fixed bugs from the new API

8 files changed, 14 insertions, 14 deletions

diff --git a/src/python/CMakeLists.txt b/src/python/CMakeLists.txt
index bed2b541..e4ac1b48 100644
--- a/src/python/CMakeLists.txt
+++ b/src/python/CMakeLists.txt
@@ -556,6 +556,7 @@ if(PYTHONINTERP_FOUND)
     # Differentiation
     if(TENSORFLOW_FOUND)
       add_gudhi_py_test(test_diff)
+    endif()
 
     # Betti curves
     if(SKLEARN_FOUND AND SCIPY_FOUND)
@@ -596,4 +597,4 @@ if(PYTHONINTERP_FOUND)
 else(PYTHONINTERP_FOUND)
   message("++ Python module will not be compiled because no Python interpreter was found")
   set(GUDHI_MISSING_MODULES ${GUDHI_MISSING_MODULES} "python" CACHE INTERNAL "GUDHI_MISSING_MODULES")
-endif(PYTHONINTERP_FOUND)
\ No newline at end of file
+endif(PYTHONINTERP_FOUND)
diff --git a/src/python/doc/cubical_complex_tflow_itf_ref.rst b/src/python/doc/cubical_complex_tflow_itf_ref.rst
index 692191ba..18b97adf 100644
--- a/src/python/doc/cubical_complex_tflow_itf_ref.rst
+++ b/src/python/doc/cubical_complex_tflow_itf_ref.rst
@@ -19,7 +19,7 @@ Example of gradient computed from cubical persistence
     cl = CubicalLayer(dimensions=[0])
 
     with tf.GradientTape() as tape:
-        dgm = cl.call(X)
+        dgm = cl.call(X)[0][0]
         loss = tf.math.reduce_sum(tf.square(.5*(dgm[:,1]-dgm[:,0])))
 
     grads = tape.gradient(loss, [X])
diff --git a/src/python/doc/ls_simplex_tree_tflow_itf_ref.rst b/src/python/doc/ls_simplex_tree_tflow_itf_ref.rst
index 3200b8e5..56bb4492 100644
--- a/src/python/doc/ls_simplex_tree_tflow_itf_ref.rst
+++ b/src/python/doc/ls_simplex_tree_tflow_itf_ref.rst
@@ -32,7 +32,7 @@ Example of gradient computed from lower-star filtration of a simplex tree
     sl = LowerStarSimplexTreeLayer(simplextree=st, dimensions=[0])
 
     with tf.GradientTape() as tape:
-        dgm = sl.call(F)
+        dgm = sl.call(F)[0][0]
         loss = tf.math.reduce_sum(tf.square(.5*(dgm[:,1]-dgm[:,0])))
 
     grads = tape.gradient(loss, [F])
diff --git a/src/python/doc/rips_complex_tflow_itf_ref.rst b/src/python/doc/rips_complex_tflow_itf_ref.rst
index fc42e5c9..104b0971 100644
--- a/src/python/doc/rips_complex_tflow_itf_ref.rst
+++ b/src/python/doc/rips_complex_tflow_itf_ref.rst
@@ -19,7 +19,7 @@ Example of gradient computed from Vietoris-Rips persistence
     rl = RipsLayer(maximum_edge_length=2., dimensions=[0])
 
     with tf.GradientTape() as tape:
-        dgm = rl.call(X)
+        dgm = rl.call(X)[0][0]
         loss = tf.math.reduce_sum(tf.square(.5*(dgm[:,1]-dgm[:,0])))
 
     grads = tape.gradient(loss, [X])
diff --git a/src/python/gudhi/tensorflow/cubical_layer.py b/src/python/gudhi/tensorflow/cubical_layer.py
index b16c512f..99d02d66 100644
--- a/src/python/gudhi/tensorflow/cubical_layer.py
+++ b/src/python/gudhi/tensorflow/cubical_layer.py
@@ -47,7 +47,7 @@ class CubicalLayer(tf.keras.layers.Layer):
         """
         super().__init__(dynamic=True, **kwargs)
         self.dimensions = dimensions
-        self.min_persistence = min_persistence if min_persistence != None else [0. for _ in range(len(self.dimensions))] 
+        self.min_persistence = min_persistence if min_persistence != None else [0.] * len(self.dimensions)
         assert len(self.min_persistence) == len(self.dimensions)
 
