Bingj rebase#1448

Open

marcharper wants to merge 14 commits intodevfrom

bingj-rebase

Commits on Jul 12, 2024

added and implemented epsilon_greedy.py

bing-j
authored and
marcharper
committed
added and implemented epsilon_greedy.py

bing-j
authored and
marcharper
committed
added and implemented epsilon_greedy.py

bing-j
authored and
marcharper
committed
epsilon-greedy added to _strategies.py and reference\strategy_index.rst

bing-j
authored and
marcharper
committed
Fixed update_rewards() to correctly refer to attributes, added test file

bing-j
authored and
marcharper
committed
random test created and passed, basic deterministic tests created and passed

bing-j
authored and
marcharper
committed
strategy tests completed and passedl; updated docstring to clarify initial behaviour.

bing-j
authored and
marcharper
committed
References correctly added and docstring updated with correct Name.

bing-j
authored and
marcharper
committed
for extendability: renamed module to armed_bandits, created Greedy superclass of EpsilonGreedy, and implemented option to weigh reward based on recency

bing-j
authored and
marcharper
committed
modified UNIFORM to use np.inf instead; fixed EpsilonGreedy.strategy() so that it correctly calls update_rewards() through the parent method; all previous tests passed.

bing-j
authored and
marcharper
committed
formatted armed_bandits.py using black.

bing-j
authored and
marcharper
committed
style changes in docstrings and comments, renamed file to bandits, updated doc test, and changed value of UNIFORM constant to -1.

bing-j
authored and
marcharper
committed
modified test file to align with changes. all tests passed!

bing-j
authored and
marcharper
committed
Update classify_strategies.rst
marcharper
authored

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Bingj rebase#1448

Bingj rebase#1448
marcharper wants to merge 14 commits intodevfrom
bingj-rebase

Commits on Jul 12, 2024

added and implemented epsilon_greedy.py

added and implemented epsilon_greedy.py

added and implemented epsilon_greedy.py

epsilon-greedy added to _strategies.py and reference\strategy_index.rst

Fixed update_rewards() to correctly refer to attributes, added test file

random test created and passed, basic deterministic tests created and passed

strategy tests completed and passedl; updated docstring to clarify initial behaviour.

References correctly added and docstring updated with correct Name.

for extendability: renamed module to armed_bandits, created Greedy superclass of EpsilonGreedy, and implemented option to weigh reward based on recency

modified UNIFORM to use np.inf instead; fixed EpsilonGreedy.strategy() so that it correctly calls update_rewards() through the parent method; all previous tests passed.

formatted armed_bandits.py using black.

style changes in docstrings and comments, renamed file to bandits, updated doc test, and changed value of UNIFORM constant to -1.

modified test file to align with changes. all tests passed!

Update classify_strategies.rst