Sining Sun

Are there any tricks during LAS model training in order to avoid overfiiting?

1

Hi, When I train LAS model using Lingvo, I found that the loss in the training set will not decrease, but the WER in the dev set keeps decrease. When...

training log

1

Hi, Do you still have the log of pre-training? I want to know how the loss changes for every epoch and the accuracy of the positive examples in each batch

Hi Xavier, I read your paper "Acoustic Beamforming for Speaker Diarization fo Meetings". You said that "All possible values range from 0 to 1 given the frequency domain amplitude normalization...

Hi, Thanks for your code. I am using python 1.4.1 and I can build libtorch.a for android in pytorch/build_android/install. However, when I make your code, I meet this error: CMake...

generate函数使用cache出错

3

为了缓存上下文，我在调用generate函数的时候，使用cache，return_dict_in_generate=True，但是传入cache之后，mask的形状不对。 ``` prompt = "Give me a short introduction to large language model." prompt = "给我介绍一下电影《大决战》." messages = [{"role": "user", "content": prompt}] text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) model_inputs = tokenizer([text],...