沧海桑田

duplicate code

于老师好！这个问题不重要，不问了

2

祝好！

首先SelectorCode32 equ LABEL_DESC_CODE32 - LABEL_GDT 下面有一句： 不难理解，当TI和RPL都为零时，选择子就变成了对应描述符相对于GDT基址的偏移，就好像我们程序中那样

bochs运行pmtest1-3就pmtest1第一次正常，然后就都出no bootable device

1

先编成bin，然后dd写入1.44的img，在bochsrc里设置，然后运行bochs

What is the model architectural difference between transductive GCN and inductive GraphSAGE?

4

Difference of the model design. It seems the difference is that GraphSAGE sample the data. But what is the difference in model architecture. Thank you very much. @tkipf

It seems self-attention is a full connected GCN. What can GCN improve to the final performance, if I combine both self-attention(BERT) and GCN to do tasks like text classification or...

I'm new to LeakGAN or SeqGAN or TextGAN. I know GAN is to generate text and let discriminator un-judge-able to real text and gen-text. LM(language model) is the task of...

run pytorch’s run_wt103_large.sh print 285170506 parameters, but the paper is 128M, and OOM.

1

RuntimeError: CUDA out of memory. My GPU is 11441MiB. How to reproduce 128M-model? Thank you @kimiyoung @zihangdai

How does ChatGPT support languages except English?

4

How does OpenAI achieve multi-language support? How did OpenAI do so many human-labeling works? I guess OpenAI is using all the world users' feedback data.

The output of TFAutoModel-save_pretrained and keras-ModelCheckpoint do not equal.

2

### Describe the bug ``` history = model.fit( tf_train_dataset, validation_split=0.01, epochs=int(training_args.num_train_epochs), callbacks=callbacks, ) model.save_pretrained(checkpoint_local) ``` output: `h5` file ``` callbacks = [tf.keras.callbacks.ModelCheckpoint(checkpoint_local)] history = model.fit( tf_train_dataset, validation_split=0.01, epochs=int(training_args.num_train_epochs), callbacks=callbacks, )...