     def call(self, X):
@@ -64,13 +64,13 @@ class CubicalLayer(tf.keras.layers.Layer):
         # Don't compute gradient for this operation
         Xflat = tf.reshape(X, [-1])
         Xdim = X.shape
-        indices = _Cubical(Xflat.numpy(), Xdim, self.dimensions)
+        indices_list = _Cubical(Xflat.numpy(), Xdim, self.dimensions)
         # Get persistence diagram by simply picking the corresponding entries in the image
-        self.dgms = [tf.reshape(tf.gather(Xflat, indice), [-1,2]) for indice in indices]
+        self.dgms = [tf.reshape(tf.gather(Xflat, indices), [-1,2]) for indices in indices_list]
         for idx_dim in range(len(self.min_persistence)):
             min_pers = self.min_persistence[idx_dim]
             if min_pers >= 0:
                 finite_dgm = self.dgms[idx_dim]
-                persistent_indices = np.argwhere(np.abs(finite_dgm[:,1]-finite_dgm[:,0]) > min_pers).ravel()
+                persistent_indices = tf.where(tf.math.abs(finite_dgm[:,1]-finite_dgm[:,0]) > min_pers)
                 self.dgms[idx_dim] = tf.gather(finite_dgm, indices=persistent_indices)
         return self.dgms
diff --git a/src/python/gudhi/tensorflow/lower_star_simplex_tree_layer.py b/src/python/gudhi/tensorflow/lower_star_simplex_tree_layer.py
index e1627944..8da1f7fe 100644
--- a/src/python/gudhi/tensorflow/lower_star_simplex_tree_layer.py
+++ b/src/python/gudhi/tensorflow/lower_star_simplex_tree_layer.py
@@ -12,8 +12,7 @@ def _LowerStarSimplexTree(simplextree, filtration, dimensions):
     #             filtration (function values on the vertices of st),
     #             dimensions (homology dimensions),
     
-    for s,_ in simplextree.get_filtration():
-        simplextree.assign_filtration(s, -1e10)
+    simplextree.reset_filtration(-np.inf, 0)
 
     # Assign new filtration values
     for i in range(simplextree.num_vertices()):
diff --git a/src/python/gudhi/tensorflow/rips_layer.py b/src/python/gudhi/tensorflow/rips_layer.py
index a5f212e3..88d501c1 100644
--- a/src/python/gudhi/tensorflow/rips_layer.py
+++ b/src/python/gudhi/tensorflow/rips_layer.py
@@ -40,7 +40,7 @@ class RipsLayer(tf.keras.layers.Layer):
     """
     TensorFlow layer for computing Rips persistence out of a point cloud
     """
-    def __init__(self, dimensions, maximum_edge_length=12, min_persistence=None, **kwargs):
+    def __init__(self, dimensions, maximum_edge_length=np.inf, min_persistence=None, **kwargs):
         """
         Constructor for the RipsLayer class
 
@@ -66,7 +66,7 @@ class RipsLayer(tf.keras.layers.Layer):
             dgms (list of tuple of TensorFlow variables): list of Rips persistence diagrams of length self.dimensions, where each element of the list is a tuple that contains the finite and essential persistence diagrams of shapes [num_finite_points, 2] and [num_essential_points, 1] respectively
         """    
         # Compute distance matrix
-        DX = tf.math.sqrt(tf.reduce_sum((tf.expand_dims(X, 1)-tf.expand_dims(X, 0))**2, 2))
+        DX = tf.norm(tf.expand_dims(X, 1)-tf.expand_dims(X, 0), axis=2)
         # Compute vertices associated to positive and negative simplices 
         # Don't compute gradient for this operation
         indices = _Rips(DX.numpy(), self.max_edge, self.dimensions)
diff --git a/src/python/test/test_diff.py b/src/python/test/test_diff.py
index e0c99d07..bab0d10c 100644
--- a/src/python/test/test_diff.py
+++ b/src/python/test/test_diff.py
@@ -22,7 +22,7 @@ def test_cubical_diff():
     cl = CubicalLayer(dimensions=[0])
 
     with tf.GradientTape() as tape:
-        dgm = cl.call(X)[0]
+        dgm = cl.call(X)[0][0]
         loss = tf.math.reduce_sum(tf.square(.5*(dgm[:,1]-dgm[:,0])))
     grads = tape.gradient(loss, [X])
     assert np.abs(grads[0].numpy()-np.array([[0.,0.,0.],[0.,.5,0.],[0.,0.,-.5]])).sum() <= 1e-6
@@ -34,7 +34,7 @@ def test_nonsquare_cubical_diff():
     cl = CubicalLayer(dimensions=[0])
 
     with tf.GradientTape() as tape:
-        dgm = cl.call(X)[0]
+        dgm = cl.call(X)[0][0]
         loss = tf.math.reduce_sum(tf.square(.5*(dgm[:,1]-dgm[:,0])))
     grads = tape.gradient(loss, [X])
     assert np.abs(grads[0].numpy()-np.array([[0.,0.5,-0.5],[0.,0.,0.]])).sum() <= 1e-6