Feedback

README.md

@@ -1,3 +1,4 @@
+[![Review Assignment Due Date](https://classroom.github.com/assets/deadline-readme-button-22041afd0340ce965d47ae6ef1cefeee28c7c493a6346c4f15d667ab976d596c.svg)](https://classroom.github.com/a/YFgwt0yY)
 # MiniTorch Module 2
 
 <img src="https://minitorch.github.io/minitorch.svg" width="50%">
@@ -15,4 +16,32 @@ python sync_previous_module.py previous-module-dir current-module-dir
 
 The files that will be synced are:
 
-        minitorch/operators.py minitorch/module.py minitorch/autodiff.py minitorch/scalar.py minitorch/scalar_functions.py minitorch/module.py project/run_manual.py project/run_scalar.py project/datasets.py
+        minitorch/operators.py minitorch/module.py minitorch/autodiff.py minitorch/scalar.py minitorch/scalar_functions.py minitorch/module.py project/run_manual.py project/run_scalar.py project/datasets.py
+
+## Simple Dataset
+<img src="images/Simple/SimpleDataSet_Model.png">
+<img src="images/Simple/SimpleDataSet_Hyper.png">
+<img src="images/Simple/SimpleDataSet_LossGraph.png">
+<img src="images/Simple/SimpleDataSet_LossTable.png">
+<img src="images/Simple/SimpleDataSet.png">
+
+## Diag Dataset
+<img src="images/Diag/DiagDataSet_Model.png">
+<img src="images/Diag/DiagDataSet_Hyper.png">
+<img src="images/Diag/DiagDataSet_LossGraph.png">
+<img src="images/Diag/DiagDataSet_LossTable.png">
+<img src="images/Diag/DiagDataSet.png">
+
+## Split Dataset
+<img src="images/Split/SplitDataSet_Model.png">
+<img src="images/Split/SplitDataSet_Hyper.png">
+<img src="images/Split/SplitDataSet_LossGraph.png">
+<img src="images/Split/SplitDataSet_LossTable.png">
+<img src="images/Split/SplitDataSet.png">
+
+## XOR Dataset
+<img src="images/XOR/XORDataSet_Model.png">
+<img src="images/XOR/XORDataSet_Hyper.png">
+<img src="images/XOR/XORDataSet_LossGraph.png">
+<img src="images/XOR/XORDataSet_LossTable.png">
+<img src="images/XOR/XORDataSet.png">

minitorch/__init__.py

@@ -1,4 +1,4 @@
-from .testing import MathTest, MathTestVariable  # type: ignore # noqa: F401,F403
+from .testing import MathTest, MathTestVariable  # type: ignore # noqa: F401,F403, D104
 from .tensor_data import *  # noqa: F401,F403
 from .tensor import *  # noqa: F401,F403
 from .tensor_ops import *  # noqa: F401,F403

minitorch/autodiff.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
-from typing import Any, Iterable, List, Tuple, Protocol
+from typing import Any, Iterable, Tuple, Protocol
 
 
 # ## Task 1.1
@@ -25,26 +25,46 @@ def central_difference(f: Any, *vals: Any, arg: int = 0, epsilon: float = 1e-6)
         An approximation of $f'_i(x_0, \ldots, x_{n-1})$
 
     """
-    raise NotImplementedError("Need to include this file from past assignment.")
+    # TODO: Implement for Task 1.1.
+
+    up_perturbed = vals[:arg] + (vals[arg] + epsilon,) + vals[arg + 1 :]
+    down_perturbed = vals[:arg] + (vals[arg] - epsilon,) + vals[arg + 1 :]
+
+    slope = (f(*up_perturbed) - f(*down_perturbed)) / (2 * epsilon)
+
+    return slope
 
 
 variable_count = 1
 
 
 class Variable(Protocol):
-    def accumulate_derivative(self, x: Any) -> None: ...
+    def accumulate_derivative(self, x: Any) -> None:
+        """Accumulates the derivative of the output with respect to this variable."""
+
+    ...
 
     @property
-    def unique_id(self) -> int: ...
+    def unique_id(self) -> int:
+        """Returns the unique identifier of this variable."""
+        ...
 
-    def is_leaf(self) -> bool: ...
+    def is_leaf(self) -> bool:
+        """Returns True if this variable is a leaf node."""
+        ...
 
-    def is_constant(self) -> bool: ...
+    def is_constant(self) -> bool:
+        """Returns True if this variable is a constant."""
+        ...
 
     @property
-    def parents(self) -> Iterable["Variable"]: ...
+    def parents(self) -> Iterable["Variable"]:
+        """Returns the parents of this variable."""
+        ...
 
-    def chain_rule(self, d_output: Any) -> Iterable[Tuple[Variable, Any]]: ...
+    def chain_rule(self, d_output: Any) -> Iterable[Tuple[Variable, Any]]:
+        """Computes gradients of inputs using the chain rule."""
+        ...
 
 
 def topological_sort(variable: Variable) -> Iterable[Variable]:
@@ -59,7 +79,30 @@ def topological_sort(variable: Variable) -> Iterable[Variable]:
         Non-constant Variables in topological order starting from the right.
 
