Cyberpunk original character by permian_polaris
2025.01.27 04:31 Permian_Polaris Cyberpunk original character by permian_polaris
 | This costume is based on a OC character illustration by one of my favorite artists, Neco! For more photos and cosplays, check out my Instagram!: permian_polaris submitted by Permian_Polaris to Cawwsplay [link] [comments] |
2025.01.27 04:31 the613daddy why is the far right supporting Chandra Arya?
MODS can delete if political posts are not allowed đ
submitted by the613daddy to ottawa [link] [comments]
2025.01.27 04:31 NoIntendedHarassment I failed the personality test đ
 | submitted by NoIntendedHarassment to teenagers [link] [comments] |
2025.01.27 04:31 JustBig8300 Sometimes Workers ATAH
This is coming from someone who has worked in food/customer service for years.
I understand that a lot of aspects of a minimum wage job may kill your mood, but what I do not understand is how common it is to blatantly ignore customers. The only subreddits where Iâve found testimonials of these types of experiences were those of couriers, mainly Doordashers and UberEats drivers; however, I know I canât be the only one who notices this just as a customer.
Post-2020, Iâve found it common to encounter workers (either in retail or food) who donât care to acknowledge my existence. I do not think employees have to smile ear to ear or force a whole conversation; how much energy they have is not my concern. It just feels odd to pass an employee as I enter a store, and for them to just blankly stare at me. I canât imagine not greeting a customer if we make eye contact for long enough. I often see the complaint of customers not responding to greetings, and while Iâve experienced it many times at my own jobs, I feel that a basic greeting is just a part of the job.
Even worse than this, Iâve had many moments where I stood waiting to order or be assisted at a fast food restaurant, and not even when itâs busy! Just tonight, I was the only customer in Chipotle (it wasnât close to closing time either) and at least one worker behind the counter saw me enter, but they continued on with their conversation. Once I got to the counter to order, I stood patiently while they all kept chatting for at least a minute. The girl who saw me walk in finally looked at me again as if she forgot I was even there, then simply pointed at me to signal another employee to help me.
So many times walking up to a checkout area while the employee has their back to me, or are more concerned with the conversation theyâre having with their coworkers. So many times being silently looked at from 50 ft away when I go to pick up my mobile or online orders.
My anxiety holds me back from getting out of these situations faster, but I feel that all of this apathy for being a decent person towards customers is a problem.
submitted by JustBig8300 to Rants [link] [comments]
2025.01.27 04:31 XavBlade 29 M4F Minnesota Gamer
Hello I'm XavBlade that's my gamertag I go by on alotta gaming platforms. I'm into reading, video games, card games, various forms of music, anime, swimming, biking, walking, and just enjoying the day. I'm 420 friendly and only drink sociallably. I'm working 2 jobs and make time for friends and ppl I consider family. I can be an awkward quiet person at times which I'm a good listener or shoulder to cry on. I can be silly and a goofball the more comfortable I become around you. Expresso milkshake or peppermint white mocha with a cinnamon twist from a local coffee shop are one of my go to once a week lil trip.
If you're a F from 25 to 35 and are into gaming and/or anime let me know whats you're favorite with a dm and let's get to know one another.
submitted by XavBlade to ForeverAloneDating [link] [comments]
2025.01.27 04:31 SmokyBluntz420 My cannabis YouTube channel
I'm growing my cannabis related vlog and need help with the growth of my channel. I'll be posting regularly and will have topic of things I'm passionate about (420 life style, sports and Pro Wrestling) if that sounds good to you please click my link to support my channel. Toke up with me in live streams
https://youtube.com/@smokybluntz4200?si=XT3pXZwCaQSs40kE
submitted by SmokyBluntz420 to Cannabis_Culture [link] [comments]
2025.01.27 04:31 totally1of1 [COD] we were a bit harsh đ rn it's ifft
![[COD] we were a bit harsh đ rn it's ifft](https://preview.redd.it/4pyzi54jsgfe1.jpeg?width=320&crop=smart&auto=webp&s=0a9b2ebf29d9e387ba7d26b7533c089623033394 "[COD] we were a bit harsh đ rn it's ifft") | submitted by totally1of1 to CallOfDuty [link] [comments] |
2025.01.27 04:31 Vic_The_Great Where to get these??
 | Where to get these? submitted by Vic_The_Great to DhGateReps [link] [comments] |
2025.01.27 04:31 Vegetable-Active7585 Optimizing DeepSeek R3 with Affine Transformation
Introducing affine transformations into DeepSeek R3 can significantly enhance feature preprocessing, index construction, query execution, and similarity computation. The overall performance improvement is estimated to be between 25%-50%, with the greatest benefits observed in high-dimensional, large-scale data scenarios. This optimization framework can also be extended to other vector search systems for general performance enhancement.
