search

hehefan / Unsupervised-Person-Re-identification-Clustering-and-Fine-tuning

219 stars 87 forks source link

Queries regarding the Algorithm and Code #7

Open sumo8291 opened 6 years ago

sumo8291 commented 6 years ago

I have two doubts : one regarding the algorithm and the second regarding the code

(1) In Algorithm 1, why do u initialize 'w' randomly in every epoch (step 2)? Shouldn't you initialize 'w' once as it is necessary that there should be atleast one sample in every cluster.

(2) In line 125 of "PUL/unsupervised.py", I was wondering why you are loading only the initial pre-trained model. Shouldn't you be loading the previous finetuned model , i.e. the variable init_model = previous_model?

hehefan commented 6 years ago

Hi @sumo8291: (1) The Algorithm 1: 5 can guarantee there is at least one sample in every cluster.

(2) Our idea is to achieve a reliable training set. For every training set, we train the model from scratch.