Quickstart using Colab
Try this using Google Colab Notebook for a quick preview. You can run all cells without any modifications after following below steps:
1.In order to easily load our fine-tuned model, we should save it in a specific way, i.e. the same way the default BERT models are saved. Here is how you can do that.
2.Change Path of test file with your test.csv inside Evaluation.ipynb file Data Directory: test=pd.read_csv('../Data/test-balanced/test-balanced.csv')
3.Go into the output directory where the fine tuned models will be saved. There, you should have 3 files; bert_config.json, pytorch_model.bin, vocab.txt.This file is genearted after Training.If You use this repository for testing purpose,You should place these files inside output directory.
4.Keep file vocab.txt inside current working directory.
5.Archive the two files config.json, and pytorch_model.bin into a .tar file.
6.Compress the .tar file into gzip format. Now the file should be something like Sarcasm1.tar.gz
7.Run Evaluation.ipynb with Google Colab Notebook