To optimize DeepSeek R3 with affine transformation, the goal is to improve its overall efficiency by refining feature distribution, simplifying index construction, and accelerating similarity computation. Below is a detailed explanation of how to apply affine transformations to each module of DeepSeek R3 and the expected efficiency improvements.
1. Optimization Objectives
- Data Preprocessing: Refine feature distribution, reducing redundancy.
- Index Construction: Simplify feature indexing and make it more efficient.
- Query Execution: Align query features to the indexed space for faster and more accurate retrieval.
- Similarity Computation: Reduce computational overhead in similarity calculations.
import numpy as np from sklearn.decomposition import PCA # Affine transformation function def affine_transform(features, A, b): """Apply affine transformation: y = Ax + b""" return np.dot(features, A.T) + b # Example: Generate and preprocess features features = np.random.rand(100000, 128) # 100k samples with 128 dimensions pca = PCA(n_components=64) # Reduce dimensions to 64 reduced_features = pca.fit_transform(features) # Define affine transformation parameters A = np.random.rand(64, 64) # Transformation matrix b = np.random.rand(64) # Bias vector # Apply transformation transformed_features = affine_transform(reduced_features, A, b) import faiss # Construct FAISS index index = faiss.IndexFlatL2(64) # L2 distance-based index index.add(transformed_features) # Add transformed features to the index def query_with_affine(query, index, A, b, k=10): """Search nearest neighbors using transformed query features""" transformed_query = affine_transform(query, A, b) distances, indices = index.search(transformed_query, k) return distances, indices # Example: Execute query query = np.random.rand(1, 64) # Single query distances, indices = query_with_affine(query, index, A, b, k=5) print("Top-5 neighbors:", indices) from sklearn.metrics.pairwise import cosine_similarity # Compute optimized similarities cos_sim = cosine_similarity(transformed_features) (2) Index Construction Affine transformations create more compact feature distributions, leading to a 15%-30% reduction in index construction time.
(3) Query Execution With aligned query features, search operations become more efficient, resulting in a 20%-35% improvement in query speed.
(4) Similarity Computation Affine transformations simplify similarity calculations, reducing computational time by 10%-20%.
Overall Performance Gains
| Module | Key Optimization | Expected Improvement |
|---|---|---|
| Data Preprocessing | Feature Standardization & Dimensionality Reduction | 20%-40% |
| Index Construction | Compact Feature Distributions | 15%-30% |
| Query Execution | Aligned Query Features | 20%-35% |
| Similarity Computation | Simplified Metrics | 10%-20% |
| Overall | Full Workflow Optimization | 25%-50% |
- Parameter Optimization: The affine transformation matrix AA and bias vector bb must be tuned to the specific dataset to maximize performance gains.
- Computational Overhead: Ensure that the additional computational cost of applying affine transformations (e.g., matrix multiplication) is outweighed by the overall efficiency gains.
- Applicability: Affine transformations are particularly beneficial for high-dimensional datasets, where dimensionality reduction and feature alignment can lead to substantial improvements.
submitted by Vegetable-Active7585 to DeepSeek [link] [comments]
2025.01.27 04:31 BigBrotherJu ORD to UIUC ride
I can give out a ride on Monday(1 27th) at around 1:30 pm from ORD to school campus. Dm me if you are interested.
submitted by BigBrotherJu to UIUC [link] [comments]
2025.01.27 04:31 AgreeableSentence541 What batch is the best in Kobe 5 Chaos?
submitted by AgreeableSentence541 to KobeReps [link] [comments]
2025.01.27 04:31 ZifferYTAndOnions What did Coffo see?
 | submitted by ZifferYTAndOnions to Sprunki [link] [comments] |
2025.01.27 04:31 ImportantAlarm5069 Hypertension Risks, Short Term and Long Term
For the past week I've noticed my systolic to be quite high 143-148 while my diastolic is normal in 70-80 range. I did go to my doctor and when he checked it was 130/73 but when I check its never that low. This has made me worried about heart attack and stroke more recently. I have been a little stressed lately but also I am very unhealthy and rarely excercise (hoping to change that soon). What I'm really wondering is that are stroke and heart attack risks elevated extremly during short term hypertension like months or is it uncontrolled hypertension over years. I am still very young (barely in 20's) and used to go on walks daily but haven't in the last year. But my diet has always been extremly unhealthy.
