rb-deeplearning-lib 0.1.5__tar.gz → 0.1.7__tar.gz

{rb_deeplearning_lib-0.1.5/src/rb_deeplearning_lib.egg-info → rb_deeplearning_lib-0.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rb-deeplearning-lib
-Version: 0.1.5
+Version: 0.1.7
 Summary: This is a machine learning--more specifically deep learning--library from my independent study on deep learning. This library is both a result of my learning and a tool for AI development.
 License-Expression: MIT
 Project-URL: Homepage, https://github.com/rylan-berry/DeepLearningIndependentStudy/tree/main/deeplearning_package

{rb_deeplearning_lib-0.1.5 → rb_deeplearning_lib-0.1.7}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "rb-deeplearning-lib"
-version = "0.1.5"
+version = "0.1.7"
 description = "This is a machine learning--more specifically deep learning--library from my independent study on deep learning. This library is both a result of my learning and a tool for AI development."
 readme = "README.md"
 requires-python = ">=3.8"

{rb_deeplearning_lib-0.1.5 → rb_deeplearning_lib-0.1.7}/src/rb_deeplearning_lib/neural_net.py

@@ -1,6 +1,7 @@
 from .autogradient import Values
 from .sequence import Sequence
 from .optimizer import Optimizer
+import numpy as np
 
 class Layer:
   def __init__(self, input,out,activ="_",rangeW=(-1,1),rangeB=(-1,1)):

rb_deeplearning_lib-0.1.7/src/rb_deeplearning_lib/sequence.py

@@ -0,0 +1,28 @@
+class Sequence:
+  def __init__(self, arr):
+    self.arr = arr
+
+  def __call__(self, x):
+    x_i = x
+    for item in self.arr:
+      x_i = item(x_i)
+    return x_i
+
+  def params(self):
+    all_params = []
+    for l in self.arr:
+      # Check if the item has a params method (e.g., Layer or Dense)
+      if hasattr(l, 'params'):
+        # layer_params should return weights and biases, which can be individual Values or lists of Values
+        layer_params = l.params()
+        # Ensure layer_params is iterable (e.g., a tuple of (weights, biases))
+        if isinstance(layer_params, tuple) or isinstance(layer_params, list):
+            for p_group in layer_params:
+                # If p_group is a list (e.g., from Dense containing multiple layers),
+                # extend with its elements, otherwise append the Value object itself.
+                if isinstance(p_group, list):
+                    all_params.extend(p_group)
+                else:
+                    all_params.append(p_group)
+
+    return all_params

{rb_deeplearning_lib-0.1.5 → rb_deeplearning_lib-0.1.7/src/rb_deeplearning_lib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rb-deeplearning-lib
-Version: 0.1.5
+Version: 0.1.7
 Summary: This is a machine learning--more specifically deep learning--library from my independent study on deep learning. This library is both a result of my learning and a tool for AI development.
 License-Expression: MIT
 Project-URL: Homepage, https://github.com/rylan-berry/DeepLearningIndependentStudy/tree/main/deeplearning_package

rb_deeplearning_lib-0.1.5/src/rb_deeplearning_lib/sequence.py

@@ -1,23 +0,0 @@
-class Sequence:
-  def __init__(self, arr):
-    self.arr = arr
-
-  def __call__(self, x):
-    x_i = x
-    for item in self.arr:
-      x_i = item(x_i)
-    return x_i
-
-  def params(self):
-    weis = []
-    biases = []
-    for l in self.arr:
-      w, b = l.params()
-      weis.append(w)
-      biases.append(b)
-
-    return weis, biases
-
-  def updateParams(self, l_rate):
-    for l in self.arr:
-      l.updateParams(l_rate)