Benjamin Weinstein

Okay, let's look into this. We want 1) default to be no checkpoint, 2) allow users to add checkpoint. First step is to make a branch with two tests that...

Creating a branch for this issue https://github.com/weecology/DeepForest/tree/model_checkpoint

Tests pass and merged into main.

@ivyleavedtoadflax is there an update to conda env we should make? I think this only a case on colab which has a version of things pre-installed, but which are old....

let me know, what's your serving use case, you want others to upload images to run against the model? Keep me in the loop, i'd be interested in setting something...

I think torchserve is probably the wrong avenue for this. I think this is the perfect use case for AIDE. https://github.com/microsoft/aerial_wildlife_detection You should contact https://twitter.com/b_kellenb who runs that project, he...

I am thinking about shelving this PR. The benefits is that we use the conftest.py correctly, the tests are faster and better organized. The negatives is that it seems somewhat...

Yes, we tried to put a key before, was not successful. On Wed, Feb 9, 2022 at 7:02 AM ethanwhite ***@***.***> wrote: > Is it always just the github API...

I am open to all ideas. In general I am worried about manipulating input data, its computationally heavy (these are going to be huge tiles) and we could lose detail...

@dassaniansh let's use this issue. The first thing we need is the evaluation score (https://deepforest.readthedocs.io/en/latest/Evaluation.html) as a function of input resolution. 1. Download a set of evaluation data. 2. Resample...