PCA and Plotting Code #4
Conversation
There was a problem hiding this comment.
Generally, seems like the code is functional, but I'm a little confused by some of the hard coded numbers - try to avoid writing these directly if you can.
Also, if you could delete any parts of scripts, or entire scripts you're not using, that'd be useful for clarity. And generally, try not to push files that aren't code, it makes it difficult to keep track of things.
When we meet, let's try to clean up the workspace and repo a little bit, and discuss results!
| env.cfg.init_state.reset_mode = "reset_to_range" | ||
|
|
||
| return env | ||
| args = get_args() |
There was a problem hiding this comment.
Why do we have to start the environment? This is just an analysis script from excel data right?
| args = get_args() | ||
| env = setup(args) | ||
|
|
||
| #import data |
There was a problem hiding this comment.
Same thing here, probably can delete this so the script runs faster
| interface.update(env) | ||
| #print(env.num_envs) | ||
| #print(env.torques.size()) | ||
| log['dof_pos_obs'] += (env.dof_pos_obs.tolist()) |
There was a problem hiding this comment.
We should probably also log the absolute dof_pos
| phase_env2 = np.empty((0,n)) | ||
| phase_env3 = np.empty((0,n)) | ||
| phase_env4 = np.empty((0,n)) | ||
| phases = [phase_env1, phase_env2, phase_env3, phase_env4] |
There was a problem hiding this comment.
Why do we have separate variables for 4 phase environments?
Includes data logging to npz file (play_ORC.py), pca calculations (skeletonpca.py), and plotting (plot.py).