This presentation took place at R ladies Taipei on July 31, 2023.
Hugging Face can be described as the GitHub for AI models. Initially, it was born to address the difficulty and incompatibility of natural language models. Now, besides open-source AI models for natural language, there are numerous other open-source AI models available on the platform. According to leaked documents from Google, open-source AI models are likely to dominate the development of the entire industry in the future. Therefore, the role of Hugging Face is becoming increasingly important.
This talk is designed for beginners in Hugging Face. It will guide you through using the natural language processing models from Hugging Face.
This presentation took place at COSCUP on July 30, 2023.
Hugging Face can be described as the GitHub for AI models. Initially, it was born to address the difficulty and incompatibility of natural language models. Now, besides open-source AI models for natural language, there are numerous other open-source AI models available on the platform. According to leaked documents from Google, open-source AI models are likely to dominate the development of the entire industry in the future. Therefore, the role of Hugging Face is becoming increasingly important.
This talk is designed for beginners in Chatbot development. It will guide you through using the natural language processing models from Hugging Face and demonstrate step by step how to integrate them into LINE to create a chatbot.
This presentation took place at R ladies Taipei on July 31, 2023.
Hugging Face can be described as the GitHub for AI models. Initially, it was born to address the difficulty and incompatibility of natural language models. Now, besides open-source AI models for natural language, there are numerous other open-source AI models available on the platform. According to leaked documents from Google, open-source AI models are likely to dominate the development of the entire industry in the future. Therefore, the role of Hugging Face is becoming increasingly important.
This talk is designed for beginners in Hugging Face. It will guide you through using the natural language processing models from Hugging Face.
This presentation took place at COSCUP on July 30, 2023.
Hugging Face can be described as the GitHub for AI models. Initially, it was born to address the difficulty and incompatibility of natural language models. Now, besides open-source AI models for natural language, there are numerous other open-source AI models available on the platform. According to leaked documents from Google, open-source AI models are likely to dominate the development of the entire industry in the future. Therefore, the role of Hugging Face is becoming increasingly important.
This talk is designed for beginners in Chatbot development. It will guide you through using the natural language processing models from Hugging Face and demonstrate step by step how to integrate them into LINE to create a chatbot.
遠距教學組成元件
鏘鏘跨入遠距教學
各種遠距平台|目前大家的評價
鏘鏘式混成型遠距教學
同步遠距平台操作實務
Google Meet
MS Teams
Jitsi Meet
遠距教學需要不同的教學方法
實體教案遠距化的個案討論
以STEAM 6E 鷹架進行同步非同步遠距教學與專案製作
類翻轉合作學習經營實務
來談談同步面授
遠距教學組成元件
鏘鏘跨入遠距教學
各種遠距平台|目前大家的評價
鏘鏘式混成型遠距教學
同步遠距平台操作實務
Google Meet
MS Teams
Jitsi Meet
遠距教學需要不同的教學方法
實體教案遠距化的個案討論
以STEAM 6E 鷹架進行同步非同步遠距教學與專案製作
類翻轉合作學習經營實務
來談談同步面授
7. Introduction to GDSC TMU
Chian-Chuen Lee 李芊淳
GDSC Lead 2023
Taipei Medical University
8. 李芊淳 Chian-Chuen Lee
Taipei Medical University
Taipei Medical University
● 牙體技術學系三年級
● GDSC Lead 2023
● GDSC Core Team Member 2022
● 北醫醫資所STEM計畫培訓學員
● TMU 3D printing teaching assistant
● 第三屆 Mifriend微接班人
LinkedIn連結
31. 理念與展望
「Do something together and make it work !」
● 實務應用:提出務實方案,持續發揮影響力
● 跨域合作:多元領域交流,提升專案全面性
● 共同成長:讓成員和業界專家分享經驗和資源
我們希冀 GDSC NTHU 不僅是一個學習與研究技術的社群,
更是激發創業熱情、培養商業直覺和建立成功產品的平台。
GDSC NTHU 社群
77. What Firebase can help you?
From build, to grow & enhance your app
• Develop
• Host your web app
• Store your data
• Enhance
• Make your app stable
• Analyze app usage
• Grow
• Remotely update UI or con
fi
gs
• Run A/B testing
80. Firebase Hosting
Super easy and fast hosting
• Backed by Google’s global network
• Suitable for all kind os front-end frameworks (Angular / React / Vue)
• Free custom domain & SSL support. Tra
ffi
c: 360 MB/day, $0.15/GB
• Need some backend logics?
