removed collect_results for efficiency

Author: rsexton2

basicrta/cluster.py

@@ -28,36 +28,38 @@ class ProcessProtein(object):
     :type cutoff: float
     """
 
-    def __init__(self, niter, prot, cutoff, gskip):
+    def __init__(self, niter, prot, cutoff, gskip, taus=None, bars=None):
         self.residues = {}
         self.niter = niter
         self.prot = prot
         self.cutoff = cutoff
         self.gskip = gskip
+        self.taus = taus
+        self.bars = bars
 
     def __getitem__(self, item):
         return getattr(self, item)
 
     def _single_residue(self, adir, process=False):
         if os.path.exists(f'{adir}/gibbs_{self.niter}.pkl'):
+            result = f'{adir}/gibbs_{self.niter}.pkl'
             try:
                 result = f'{adir}/gibbs_{self.niter}.pkl'
                 g = Gibbs().load(result)
                 if process:
                     g.gskip = self.gskip
                     g.process_gibbs()
+                tau = g.estimate_tau()
             except ValueError:
                 result = None
+                tau = [0, 0, 0]
         else:
-            print(f'results for {adir} do not exist')
             result = None
-        return result
+            tau = [0, 0, 0]
 
-    def _single_result(self, adir):
-        result = self._single_residue(adir)
         residue = adir.split('/')[-1]
-        self.residues[residue] = result
-
+        setattr(self.residues, 'f{residue}', result)
+        setattr(self.residues.residue, 'tau', tau)
 
     def reprocess(self, nproc=1):
         """Rerun processing and clustering on :class:`Gibbs` data.
@@ -83,30 +85,6 @@ def reprocess(self, nproc=1):
             except KeyboardInterrupt:
                 pass
 
-    def collect_results(self, nproc=1):
-        """Collect names of results for each residue in the `basicrta-{cutoff}`
-        directory in a dictionary stored in :attr:`ProcessProtein.results`.
-        """
-        from glob import glob
-
-        dirs = np.array(glob(f'basicrta-{self.cutoff}/?[0-9]*'))
-        sorted_inds = (np.array([int(adir.split('/')[-1][1:]) for adir in dirs])
-                       .argsort())
-        dirs = dirs[sorted_inds]
-        with (Pool(nproc, initializer=tqdm.set_lock,
-                   initargs=(Lock(),)) as p):
-            try:
-                for _ in tqdm(p.imap(self._single_result, dirs),
-                              total=len(dirs), position=0,
-                              desc='overall progress'):
-                    pass
-                #for adir in tqdm(dirs, desc='collecting results'):
-                #    result = self._single_residue(adir)
-                #    residue = adir.split('/')[-1]
-                #    self.residues[residue] = result
-            except KeyboardInterrupt:
-                pass
-
     def get_taus(self):
         r"""Get :math:`\tau` and 95\% confidence interval bounds for the slowest
         process for each residue. 
@@ -120,17 +98,20 @@ def get_taus(self):
 
         taus = []
         for res in tqdm(self.residues, total=len(self.residues)):
-            if self.residues[res] is None:
-                result = [0, 0, 0]
-            else:
-                try:
-                    gib = Gibbs().load(self.residues[res])
-                    result = gib.estimate_tau()
-                except AttributeError:
-                    result = [0, 0, 0]
-            taus.append(result)
+            taus.append(res.tau)
+            #if self.residues[res] is None:
+            #    result = [0, 0, 0]
+            #else:
+            #    try:
+            #        gib = Gibbs().load(self.residues[res])
+            #        result = gib.estimate_tau()
+            #    except AttributeError:
+            #        result = [0, 0, 0]
+            #taus.append(result)
         taus = np.array(taus)
         bars = get_bars(taus)
+        setattr(self, 'taus', taus[:, 1])
+        setattr(self, 'bars', bars)
         return taus[:, 1], bars
 
     def write_data(self, fname='tausout'):