Issue with pl.Trainer.from_argparse_args(...)

[amitness]

🐛 Bug

To Reproduce

Steps to reproduce the behavior:

1. Use parser = pl.Trainer.add_argparse_args(parser)
2. Run python main.py --overfit_batches 1
3. The training runs over the whole dataset instead of running on a single batch

Code sample

Expected behavior

Only one batch should have run.

Environment

• CUDA:
  • GPU:
    • Tesla P100-PCIE-16GB
  • available: True
  • version: 10.1
• Packages:
  • numpy: 1.18.5
  • pyTorch_debug: False
  • pyTorch_version: 1.6.0+cu101
  • pytorch-lightning: 0.8.5
  • tensorboard: 2.3.0
  • tqdm: 4.41.1
• System:
  • OS: Linux
  • architecture:
    • 64bit
  • processor: x86_64
  • python: 3.6.9
  • version: Proposal for help #1 SMP Thu Jul 23 08:00:38 PDT 2020

Additional context

[williamFalcon]

thanks. looking at it