00:00Two tabs, instant predictions, beautiful design. This is your new portfolio star.
00:06I already deployed mine. It's addictive.
00:10Professional, fast, and 100% your code.
00:16Create requirements.txt. Exactly these four lines. This is what Streamlit Cloud needs.
00:23Streamlit, TensorFlow, Pillow, Numpy. That's literally everything.
00:27Pin versions if you want extra safety.
00:33We start with imports and page config. This makes it look professional instantly.
00:39Wide layout plus custom title equals instant wow. Factor.
00:44Set page config. Runs only once. Put it first.
00:50This is day 83 magic. Both models load once and stay in memory forever.
00:55No waiting. No retraining. Instant AI.
01:01At Steak Cash Resource is the secret sauce.
01:05Two tabs. One for images. One for text. Clean and beautiful.
01:11Tabs make it feel like a real product.
01:13Users love simple navigation.
01:18First tab. Image classifier from day 80. Let's make it gorgeous.
01:24I can't wait to upload cat pictures.
01:28Same model you trained weeks ago. Still perfect.
01:33Upload any photo. We open with pill and display instantly.
01:37Selfies work too. I tested.
01:42Convert RGB ensures three channels.
01:47Resize to 32x32 and normalize.
01:50Exactly like training in day 80.
01:54It's so satisfying when it instantly says cat.
01:58Same preprocessing equals same accuracy.
02:01Predict show result with confidence.
02:04Balloons if 95%.
02:06Balloons make it addictive.
02:10NP.argmax plus classlist equals human readable.
02:16Second tab. Sentiment from day 82.
02:19Write anything.
02:21I wrote a love letter to Ethan. It said positive.
02:26Anastasia. Can we make it taller?
02:28We use a super simple tokenizer. Good enough for live demo.
02:34It works shockingly well.
02:37For production, you'd use the real one.
02:40But this is perfect for today.
02:43Predict show sentiment balloons for positive.
02:45Snow for negative.
02:48Snow when it hates my review. So dramatic.
02:52Saintmetric is perfect for this.
02:55Add a cute title and emoji.
02:57Makes it feel alive.
03:00I used robot and sparkles. Perfect combo.
03:04Small details make it portfolio ready.
03:08Sidebar with your name, GitHub, and Ko-fi link.
03:11Looks pro.
03:13I put my Instagram. Recruiters find me.
03:17Great place for credits and version.
03:20I'll write the perfect sidebar.
03:22Always test locally before going public.
03:26Catch bugs early.
03:28I always forget and deploy broken code.
03:31Lesson learned.
03:33Streamlit run app.py instant preview.
03:37I'll run it live.
03:39Watch it work.
03:40Anastasia, run it for me.
03:43New public repo upload app.py plus requirements.
03:47It's .txt plus models, or .gitignore large files.
03:52I always make it public, instant portfolio.
03:56Use .gitignore for models greater than 100 megabytes.
03:59I'll show the perfect repo structure.
04:03Ethan, help me name my repo.
04:06Look, hundreds of you already deployed.
04:08Your AI is officially on the internet.
04:11I'm refreshing Instagram.
04:13So many links.
04:14From day one notebook to day 85 live app.
04:20Incredible journey.
04:22You all crushed it.
04:24Image classification, NLP, saving models, deployment.
04:28You are now full stack AI wizards.
04:32This app is better than most junior dev portfolios.
04:36Professional, clean, and 100% yours.
04:40I'm proud.
04:42You're all legends.
04:43ync.
04:44I'm proud.
04:46You're all tokens.
04:55Amazing.
04:57m Bye.
04:58No.
04:58Beef.
04:59FEMA.
04:59Young.
05:00Beautiful.
05:00Glories.
05:00Let's kalen.
05:01Poll rails.
05:01Not that's right.
05:01You're all wifi.
05:02Now you're right.
05:02What?
05:04You're all gone.
05:04Come onto my one book.
05:05Now you're waiting.
05:06Yeah, that's reallyおお.
05:08What group is..
Comments