Dropout Last Layer - option to remove node on the last layer and seed dropouts for reproducible animations (#30)

Option to remove node on the last layer and seed dropouts for reproducible animations

.gitignore

@@ -9,4 +9,4 @@ setup.cfg
 !examples/media
 examples/media/videos
 examples/media/text
-examples/media/images
+examples/media/images

Readme.md

@@ -347,4 +347,40 @@ We can now render with:
 $ manim -pql example.py
 ```
 
-<img src="assets/readme/dropout.gif">
+<img src="assets/readme/dropout.gif">
+
+#### Seed the Dropouts:
+
+```py
+self.play(
+    make_neural_network_dropout_animation(
+        nn, dropout_rate=0.25, do_forward_pass=True, seed=4
+    )
+)
+```
+
+<img src="assets/readme/dropout_seed_4.gif">
+
+#### Seed the Dropouts with First layer static:
+
+```py
+self.play(
+    make_neural_network_dropout_animation(
+        nn, dropout_rate=0.25, do_forward_pass=True, seed=4,  first_layer_stable=True
+    )
+)
+```
+
+<img src="assets/readme/dropout_seed_4_first.gif">
+
+#### Seed the Dropouts with First and Last layers static:
+
+```py
+self.play(
+    make_neural_network_dropout_animation(
+        nn, dropout_rate=0.25, do_forward_pass=True, seed=4, first_layer_stable=True, last_layer_stable=True
+    )
+)
+```
+
+<img src="assets/readme/dropout_seed_4_first_last.gif">

manim_ml/neural_network/animations/dropout.py

@@ -192,7 +192,7 @@ def make_forward_pass_with_dropout_animation(
 
 
 def make_neural_network_dropout_animation(
-    neural_network, dropout_rate=0.5, do_forward_pass=True
+    neural_network, dropout_rate=0.5, do_forward_pass=True, last_layer_stable=False, first_layer_stable=False, seed=None
 ):
     """
     Makes a dropout animation for a given neural network.
@@ -204,6 +204,8 @@ def make_neural_network_dropout_animation(
     3. Revert network to pre-dropout appearance
     """
     # Go through the network and get the FeedForwardLayer instances
+    if seed is not None:
+        random.seed(seed)
     feed_forward_layers = neural_network.filter_layers(
         lambda layer: isinstance(layer, FeedForwardLayer)
     )
@@ -213,12 +215,16 @@ def make_neural_network_dropout_animation(
     )
     # Get random nodes to drop out for each FeedForward Layer
     layers_to_nodes_to_drop_out = {}
-    for feed_forward_layer in feed_forward_layers:
+    for idx, feed_forward_layer in enumerate(feed_forward_layers):
         num_nodes = feed_forward_layer.num_nodes
         nodes_to_drop_out = []
         # Compute random probability that each node is dropped out
         for node_index in range(num_nodes):
             dropout_prob = random.random()
+            if last_layer_stable and idx==len(feed_forward_layers)-1:
+                    continue
+            if first_layer_stable and idx==0:
+                continue
             if dropout_prob < dropout_rate:
                 nodes_to_drop_out.append(node_index)
         # Add the mapping to the dict