Natural language processing

Environment info

• transformers version: 4.11.3
• Platform: Linux-4.19.128-microsoft-standard-x86_64-with-glibc2.2.5
• Python version: 3.8.12
• PyTorch version (GPU?): 1.9.1+cu102 (False)
• Tensorflow version (GPU?): not install

Describe the bug
I'm having major trouble with from_csv.

Context: I'm writing tutorial for build simple text search engine with Jina + Hub. I don't want to include a whole section of processing datasets, hence just passing a CSV into from_csv. I tried with meme dataset (converted tsv) before, and now using [superhero dataset](https://www.kaggle.com/jonathanbesomi/superheroes-nlp-datas

Is your feature request related to a problem? Please describe.
I typically used compressed datasets (e.g. gzipped) to save disk space. This works fine with AllenNLP during training because I can write my dataset reader to load the compressed data. However, the predict command opens the file and reads lines for the Predictor. This fails when it tries to load data from my compressed files.

When using the SmoothingFunction's method4 with a a single word as a hypothesis I get the following error. Which is of course because math.log(1) is 0.
I ended up using just method2 or method1 which doesn't use the function that throws the error, but I guess it would be useful to report it, as some other poor soul may