Mean Absolute Error (MAE) is a standard and famous evaluation metric, it usually appears in the first class or the Nth class of the machine learning course... it depends on how your teacher arranges the textbook. But to me, it's like Mario's first mushroom monster after starting level 1-1.
Read More »[Machine Learning] Introduction To MAE Metric (With Example)CodeBERT is a pre-trained model based on transformer architecture, it is proposed from CodeBERT: A Pre-Trained Model for Programming and Natural Languages.
Read More »[Machine Learning] CodeBERT Introduction (With Example)
At the end of 2018, the transformer model BERT occupied the rankings of major NLP competitions, and performed quite well. I have been interested in transform models such as BERT, so today I started to record how to use the transformers package developed by HuggingFace.
Read More »[PyTorch] How to Use HuggingFace Transformers Package (With BERT Example)Early stopping is a technique applied to machine learning and deep learning, just as it means: early stopping. In the process of supervised learning, this is likely to be a way to find the time point for the model to converge.
Read More »[PyTorch] Use Early Stopping To Stop Model Training At A Better Convergence TimeIf we have a need to split our data set for deep learning, we can use PyTorch built-in data split function random_split() to split our data for dataset.
PyTorch is a famous deep learning framework. As you can see from the name, it is called using Python syntax.
Read More »[PyTorch] Set Seed To Reproduce Model Training Results
When I first started using the PyTorch framework to gradually learn how to build models, I read and executed many examples from the official tutorials. At that time, I often encountered functions like squeeze() and unsqueeze() in example code, but I didn't quite understand their purposes.
PyTorch a is deep learning framework based on Python, we can use the module and function in PyTorch to simple implement the model architecture we want.
Read More »[PyTorch] Use view() and permute() To Change Dimension ShapeWhen we use RNN network (such as LSTM and GRU), we can use Embedding layer provided from PyTorch, and receive many different length sequence sentence input.
Read More »[PyTorch] How To Use pad_packed_sequence() And pack_padded_sequence() To Adjust Sequence LengthToday, when I was trying to use nn.Embedding() to splice and calculate sentence vectors by myself, I got an error like this on the DataLoader I prepared in advance:
