Fix newer Gym support

Replace np.bool with bool (makes np > 1.20 usable) (#712)

Add instructions for adding related projects

Dockerfile: Add -dev and -data-processing stages (#575)

* Dockerfile: Add -dev and -data-processing stages

The dev stage includes an install of the latest minerl `dev` commit (upon
Docker image build). The data-processing stage adds packages,
environment variables, and volumes used for processing raw demonstration
data into datasets.

* typos

* simplify examples by using ~/minerl.data

* Dockerfile: Fix incorrect 7z dependency

setup.py: Include requirements-docs.txt as data_file

Otherwise PyPI install fails, because this file is not uploaded
along with the rest of the package.

handers..equip.py: Reduce from_universal warning verbosity (#540)

* handers..equip.py: Reduce from_universal warning verbosity

* even more filtering

Merge pull request #334 from juliusfrost/multiple-environment-test

Multiple environment test

Don't try to download missing files (#316)

* Don't try to download missing files
Fix dataset url path for experiment downloading

* Add SUPER fancy fast mirror selection
Should save the average user over 1 second of time!

MineRL 0.2.0 [rebased] (#136)

* Add ability to observe placement of non-agent placeable blocks

* Fix loud broken pipe error on data loading framework

* Add action_space and observation_space to data pipeline

* Made obsevation space a member of data class

* Add spaces to package

* Add other type to handlers

* test handlers

* Add debug flags to data pipeline to reduce typical printing

* Enable debug flag in pipeline

* use logger log levels

* Expose logger

* Unexpose logger

* Add queue size parameter

* Add debug statements

* debug for days

* debug for days

* Add ability to observe placement of non-agent placeable blocks

* Fix loud broken pipe error on data loading framework

* Add action_space and observation_space to data pipeline

* Made obsevation space a member of data class

* Add other type to handlers

* test handlers

* Add debug flags to data pipeline to reduce typical printing

* use logger log levels

* Expose logger

* Unexpose logger

* Add queue size parameter

* Reveering changes.

* Addding data versioning.

* Updating Malmo for gamma fix.

* Update load_data to use current environment

* Update load_data function

* Align data based on actual number of frames in recording

* A workable stream viewer!

* Making stream viewer a module in minerl.

* Adding a new streamviewier script.

* viewier.

* Windows requirement.txt

* Updating.

* Updating one more time.

* Importing on windows.

* need more color

* Updating your cool applicaton.xtxt

* yeah dude!

* remove debug

* Added version support to download script

* Updating the head to support the new viewer.

* Updating viewer to support reward plots.

* Updating the new data viewer.

* Add ability to return metadata in seq_iter

* Fix metadata

* Fix metadata

* Pushing. code.

* Update load data

* Add metadat if not there

* Fix

* Deprecating seq_iter.

* Increase timeout in download script
closes #131

* Remove teleport observation for final state

* Allowing for random streams.

* Fix minerl data download version

* Fix minerl data download version

* Update documentation for data pipeline

* Update setup.py

* Delete mc_1.log

* Update documentation for data itter

* Updating the documentation.

* Updating Malmo version

* Updating the viewer for the new version of python3.5. Why oh why matplotlib :(

* Adding a dependency for pySmartDL.

* Adding new updates.

* Yes.

MineRL 0.2.0 [rebased] (#136)

* Add ability to observe placement of non-agent placeable blocks

* Fix loud broken pipe error on data loading framework

* Add action_space and observation_space to data pipeline

* Made obsevation space a member of data class

* Add spaces to package

* Add other type to handlers

* test handlers

* Add debug flags to data pipeline to reduce typical printing

* Enable debug flag in pipeline

* use logger log levels

* Expose logger

* Unexpose logger

* Add queue size parameter

* Add debug statements

* debug for days

* debug for days

* Add ability to observe placement of non-agent placeable blocks

* Fix loud broken pipe error on data loading framework

* Add action_space and observation_space to data pipeline

* Made obsevation space a member of data class

* Add other type to handlers

* test handlers

* Add debug flags to data pipeline to reduce typical printing

* use logger log levels

* Expose logger

* Unexpose logger

* Add queue size parameter

* Reveering changes.

* Addding data versioning.

* Updating Malmo for gamma fix.

* Update load_data to use current environment

* Update load_data function

* Align data based on actual number of frames in recording

* A workable stream viewer!

* Making stream viewer a module in minerl.

* Adding a new streamviewier script.

* viewier.

* Windows requirement.txt

* Updating.

* Updating one more time.

* Importing on windows.

* need more color

* Updating your cool applicaton.xtxt

* yeah dude!

* remove debug

* Added version support to download script

* Updating the head to support the new viewer.

* Updating viewer to support reward plots.

* Updating the new data viewer.

* Add ability to return metadata in seq_iter

* Fix metadata

* Fix metadata

* Pushing. code.

* Update load data

* Add metadat if not there

* Fix

* Deprecating seq_iter.

* Increase timeout in download script
closes #131

* Remove teleport observation for final state

* Allowing for random streams.

* Fix minerl data download version

* Fix minerl data download version

* Update documentation for data pipeline

* Update setup.py

* Delete mc_1.log

* Update documentation for data itter

* Updating the documentation.

* Updating Malmo version

* Updating the viewer for the new version of python3.5. Why oh why matplotlib :(

* Adding a dependency for pySmartDL.

* Adding new updates.

* Yes.