NELSONZHAO/zhihu — reverse-engineered prompt

Build me a Python Jupyter notebook learning repo for deep learning examples, like a hands on companion to a Chinese Zhihu column. I want it to be easy for readers to open each notebook, run it, and understand what the model is doing without needing a full app.

Include TensorFlow based examples for text generation with LSTM, skip gram word embeddings, lyric generation, a basic Seq2Seq toy task, image denoising with an autoencoder, CIFAR image classification, MNIST GAN, DCGAN image generation, batch normalization experiments, English to French machine translation, an attention plus BiRNN translation demo, sentiment analysis using DNN, LSTM, and CNN, image style transfer, and CTR prediction models like DeepFM, Deep and Cross, xDeepFM, and AutoInt.

Please add clear explanations in Chinese, simple training and prediction cells, sample output images where relevant, and a README that links everything together. Keep it educational and runnable rather than production focused.