In my opinion, if its the sole contribution or novelty of the paper you are planning to publish, then it do not think so it is sufficient enough to publish.
You may make a technical report on it and upload in arXiv. But, if you want to publish, then I think you have to some more analysis and perform experiments to make the novelty of the paper significant.
Another possible direction may be to implement a set of related papers, possibly benchmark models on common datasets and compare their results and provide inferences. Some conferences like ECIR (European Conference on Information Retrieval) have started a “Reproducibility Track” specifically for such studies.
ECIR also strongly encourages the submission of reproducibility papers that repeat and analyze prior work. In particular we solicit classical reproducibility papers, which replicate prior experiments and show how, why, and when the methods work (or not), along with two other types of reproducibility papers: generalizability papers, that focus on assessing how well technology performs in new contexts (e.g., different time, location, access device, task), and predictability papers, that focus on developing theory and methods that assess and evaluate how generalizable methods are and whether they will work in other contexts.ECIR website, Call for papers section
This is my answer on Quora to the question “Can we publish a paper on the implementation of an already published paper (implemented on a different platform)?“
Please comment below if you would like to add something.