@toqee2 Profile picture

Muhammad Toqeer

@toqee2

Data Scientist , FASTIAN

Joined April 2014
Similar User
Sera🧚🏽‍♀️ of the soft life✨. photo

@bellaphilia

JD Solanki ― WrapWithAI ✨ photo

@me_jd_solanki

Urszula photo

@cyber_ula

kadir1990 photo

@cryptoko6

Maqsood Ali photo

@maqsood132333

Cyrus photo

@CyrusNChege

Ramansh Sharma photo

@ramanshsharma1

The Thug U Luv To Hate photo

@LifeOfDecent

chan_kaku photo

@chan_kakuz

Alex Southgate photo

@fromagie

Lily Chang photo

@LilyLChang

SizzlerWA photo

@SizzlerWA

Fiza photo

@Fiza30894889

Shirai Ryu | Q.A photo

@Quayex

Muhammad Toqeer Reposted

تحریک انصاف ایمرجنسی درخواست‼⚠️ گوال منڈی کی بدبودار مجرمہ کے ٹویٹ پر میں نے کمنٹ کیا ان کے کچھ پالتو کتے میرے ID کو رپورٹ کر رہے ہیں اپ لوگ نیچے کمنٹ اور ری ٹویٹ کرتے جائیں رکنا نہیں, نہیں تو اکاؤنٹ سسپینڈ ہو جائے گا


Muhammad Toqeer Reposted

If Pakistan’s elections are held under the current circumstances, “they would be a disaster,” the former prime minister warns. “The only viable way forward for Pakistan is fair and free elections,” he argues in a guest essay econ.st/48jp5Nu


Muhammad Toqeer Reposted

🔥The WizardLM-30B V1.0 model surpasses OpenAI's code-cushman-001 and the best open-source code-LLM StarCoder on HumanEval. For more details, please refer to: github.com/nlpxucan/Wizar…

WizardLM_AI's tweet image. 🔥The WizardLM-30B V1.0 model surpasses OpenAI's code-cushman-001 and the best open-source code-LLM StarCoder on HumanEval.

For more details, please refer to:
<a style="text-decoration: none;" rel="nofollow" target="_blank" href="https://t.co/4czrcBYJbr">github.com/nlpxucan/Wizar…</a>

Muhammad Toqeer Reposted

🎲 Interested in using Monte Carlo simulations for predicting stock prices? Check out this step-by-step tutorial from Mats Stellwall where he creates a Streamlit app with #Snowpark to do exactly that. 📈 📖 Learn more: blog.streamlit.io/monte-carlo-si… #python #stocks @SnowflakeDB


Muhammad Toqeer Reposted

Okay everyone, my latest LORA for 7B is uploaded. Llama-DEUS-v3-7b is trained on: -GPTeacher -RoleplayV2 (unreleased) -WizardLM Uncensored -GPT4-LLM Uncensored -CamelAI's Bio/Phys/Chem/Math -Codealpaca More info and DL here: huggingface.co/teknium/llama-…


Muhammad Toqeer Reposted

Learn how to run PyTorch models for language identification from audio samples using the Intel®️ Extension for PyTorch. Now up on our blog: hubs.la/Q01PWBPJ0

PyTorch's tweet image. Learn how to run PyTorch models for language identification from audio samples using the Intel®️ Extension for PyTorch.

Now up on our blog: <a style="text-decoration: none;" rel="nofollow" target="_blank" href="https://t.co/mR8zKS0jLu">hubs.la/Q01PWBPJ0</a>

Muhammad Toqeer Reposted

کچھ نہیں ، بس لمبر ون کا سوفٹویئر اپڈیٹ کرکے آیا ہوں 🔥🔥🔥


Muhammad Toqeer Reposted

🚀 @LangChainAI JS/TS 0.0.73 out now ⛓️ 2 new chains: MultiRetrievalQAChain and MultiPromptChain 📄 1 new output parser: CustomListOutputParser 📦 1 new Chat Message History: DynamoDB 🕸️ Various bug fixes and improvements 🧵


Muhammad Toqeer Reposted

This is just 🤯 Aaron showed how we can combine LLMs with eye-tracking tools. So we can ask questions to LLM about the object we're looking at! Imagine extrapolating this to real-life situations with smart glasses.

