From 0a41f491cddc12ac86e2929cdbe4779897ea65a3 Mon Sep 17 00:00:00 2001
From: Muammar El Khatib <muammarelkhatib@gmail.com>
Date: Sun, 19 Jan 2020 12:29:53 -0800
Subject: [PATCH] Use client.submit instead of dask.compute for SVM
 computations Gaussian.

---
 ml4chem/data/visualization.py | 7 +++----
 ml4chem/features/gaussian.py  | 8 +++-----
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/ml4chem/data/visualization.py b/ml4chem/data/visualization.py
index 9ce7a47..7fc7d71 100644
--- a/ml4chem/data/visualization.py
+++ b/ml4chem/data/visualization.py
@@ -5,7 +5,6 @@
 import matplotlib.pyplot as plt
 from sklearn.metrics import mean_squared_error, r2_score, mean_absolute_error
 from ml4chem.data.serialization import load
-import time
 
 
 def parity(predictions, true, scores=False, filename=None, **kwargs):
@@ -30,7 +29,7 @@ def parity(predictions, true, scores=False, filename=None, **kwargs):
 
     min_val = min(true)
     max_val = max(true)
-    fig = plt.figure(figsize=(6.0, 6.0))
+    fig = plt.figure(figsize=(6, 6))
     ax = fig.add_subplot(111)
     ax.plot(true, predictions, "r.")
     ax.plot([min_val, max_val], [min_val, max_val], "k-", lw=0.3)
@@ -284,7 +283,7 @@ def plot_atomic_features(
 
         labels = {str(axis[i]): "t-SNE-{}".format(i + 1) for i in range(len(axis))}
 
-        tsne = manifold.TSNE(n_components=dimensions)
+        tsne = manifold.TSNE(n_components=dimensions, perplexity=5)
 
         tsne_result = tsne.fit_transform(full_ls)
 
@@ -324,4 +323,4 @@ def plot_atomic_features(
     except:
         pass
 
-    return plt
+    return plt, df
diff --git a/ml4chem/features/gaussian.py b/ml4chem/features/gaussian.py
index 6eace27..edffd9c 100644
--- a/ml4chem/features/gaussian.py
+++ b/ml4chem/features/gaussian.py
@@ -365,7 +365,6 @@ def calculate(self, images=None, purpose="training", data=None, svm=False):
 
                 scaled_feature_space.append(features)
 
-            # scaled_feature_space = client.gather(scaled_feature_space)
 
         else:
             scaled_feature_space = []
@@ -397,13 +396,12 @@ def calculate(self, images=None, purpose="training", data=None, svm=False):
 
                 # image = (hash, ase_image) -> tuple
                 for atom in image[1]:
-                    reference_space.append(
-                        self.restack_atom(i, atom, scaled_feature_space)
-                    )
+                    restacked_atom = client.submit(self.restack_atom, *(i, atom, scaled_feature_space))
+                    reference_space.append(restacked_atom)
 
                 feature_space.append(restacked)
 
-            reference_space = dask.compute(*reference_space, scheduler=self.scheduler)
+            reference_space = client.gather(reference_space)
 
         elif svm is False and purpose == "training":
             for i, image in enumerate(images.items()):