Experiment runing problem

[DaDaPi3]

I am trying test the code with a json file from 4_20_17. The variable "epoch" is defined within the "train_stages" in this json. But the rastervision/common/options.py uses the epoch variable without retrieving the train_stages. So when I run with "python -m rastervision.run experiments/....", it gives the following error:

root@0a7cd1dcc9ef:/opt/src# python -m rastervision.run experiments/semseg/4_20_17/fcn_0.json
Using TensorFlow backend.
Traceback (most recent call last):
  File "/opt/conda/lib/python3.5/runpy.py", line 170, in _run_module_as_main
    "__main__", mod_spec)
  File "/opt/conda/lib/python3.5/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/opt/src/rastervision/run.py", line 47, in <module>
    run_tasks()
  File "/opt/src/rastervision/run.py", line 34, in run_tasks
    options = make_options(options_dict)
  File "/opt/src/rastervision/options.py", line 15, in make_options
    options = SemsegOptions(options_dict)
  File "/opt/src/rastervision/semseg/options.py", line 16, in __init__
    super().__init__(options)
  File "/opt/src/rastervision/common/options.py", line 18, in __init__
    self.epochs = options['epochs']
KeyError: 'epochs'

Can somebody please help to look at it?

[lewfish]

I just merged #51 that fixes the issue you reported and some other things. I was able to get python -m rastervision.run experiments/semseg/4_20_17/fcn_0.json to run without errors using the following updated experiment file. Remember that experiment files aren't backward compatible, so to get it to run with the develop branch, I made some necessary changes which were

    "problem_type": "semseg",
    "dataset_name": "isprs/potsdam",

I also made various changes so it would run quickly for testing purposes.

{
    "batch_size": 1,
    "problem_type": "semseg",
    "dataset_name": "isprs/potsdam",
    "generator_name": "numpy",
    "active_input_inds": [3, 0, 1],
    "use_pretraining": false,
    "freeze_base": false,
    "train_stages": [
        {
            "epochs": 1,
            "init_lr": 1e-5
        },
        {
            "epochs": 2,
            "init_lr": 1e-6
        }
    ],
    "kernel_size": [5, 5],
    "train_ratio": 0.8,
    "optimizer": "adam",
    "target_size": [256, 256],
    "eval_target_size": [2000, 2000],
    "nb_eval_samples": 1,
    "nb_eval_plot_samples": 1,
    "model_type": "conv_logistic",
    "run_name": "semseg/4_20_17/fcn_0",
    "validation_steps": 1,
    "steps_per_epoch": 1
}

[lewfish]

Thanks for the bug report and sorry for any inconvenience. When we get the time, we'd like to come up with a better testing strategy.