GPT-4 + Eye Tracking = an AI that knows what you’re looking at. The line where our tools end and we begin continues to blur.



Muhammad Toqeer Reposted

Amazon Prime Video Replaces Microservices with a Simplified Monolith in their Live Stream Monitoring Service, Resulting in a 90% Architecture Cost Reduction. The team accomplished this by: - Pinpointing Overhead in the Distributed System - Combining Tightly-Coupled…

hnasr's tweet image. Amazon Prime Video Replaces Microservices with a Simplified Monolith in their Live Stream Monitoring Service, Resulting in a 90% Architecture Cost Reduction.

The team accomplished this by:

- Pinpointing Overhead in the Distributed System
- Combining Tightly-Coupled…

Muhammad Toqeer Reposted

Today I have finally after much effort cleaning the dataset, uploaded the GPT4 Instruct Code dataset. It's ~4500 GPT-4 Generated Programming Instructions, which may help fine tunes build better code-instruct models. You can find it in my GPTeacher repo: github.com/teknium1/GPTea…


Muhammad Toqeer Reposted

Pinched Nerve Remedies: At-Home Treatments 1. Migraine

CadioArena's tweet image. Pinched Nerve Remedies: At-Home Treatments

1. Migraine

Muhammad Toqeer Reposted

Most projects using agents have used one agent. Or, if theres been many agents, they typically interact in a sequential manner Excited to share a new guest blog post from @cyrusofeden that pushes the boundary - multiple agents acting in parallel blog.langchain.dev/unleashing-the…


Muhammad Toqeer Reposted

Build and Deploy React and Laravel Project in 10 Hours #react #laravel morioh.com/p/a8665978d4e4…


Muhammad Toqeer Reposted

Using DeepSpeed inference can get you nearly 2x the inference speed vs standard PyTorch. Just by wrapping the model in deepspeed.init_inference Left is with DeepSpeed 10 seconds~ for 512 tokens, right is standard fp16 with PyTorch 21 seconds~

abacaj's tweet image. Using DeepSpeed inference can get you nearly 2x the inference speed vs standard PyTorch. Just by wrapping the model in deepspeed.init_inference

Left is with DeepSpeed 10 seconds~ for 512 tokens, right is standard fp16 with PyTorch 21 seconds~
abacaj's tweet image. Using DeepSpeed inference can get you nearly 2x the inference speed vs standard PyTorch. Just by wrapping the model in deepspeed.init_inference

Left is with DeepSpeed 10 seconds~ for 512 tokens, right is standard fp16 with PyTorch 21 seconds~

Muhammad Toqeer Reposted

Did you know, you can now easily try out autonomous agents such as AutoGPT & BabyAGI with LangChain. This allows maximum flexibility: - Easy switching of LLMs - Easy switching of VectorStores - Connectivity to different tools - Connectivity to the LangChain ecosystem

AssemblyAI's tweet image. Did you know, you can now easily try out autonomous agents such as AutoGPT &amp; BabyAGI with LangChain.

This allows maximum flexibility:
- Easy switching of LLMs
- Easy switching of VectorStores
- Connectivity to different tools
- Connectivity to the LangChain ecosystem

Muhammad Toqeer Reposted

OpenAI's speech-to-text API "Whisper" just got supercharged: This tool transcribes audio 70x faster than Whisper A 2-hour podcast can now be transcribed in ~30 seconds using Whisper JAX: The Fastest Whisper API Try it here using your mic: huggingface.co/spaces/sanchit…

AiBreakfast's tweet image. OpenAI's speech-to-text API "Whisper" just got supercharged:

This tool transcribes audio 70x faster than Whisper

A 2-hour podcast can now be transcribed in ~30 seconds using Whisper JAX: The Fastest Whisper API

Try it here using your mic: <a style="text-decoration: none;" rel="nofollow" target="_blank" href="https://t.co/avz4i3vzwE">huggingface.co/spaces/sanchit…</a>

Loading...

Something went wrong.


Something went wrong.