James Adams

As with the OpenImages module (`cvdata.openimages`) we should provide a module that allows for retrieval of images and corresponding annotations from the [YouTube-BoundingBoxes Dataset](https://research.google.com/youtube-bb/). This will likely involve the extraction...

Create a dataset API that allows a user to specify - [ ] download source(s) (OpenImages, COCO, private repo, etc.) - [ ] labels - [ ] resize dimensions -...

images: https://sci2s.ugr.es/sites/default/files/files/TematicWebSites/WeaponsDetection/BasesDeDatos/WeaponS.zip annotations (in PASCAL VOC format): https://sci2s.ugr.es/sites/default/files/files/TematicWebSites/WeaponsDetection/BasesDeDatos/WeaponS_bbox.zip

Add KITTI format as an output option for OpenImages download

1

Add KITTI as a supported output annotation format (currently only PASCAL VOC is supported).

Develop a simple RNN/LSTM model using Keras

1

Begin development of a notebook that includes an RNN/LSTM model implemented using Keras.

Create a Pipeline for managing machine learning workflow

3

Develop an initial [Pipeline](http://scikit-learn.org/stable/modules/generated/sklearn.pipeline.Pipeline.html) for managing machine learning workflow.

We currently expect data for prediction to match the shape of the fit inputs, i.e. there is currently no allowance for batching in the model(s).

We've been doing scaling erroneously using a scaler for each lat/lon location, rather than a scaler per feature/label.

Create a simple CNN with Keras to evaluate its performance and predict variable values.

Create PBS batch processing script for running a ML model using CAM inputs/outputs. Script template: ``` #### PBS preamble #PBS -N PBS_test_script #PBS -M [email protected] #PBS -m abe #PBS -A...