ChatTTS：开源文转音模型

支持中英混合，实测音色稍浑浊，中文部分的断句还不是特别自然。

使用：
1️⃣ git clone 拉下来
2️⃣ pip install vocos vector_quantize_pytorch
3️⃣ 运行文件夹里面的 infer 即可

github.com/2noise/ChatTTS

Today’s my first day of #30NitesOfCode !

100 days felt too long, and I never finished 😅, so the sweet Ellie 🌟 suggested I start with 30 nights.

I began the Intro to Deep Learning with PyTorch course on Udacity and made it up to Logistic Regression.

Visualization ( #immersivespace ) and sonification 🎹 of the GitHub commit history of the PyTorch repository.

me when the homies ask how to write pytorch code that breaks for no apparent reason

🎉Sonunda o gün geldi, yeni kursum yayında!

Bu kurs ile birlikte kendi fotoğraflarınızla kendi görüntü sınıflandırıcınızı eğitip bunu herkesin kullanımına açabileceğiniz ve bir web uygulaması olarak yayınlayabileceğiniz tam teşekküllü bir PyTorch - Gradio uygulaması yapıyoruz.

I'm happy to announce that skorch has reached version 🎊 1.0.0 🎉

This means that skorch is stable and we will avoid any breaking changes (which were already very rare). If you are interested in using PyTorch with an sklearn API, try it out: skorch.readthedocs.io

ml/life journey day 22-23
learned for ~6hrs

> did some Anki decks
> trained micrograd using Pytorch
> started doing makemore by Andrej Karpathy but realised i need to accelerate if I want to implement transformers
> starting to work my way through Attention is all you need

i'm

pytorch unstable 4me again

#Hackment Day 6

Dev - Some more working upon improving the functionality of my Shodan tool!!! and more of learning PyTorch

Hacking - Found an open redirect and CVE 2024-4956 on a target and reported!!!!!!

#BugBounty #Development #Pytorch #infosec

A neat feature of Keras is that its `keras.ops` API follows the NumPy API syntax exactly. And since it runs on JAX, PyTorch, TensorFlow, and NumPy itself, you can use Keras to write GPU-accelerated NumPy code that runs with any framework.

Check out this blog post on how to use the newest MATLAB functions to convert deep-learning models between MATLAB, PyTorch, and TensorFlow 👉 spr.ly/6019eDRvX

#DeepLearning #MATLAB #PyTorch #TensorFlow

📌 Quick fundamental - Pruning in Neural Network with built-in Pytorch util

Simply put, with Pruning we removes weights or biases (parameters) from a neural network. If done right, this reduces the memory footprint of the model, improves generalization, speeds-up inference, and

推荐本优质的开源书籍《Pytorch实用教程》
PyTorch 是一个开源的深度学习框架，它提供了一个灵活、高效的平台，用于构建和训练各种深度学习模型，
《Pytorch实用教程》（第二版）无论是零基础入门，还是CV、NLP、LLM项目应用，或是进阶工程化部署落地，在这里都有。本书的帮助下，读者将能够轻松掌握

Day 38 of ML:
> did a lot of linear algebra today, since I have an exam on the same tomorrow (good for ml)
> made MNIST model on kaggle for the competition (needs some improvements, tomm)
> no pytorch today, but will do it tom
also came across this great vid, check it out

Cya :)

Day 33 of ML:
> Decided to learn pytorch in detail and found this video
> learnt everything about tensors and various functions for mat mul and others (can't build stuff due to bad internet + compute issue)
> Finally ended the day with regular CPP.

Cya ;)

Day 36 of ML:
> almost 10.5 hours into pytorch and learnt about making classification models
> Learnt about 'Sequential'. Had a lot of doubts around how to add activations across each layer but got them
> No cpp today :(
> also stumbled upon a great video by Samson Zhang!

Cya;)

Head of Ted is calling Meta reckless for releasing Open source models.

While its a fact that Meta is really the leader for OSS contributions.

Beyond Llama-3 - we have all the below from them

(and this not an exhaustive list)

- React
- PyTorch
- React Native
- GraphQL
- Jest
-

PyTorch on M1 MacBooks is broken