Update CHANGELOG

Author: bittremieux

CHANGELOG.md

@@ -8,11 +8,12 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 ### Added
 
-- During training, model checkpoints will now be saved at the end of each training epoch in addition to the checkpoints saved at the end of every validation run.
+- During training, model checkpoints will be saved at the end of each training epoch in addition to the checkpoints saved at the end of every validation run.
+- Besides as a local file, model weights can be specified from a URL. Upon initial download, the weights file is cached for future re-use.
 
 ### Fixed
 
-- Precursor charges are now exported as integers instead of floats in the mzTab output file, in compliance with the mzTab specification.
+- Precursor charges are exported as integers instead of floats in the mzTab output file, in compliance with the mzTab specification.
 
 ## [4.2.1] - 2024-06-25
 

docs/file_formats.md

@@ -2,7 +2,7 @@
 
 ## Input file formats for Casanovo
 
-### MS/MS Spectra
+### MS/MS spectra
 
 When you're ready to use Casanovo for *de novo* peptide sequencing, you can input your MS/MS spectra in one of the following formats:
 
@@ -13,15 +13,17 @@ When you're ready to use Casanovo for *de novo* peptide sequencing, you can inpu
 All three of the above file formats can be used as input to Casanovo for *de novo* peptide sequencing.
 As the official PSI standard format containing the complete information from a mass spectrometry run, mzML should typically be preferred.
 
-### Model Weights
+### Model weights
 
-In addition to MS/MS spectra, Casanovo also optionally accepts a model weights (.ckpt) input file when running in training, sequencing or evaluating mode.
-If no input weight file is provided, Casanovo will automatically use the most recent compatible weights from the official Casanovo github repo, which will be downloaded and cached if they are not already.
-Model weights are retrieved by matching release version.
-If no model weights for an identical release (major, minor, patch), alternative releases with matching (i) major and minor, or (ii) major versions will be used.
+In addition to MS/MS spectra, Casanovo also optionally accepts a model weights (.ckpt extension) input file when running in training, sequencing, or evaluating mode.
+These weights define the functionality of the Casanovo neural network.
+
+If no input weights file is provided, Casanovo will automatically use the most recent compatible weights from the [official Casanovo GitHub repository](https://github.com/Noble-Lab/casanovo), which will be downloaded and cached locally if they are not already.
+Model weights are retrieved by matching Casanovo release version, which is of the form (major, minor, patch).
+If no model weights for an identical release are available, alternative releases with matching (i) major and minor, or (ii) major versions will be used.
 
 Alternatively, you can input custom model weights in the form of a local file system path or a URL pointing to a compatible Casanovo model weights file.
-If a URL is provided the upstream weights file will be downloaded and cached for later use.
+If a URL is provided, the upstream weights file will be downloaded and cached locally for later use.
 See the [command line interface documentation](cli.rst) for more details.
 
 ## Output: Understanding the mzTab format