• Integrate with Firebase Cloud Functions (Node.js)
• Integrate with Cloud Run (Docker-based environment)
84. Cloud Firestore
If you know MongoDB, you know Firestore
• One of the best “serverless database” available in the world
• Support all major platforms across web & app
• Schema-free, store whatever you want
• Built-in o
ffl
ine support, and realtime sync between devices
• Integrate with other Firebase products well
85. import { collection, addDoc } from "firebase/firestore";
try {
const docRef = await addDoc(collection(db, "users"), {
first: "Ada",
last: "Lovelace",
born: 1815
});
console.log("Document written with ID: ", docRef.id);
} catch (e) {
console.error("Error adding document: ", e);
}
Add data via Firestore
86. import { collection, getDocs } from "firebase/firestore";
const querySnapshot = await getDocs(collection(db, "users"));
querySnapshot.forEach((doc)
=
>
{
console.log(`${doc.id}
=
>
${doc.data()}`);
});
Read data via Firestore
89. Firebase Crashlytics
Lightweight, realtime crash reporter
• The leading crash reporting service on Apple, Android, Flutter, and Unity
• Easy to integrate, takes less than 10 mins
• Receive realtime crash report via Email or Slack
• Try to keep 99% crash-free sessions
• Native integration with Android Studio & Google Play
93. Google Analytics (fka Firebase Analytics)
De-facto standard for app/web analytics
• Firebase’s Analytics product now becomes Google Analytics 4
• Google Analytics is the most used analytic app by internet professionals
• The legacy Google Analytics will be end-of-life on July 1, 2023
• You can analyze cross platforms usage and reduce duplications
• Powered by Google Signals
• Di
ff
erent platform installed as “streams” of events into same GA property
99. Firebase Remote Config
Change behaviors or UI without publishing an app update
• It works on native apps and also web!
• Basically it just a remotely fetched key-value dictionary
• With ability to provide di
ff
erent value for di
ff
erent kind of audiences
• We actually use it for “lightweight APIs”
• Banner / copywriting management for marketing
• Dynamically enable features for a speci
fi
c period of time (e.g. LNY)
• Caution: The update might not be real-time. Be careful to add critical logics!
103. Firebase A/B Testing
A/B testing for various Firebase features
• To use A/B testing, you need Google Analytics
• For measure and targeting
• Then Firebase A/B testing will help you
• Split targets into groups
• Run experiments with di
ff
erent values
• Calculate the winning variant
104.
105. Takeaways
So many features covered in this session I know…
• The must haves
• Firebase Crashlytics - Try to make sure you have 99% crash-free sessions
• Google Analytics - De-facto standard for app/web analytics
• Nice to haves
• Firebase Remote Con
fi
g - Save time APIs for marketing or testing
• Firebase A/B Testing - Easy way to do A/B testing right
• Others: Depends on your cases
106. What others Firebase can help you?
Some advanced use cases
• Firebase Messaging
• Send push messages
• Show in-app prompts
• Firebase Authentication
• Let user login with email / phone
• Support Google / Facebook /
GitHub logins
• Firebase Test Lab
• Real devices on cloud for testing
• Firebase Extensions
• Pre-packaged solutions
107. Final remarks
As a student developer, please don’t miss some good things
• GitHub Student Developer Pack - Free domain / cloud / tools
• Google Summer of Code - Learn from open source experts
• Google Developer Student Club - Yes, this event
• Internship - The best chance to get into dream companies
• Graduate level courses - Make you di
ff
erent than others