hsbang09 / iFEED_experiment_result_analysis_2019

1 stars 1 forks source link

Capability to generate a scatter plot showing the feature space #1

Closed hsbang09 closed 5 years ago

hsbang09 commented 5 years ago

Implement the capability to plot a scatter plot showing the features from the feature synthesis task.

Each Subject class instance has a member variable called "feature_synthesis_task_data", which is a python dict. Under it, it contains "features_found" key. Use the features there to generate a scatter plot. The axes of the scatter plot should be metrics[2] and metrics[3], which are specificity (a.k.a. precision) and coverage (a.k.a. recall), respectively.

The function should take one Subject instance or a list of Subject instances as an argument, and generate a scatter plot using matplotlib.pyplot. Features from each subject should be colored differently.

hsbang09 commented 5 years ago

Extra:

Make it possible to pass groups of subject instances as arguments. Then, each group will be colored differently (or use marker) on the scatter plot.

The argument passed to the function maybe a list or a dict. You can choose to implement it in either way:

If it is a list, it would look like this: [[s0, s2, s4, ... ], [s1, s3, s5, ... ]] Or if it is a dict, it would look like this: {"group1_name": [s0, s2, s4, ...], "group2_name": [s1, s3, s5, ...]}

hsbang09 commented 5 years ago

Capability implemented