Releasing v: 0.2 #57
Releasing v: 0.2 #57
Conversation
|
Integrated the Planner. |
| /// Uncompresses the input file/directory | ||
| #[arg(short, action)] | ||
| uncompress: bool, | ||
|
|
||
| /// Verbose output, dumps everysample in the input file (for compression) and in the ouput file (for decompression) | ||
| #[arg(long, action)] | ||
| verbose: bool, |
There was a problem hiding this comment.
maybe we should emit these as a debug!() or trace!() level logs instead.
That way the user can configure the log level to their needs and we avoid introducing multiple logging systems.
There was a problem hiding this comment.
Moving this comment into a new PR with a discussion around this. I would also prefer that route, but, traditional UNIX CLI tools always have a flag for verbosity, that is why I created this one.
My idea is that Logs are more of a developer thing, we enable a certain level of logs to check the internals of the code. A user might want to have a flag to have more insight into the functioning of the tool.
Another option is: Can we fuse the flag with the logging levels?
There was a problem hiding this comment.
You could certainly set the flag to enable a specific log level.
| @@ -22,6 +22,23 @@ pub fn calculate_error(vec1: &[f64], vec2: &Vec<f64>) -> Option<f64> { | |||
| Some(squared_error / min_length as f64) | |||
| } | |||
|
|
|||
| /// Computes the Normalized Mean Square Error between 2 signals | |||
| pub fn nmsqe(original: &[f64], generated: &[f64]) -> Option<f64> { | |||
| if original.len() != generated.len() { | |||
There was a problem hiding this comment.
If this is an error case maybe we should return an Err instead of None here?
There was a problem hiding this comment.
Now that you touch this topic, I think we should panic here. Error calculation is fundamental for getting the compressor settings right. Failing to calculate an error is not recoverable (Which samples should we drop? How did we even get different sizes to start with?).
There was a problem hiding this comment.
Panicking sounds good to me!
There was a problem hiding this comment.
I would leave as an error if there is any possible input to brro that could trigger this case. Otherwise if it's purely internal then panic away.
Releasing 0.2