submitted by ImportantAlarm5069 to hypertension [link] [comments]
2025.01.27 04:31 dhruan Kaweco KV1 & V12
 | Received this Kaweco KV1 from the 1960s, pictured here with a Kaweco V12 from the same era (wuite a bit of parts commonality between the two). While it is a neat little pen, the barrel sadly has some hairline cracks at the piston knob end that seem to make it not 100% leak proof⌠still a nice example to have as it is in great shape otherwise (and it only cost like 36⏠shipping included). That V12 though⌠bonkers little pen as that âBâ nib is more like a cursive italic BB đ I love it to bits. submitted by dhruan to fountainpens [link] [comments] |
2025.01.27 04:31 Odd-Mathematician488 SUPERMAN Full Movie 2025: Man of Justice | FullHDvideos4me Action Movies 2024 English (Game Movie)
submitted by Odd-Mathematician488 to LUCIFERSTAR [link] [comments]
2025.01.27 04:31 Solko7 [WTS] Steiner P4xi, Streamlight TLR1-HL
Timestamp: https://imgur.com/a/yUniaa0
P4xi serial: https://imgur.com/a/iD73im5
Evening! Letting a couple things go to make room. Prices shipped to the lower 48 via PP F&F or Venmo. Dibs system honored. Please PM, no chat. Thanks!
- Minty LNIB Steiner P4xi - P3TR reticle. Comes with box and manual. Mounted never fired, went with a 1-6x instead. -$500
- Streamlight TLR1-HL - no box, all keys, outstanding condition - $100
2025.01.27 04:31 ThatSaltyTaste1 Honest ratings â¨
 | Kinda doing this one for fun, I know imma specific taste in terms of physical appearance submitted by ThatSaltyTaste1 to GayRateMe [link] [comments] |
2025.01.27 04:31 jpower-27 Insufficient Eth balance on Dackie LP
 | I just saw an aerobud team member posted the new Aerobud/ETH LP on Dackie. I wanted to go add liquidity but for some reason Dackie is telling me I only have .0005 ETH when I have .08. I tried disconnecting/reconnecting my wallet and clearing cache but nothing worked. if anyone knows anything thatâd be great, thanks đđź submitted by jpower-27 to Aerobud [link] [comments] |
2025.01.27 04:31 First-Loss-8540 âKiss Of The Spider Womanâ: World Premiere Of Jennifer Lopez Musical Receives Standing Ovation At Sundance
 | submitted by First-Loss-8540 to oscarrace [link] [comments] |
2025.01.27 04:31 NipponPunk First truck?đ
Yo thinking about getting a truck for the first time and only want the Colorado, is this the superior entry level truck?
Iâm coming from a shitty ford escapeđ
submitted by NipponPunk to chevycolorado [link] [comments]
2025.01.27 04:31 lemkowidmak Eagles Nation @PHLEaglesNation: RT @NickFoles: Had to celebrate a little on Broad Street with Tori! #dadszn #phillyphillydad https://t.co/HN1MEypppt
 | submitted by lemkowidmak to phillysports [link] [comments] |
2025.01.27 04:31 RobDemersArt Sketch cards I created for Star Wars Galaxy 2024 #1 - 20
 | submitted by RobDemersArt to starwarscollecting [link] [comments] |
2025.01.27 04:31 key-tie9871 Kevin Pietersen on Bazball's failure in India
 | submitted by key-tie9871 to CricketShitpost [link] [comments] |
2025.01.27 04:31 barnwater_828 1/26/25 - The USA has a bright light over it since the 2024 election.
 | submitted by barnwater_828 to trumptweets [link] [comments] |
2025.01.27 04:31 Beneficial_Guard819 Precossidade.
Estou tendo dificuldades na hora da penetração, nunca fui um rapaz que demorava horrores, mais eu conseguia satisfazer quem estivesse ao meu lado, e se satisfazer... comecei a namorar uma moça que diz ser apaixonada por mim, jå ficamos vårias vezes antes de começar a namorar, ela sempre ficava louca, e eu gostava bastante, enfim começamos a namorar, nas duas primeiras semanas foi tranquilo, tudo maravilhoso... os dois se satisfazendo muito... mais ela veio me visitar esse fim de semana e eu não conseguia dar 4 bombada... eu fiquei ser chão. Estou desesperado pois fomos tentar denovo... foi 10/15 bombada e jå foi denovo... ela disse que estava tudo bem, que sempre foi tudo perfeito pra ela, e que eu estava muito eufórico e que poderia ser isso... estou querendo procurar um urologista pra ver se tem algo de errado antes de revê-la novamente...
submitted by Beneficial_Guard819 to desabafos [link] [comments]