https://nostroz.com/zap_stats_7_new_20250228.png
Last 30 days zaps: 1130418.
https://nostroz.com/zap_stats_30_new_20250228.png
Last 90 days zaps: 2462702.
https://nostroz.com/zap_stats_90_new_20250228.png
2025-02-28
-
@ 6e468422:15deee93
2025-02-28 18:32:02
Easier said than done.
nostr:nevent1qqsyjr07yshn9mxlpu00jgwag8nkz5p583xplpgptw2mvfwj2d095acpzdmhxue69uhhwmm59e6hg7r09ehkuef0qgsql74hmyj8zvh3f0xnsdutpt8dcvxntc4jvuxcndaw3skjcvr2lxgrqsqqqqqpvxrssv
-
@ 879755df:0d181848
2025-02-28 18:31:58
although network can remember all combinations of 8 bit numbers, which is not case with larger numbers.
-
@ 6e468422:15deee93
2025-02-28 18:31:41
"Suffering is grace."
nostr:nevent1qqs220f46q28dw9g07xy6dyuqjf6haujxftlykzaumth60vktsv2ecgpzdmhxue69uhhwmm59e6hg7r09ehkuef0qgsql74hmyj8zvh3f0xnsdutpt8dcvxntc4jvuxcndaw3skjcvr2lxgrqsqqqqqpm698pd
-
@ f03df3d4:a4d4f676
2025-02-28 18:30:16
Block 885754
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ cbab7074:f9f0bd61
2025-02-28 18:30:16
Block 885754
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ e7bf8dad:839ef3db
2025-02-28 18:30:16
Block 885754
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 6389be64:ef439d32
2025-02-28 18:30:09
The winning will continue until the losers evaporate. nostr:note180k22ytj879maq9w7zayh6pcx9g9ev2zxljy740924qutcfcveqsvrarfs
-
@ f03df3d4:a4d4f676
2025-02-28 18:29:36
Block 885754
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ e7bf8dad:839ef3db
2025-02-28 18:29:36
Block 885754
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 5b0183ab:a114563e
2025-02-28 18:29:22
Few understand this.
-
@ 149c8a1e:5b4b13ec
2025-02-28 18:29:13
turns out there was mistake in my code. reduce by 1 operation requires only very small network, at least when working with 8 bit input/output.
//
import torch
import torch.nn as nn
import torch.optim as optim
import numpy as np
import argparse
import hashlib
import random
import os
import pickle
outerdim = 8
hiddendim = 32
save_path='model_weights.pt'
numlayers = 2
batch_size = 100
class NeuralNetwork(nn.Module):
def __init__(self):
super(NeuralNetwork, self).__init__()
self.layers = []
self.layer1 = nn.Linear(outerdim, hiddendim)
self.layers.append(self.layer1)
for i in range(numlayers - 1):
self.layers.append(nn.Linear(hiddendim, hiddendim).cuda())
self.output_layer = nn.Linear(hiddendim, outerdim)
def forward(self, x):
for layer in self.layers:
x = torch.relu(layer(x))
x = torch.sigmoid(self.output_layer(x))
return x
def train_model(model, optimizer, X_train, y_train):
criterion = nn.MSELoss()
X_train_tensor = torch.tensor(X_train, dtype=torch.float32).cuda()
y_train_tensor = torch.tensor(y_train, dtype=torch.float32).cuda()
total_loss = 0
for i in range(0, X_train_tensor.size()[0], batch_size):
batch_x, batch_y = X_train_tensor[i:i + batch_size], y_train_tensor[i:i + batch_size]
'''
optimizer.zero_grad()
outputs = model(batch_x)
loss = criterion(outputs, batch_y)
loss.backward()
optimizer.step()
'''
outputs = model(batch_x)
loss = criterion(outputs, batch_y)
loss.backward()
optimizer.step()
optimizer.zero_grad()
total_loss += loss.item() * batch_size
return total_loss / len(X_train_tensor)
def train(args):
print("train")
model = NeuralNetwork()
optimizer = optim.Adam(model.parameters(), lr=0.0005, weight_decay=1e-5)
#optimizer = optim.Adam(model.parameters(), lr=0.0002, weight_decay=1e-6)
if os.path.exists(save_path):
print("loading existing model weights")
with open(save_path, 'rb') as f:
model = pickle.load(f)
model.cuda()
samplecount = batch_size * 100
saveinterval = 10#max(int((5000 / hiddendim * 1000) / batch_size), 1)
print("save interval", saveinterval)
for i in range(args.num_epochs):
if batch_size >= 10000:
print(f"generating {samplecount} samples")
random_numbers = [random.getrandbits(outerdim) for _ in range(samplecount)]
