fix discovery example, make len more stable in DiscoveryModel

Author: levimcclenny

examples/AC-inference.py

@@ -46,6 +46,7 @@ def f_model(u_model, vars, x, t):
 x = X_star[:, 0:1]
 t = X_star[:, 1:2]
 
+print(np.shape(x))
 # append to a list for input to model.fit
 X = [x, t]
 
@@ -54,7 +55,7 @@ def f_model(u_model, vars, x, t):
 
 # initialize, compile, train model
 model = DiscoveryModel()
-model.compile(layer_sizes, f_model, X, u_star, vars, col_weights=col_weights) # baseline approach can be done by simply removing the col_weights arg
+model.compile(layer_sizes, f_model, X, u_star, params, col_weights=col_weights) # baseline approach can be done by simply removing the col_weights arg
 model.tf_optimizer_weights = tf.keras.optimizers.Adam(lr=0.005, beta_1=.95) # an example as to how one could modify an optimizer, in this case the col_weights optimizer
 
 # train loop

tensordiffeq.egg-info/PKG-INFO

@@ -1,12 +1,12 @@
 Metadata-Version: 2.1
 Name: tensordiffeq
-Version: 0.1.6
+Version: 0.1.6.2
 Summary: Distributed PDE Solver in Tensorflow
 Home-page: https://github.com/tensordiffeq/tensordiffeq
 Author: Levi McClenny
 Author-email: [email protected]
 License: UNKNOWN
-Download-URL: https://github.com/tensordiffeq/tensordiffeq/tarball/v0.1.6
+Download-URL: https://github.com/tensordiffeq/tensordiffeq/tarball/v0.1.6.2
 Description: 
         ![TensorDiffEq logo](tdq-banner.png)
 

tensordiffeq.egg-info/SOURCES.txt

@@ -9,6 +9,7 @@ tensordiffeq/models.py
 tensordiffeq/networks.py
 tensordiffeq/optimizers.py
 tensordiffeq/plotting.py
+tensordiffeq/sampling.py
 tensordiffeq/utils.py
 tensordiffeq.egg-info/PKG-INFO
 tensordiffeq.egg-info/SOURCES.txt

tensordiffeq.egg-info/requires.txt

@@ -3,4 +3,4 @@ numpy
 scipy
 tensorflow
 tensorflow_probability
-smt
+pyDOE2

tensordiffeq/models.py

@@ -129,12 +129,13 @@ def predict(self, X_star):
 # WIP
 # TODO DiscoveryModel
 class DiscoveryModel():
-    def compile(self, layer_sizes, f_model, X, u, vars, col_weights=None):
+    def compile(self, layer_sizes, f_model, X, u, var, col_weights=None):
         self.layer_sizes = layer_sizes
         self.f_model = get_tf_model(f_model)
         self.X = X
         self.u = u
-        self.vars = vars
+        self.vars = var
+        self.len_ = len(var)
         self.u_model = neural_net(self.layer_sizes)
         self.tf_optimizer = tf.keras.optimizers.Adam(lr=0.005, beta_1=.99)
         self.tf_optimizer_vars = tf.keras.optimizers.Adam(lr=0.005, beta_1=.99)
@@ -167,7 +168,7 @@ def grad(self):
     @tf.function
     def train_op(self):
         self.variables = self.u_model.trainable_variables
-        len_ = len(self.vars)
+        len_ = self.len_
         if self.col_weights is not None:
 
             self.variables.extend([self.col_weights])