Hello! I have recently trained a sagemaker estimator called blazingtext for text classification. I am struggling on how to log the accuracy and loss with wandb.log(). Would love some help with this. Thanks a ton!
I’m glad to hear you’re interested in using W&B Sagemaker. We offer a variety of resources which show how to train a Sagemaker estimator alongside Weights & Biases. I have personally not used the BlazingText offering that they provide so any details/code you can provide will help us provide a more pointed solution for your question.
In the interim let me provide you resources which can make this W&B Sagemaker process easier.
- Docs: https://docs.wandb.ai/guides/integrations/other/sagemaker
- Blog: Optimizing CIFAR-10 Hyperparameters with W&B and SageMaker on Weights & Biases
- Report: Weights & Biases
- Example: examples/text_classification.ipynb at master · wandb/examples · GitHub
I hope this is able to help!
Hello! I was wondering if there was anything else I could do to help here/if the resources I provided were helpful!