Today I used transformers package to build an ALBERT model to process a NLP task in the offline environment, and I got an error message when loading the ALBERT tokenizer:
Read More »[Solved] “TypeError: not a string” at Using AutoTokenizer.from_pretrained()
Today I was testing the simpletransformers that is a well-known python deep learning package. But on my notebook, my training was stuck and not continue.
Read More »[Solved] Simpletransformers Stuck at Training And Not Continue
In the data processing, sometimes we want to replace some column values into others. It may be missing value, or the wrong value…, of course, python have many tools, packages, functions can do the replacement tasks.
Read More »[Python] How To Replace The Value Of Pandas DataFrame
Today I trained the model with simpletransformers package, I got an warning message that never seen:
Read More »[Solved] huggingface/tokenizers: The current process just got forked. after parallelism has already been used. Disabling parallelism to avoid deadlocks
Kaggle Notebook offers a not-so-short GPU-acceleration time every week, and updated every Saturday. Because of this, I recommend using Kaggle Notebook over Google Colab.
Read More »[Machine Learning] Continuously Train the Model on Kaggle Notebook (Can Close the Session)
PyTorch Lightning is a framework that encapsulates native PyTorch in a more advanced level, just like Keras does to Tensorflow (although I remember a lot of backends that Keras can support).
Read More »[Machine Learning] Introduction of ‘pytorch-lightning’ packageI have always wanted to be able to save the optimizer used by PyTorch to train the model, so that can continue training after the model finishes the previous training.
Read More »[PyTorch] Save the Optimizer to Continue Training the Model
aitextgen is a great Python pacakge, it allows user just need to write a few codes to configure the complex AI model. Its architecture is used OpenAI’s GPT-2 and EleutherAI’s GPT Neo/GPT-3, and it can receive pre-trained model to fine-tune.
Read More »[NLP] Use aitextgen Python Package To Generate TextMean 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)
Symmetric Mean Absolute Percentage Error (SMAPE) is a classic evaluation metric for “predicted value and actual value“.
Read More »[Machine Learning] Introduction To SMAPE Metric (With Example)