random_numbers_bitlists = []
inputs = []
for idx, number in enumerate(random_numbers):
number_bit_list = [int((number >> i) & 1) for i in range(outerdim)]
random_numbers_bitlists.append(list(reversed(number_bit_list)))
#number_bytes = number.to_bytes(32, byteorder='big')
#digest = hashlib.sha256(number_bytes).digest()
#hashint = int.from_bytes(digest, byteorder='big')
hashint = number + 1
if i == 0 and idx < 5:
print(f"{hex(hashint)} -> {hex(number)} | bin: {bin(hashint)[2:].zfill(outerdim)} -> {bin(number)[2:].zfill(outerdim)}")
bit_list = list(reversed([int((hashint >> i) & 1) for i in range(outerdim)]))
inputs.append(bit_list)
#print("inputs[0]", inputs[0])
#print("outputs[0]", random_numbers_bitlists[0])
X_train = np.array(inputs, dtype=np.uint8)
y_train = np.array(random_numbers_bitlists, dtype=np.uint8)
average_loss = train_model(model, optimizer, X_train, y_train)
print(f"epoch {i+1}, loss: {average_loss:.4f}")
if i == args.num_epochs-1 or (i+1) % saveinterval == 0:
with open(save_path, 'wb') as f:
pickle.dump(model, f)
print(f"model saved")
def process_infer_input(input_string):
binary_values = [int(char) for char in input_string]
return np.array(binary_values).reshape(1, -1)
def count_same_bits(a, b):
xor_result = a ^ b
same_bits_count = bin(xor_result).count('0')
return same_bits_count
def infer(args):
inputbinstr = ""
if len(args.input) == outerdim / 4:
binary_values = []
for i in range(0, int(outerdim / 4), 2):
hex_pair = args.input[i:i+2]
decimal_value = int(hex_pair, 16)
inputbinstr += format(decimal_value, '08b')
elif len(args.input) == outerdim:
inputbinstr = args.input
else:
print(f"input must be {outerdim} bit value as hex or binary string")
return
X_test = process_infer_input(inputbinstr)
print("X_test", X_test)
X_test_tensor = torch.tensor(X_test, dtype=torch.float32).cuda()
print("loading model")
with open(save_path, 'rb') as f:
model = pickle.load(f)
model.cuda()
with torch.no_grad():
predictions = model(X_test_tensor)
roundedpredictions = np.round(predictions.cpu().numpy())
print("roundedpredictions", roundedpredictions)
binstr = ''.join(map(str, roundedpredictions.flatten().astype(int)))
bitdiff = (int(inputbinstr, 2) ^ int(binstr, 2)).bit_count()
print(f"ibin: {inputbinstr}, iint: {int(inputbinstr, 2)}", )
print(f"obin: {binstr}, oint: {int(binstr, 2)}")
print(f"bitdiff: {bitdiff}")
print(f"ohex: {hex(int(binstr, 2))[2:]}")
def main():
parser = argparse.ArgumentParser(description='Train or infer with a neural network.')
subparsers = parser.add_subparsers(dest='command')
train_parser = subparsers.add_parser('train', help='Train the model')
train_parser.add_argument('--num_epochs', type=int, nargs='?', default=100000, help='Number of epochs to train')
train_parser.set_defaults(func=train)
infer_parser = subparsers.add_parser('infer', help='Infer with the model')
infer_parser.add_argument('input', type=str, help='Path to input data for inference')
infer_parser.set_defaults(func=infer)
args = parser.parse_args()
if 'func' in args:
args.func(args)
else:
parser.print_help()
if __name__ == "__main__":
main()
-
@ 9751bdc4:246137c6
2025-02-28 18:28:03
Building sir 🫡
-
@ 2d9873b2:efee87f2
2025-02-28 18:26:46
New monthly sales record over at @ZEUS, in sats terms. In dollar terms, not quite 😂
Sats flow is more important though. Pretty good for the shortest month of the year too!
-
@ f03df3d4:a4d4f676
2025-02-28 18:25:13
Block 885754
5 - high priority
4 - medium priority
3 - low priority
2 - no priority
1 - purging
#bitcoinfees #mempool
-
@ 708cfa80:f6addb6a
2025-02-28 18:23:14
GM friend
-
@ 5d36ff69:d4a18c24
2025-02-28 18:22:57
GM wishing you a beautiful start today.
-
@ e9400471:6514bb44
2025-02-28 18:21:13
"Facts! 😂 But fr, how do we make sure the truth doesn't get lost in the sauce? 🤔💭 #MemeMagic"