     """
-    raise NotImplementedError("Need to include this file from past assignment.")
+    # create a set to store visited variables
+    visited = set()
+
+    # create a list to store sorted variables
+    sorted_vars = []
+
+    def visit(node: Variable) -> None:
+        """Visits the variable and its parents recursively."""
+        if node.unique_id in visited or node.is_constant():
+            return
+        if not node.is_leaf():
+            # visit all the parents of the variable
+            for parent in node.parents:
+                if not parent.is_constant():
+                    visit(parent)
+        # mark the variable as visited
+        visited.add(node.unique_id)
+
+        # once all the parents have been visited, add the variable to the sorted list
+        sorted_vars.insert(0, node)
+
+    visit(variable)
+
+    return sorted_vars
 
 
 def backpropagate(variable: Variable, deriv: Any) -> None:
@@ -69,12 +112,39 @@ def backpropagate(variable: Variable, deriv: Any) -> None:
     Args:
     ----
         variable: The right-most variable
-        deriv  : Its derivative that we want to propagate backward to the leaves.
+        deriv: Its derivative that we want to propagate backward to the leaves.
 
-    No return. Should write to its results to the derivative values of each leaf through `accumulate_derivative`.
+    Returns:
+    -------
+        No return. Should write to its results to the derivative values of each leaf through `accumulate_derivative`.
 
     """
-    raise NotImplementedError("Need to include this file from past assignment.")
+    # call topological sort
+    sorted_vars = topological_sort(variable)
+
+    # create dictionary to store variables and their derivatives
+    derivatives = {}
+    derivatives[variable.unique_id] = deriv
+
+    # iterate through the sorted variables in backward order
+    for var in sorted_vars:
+        # if the variable is a leaf node, accumulate the derivative
+        if var.is_leaf():
+            # accumulate the derivative
+            var.accumulate_derivative(derivatives[var.unique_id])
+
+        # if the variable is not a leaf node
+        else:
+            # call .chain_rule on the last function in the history of the variable
+            grads = var.chain_rule(derivatives[var.unique_id])
+
+            # loop through all the Scalars+derivatives provided by the chain rule
+            for parent, derivative in grads:
+                if parent.is_constant():
+                    continue
+                # accumulate derivatives for the Scalar in the dictionary
+                derivatives.setdefault(parent.unique_id, 0)
+                derivatives[parent.unique_id] += derivative
 
 
 @dataclass
@@ -92,4 +162,5 @@ def save_for_backward(self, *values: Any) -> None:
 
     @property
     def saved_tensors(self) -> Tuple[Any, ...]:
+        """Returns the saved values."""
         return self.saved_values

minitorch/datasets.py

@@ -5,6 +5,17 @@
 
 
 def make_pts(N: int) -> List[Tuple[float, float]]:
+    """Generate a list of N random 2D points.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A list of N points, each represented as a tuple of two floats.
+
+    """
     X = []
     for i in range(N):
         x_1 = random.random()
@@ -21,6 +32,17 @@ class Graph:
 
 
 def simple(N: int) -> Graph:
+    """Generate a simple dataset where the label is 1 if the first coordinate is less than 0.5 and 0 otherwise.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A Graph object containing the generated data.
+
+    """
     X = make_pts(N)
     y = []
     for x_1, x_2 in X:
@@ -30,6 +52,17 @@ def simple(N: int) -> Graph:
 
 
 def diag(N: int) -> Graph:
+    """Generate a dataset where the label is 1 if the sum of the coordinates is less than 0.5 and 0 otherwise.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A Graph object containing the generated data.
+
+    """
     X = make_pts(N)
     y = []
     for x_1, x_2 in X:
@@ -39,6 +72,17 @@ def diag(N: int) -> Graph:
 
 
 def split(N: int) -> Graph:
+    """Generate a dataset where the label is 1 if the first coordinate is less than 0.2 or greater than 0.8 and 0 otherwise.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A Graph object containing the generated data.
+
+    """
     X = make_pts(N)
     y = []
     for x_1, x_2 in X:
@@ -48,6 +92,17 @@ def split(N: int) -> Graph:
 
 
 def xor(N: int) -> Graph:
+    """Generate a dataset where the label is 1 if the first coordinate is less than 0.5 and the second coordinate is greater than 0.5 or vice versa and 0 otherwise.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A Graph object containing the generated data.
+
+    """
     X = make_pts(N)
     y = []
     for x_1, x_2 in X:
@@ -57,6 +112,17 @@ def xor(N: int) -> Graph:
 
 
 def circle(N: int) -> Graph:
+    """Generate a dataset where the label is 1 if the point's distance from the origin (after shifting by 0.5) is greater than sqrt(0.1), and 0 otherwise.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A Graph object containing the generated data.
+
+    """
     X = make_pts(N)
     y = []
     for x_1, x_2 in X:
@@ -67,10 +133,24 @@ def circle(N: int) -> Graph:
 
 
 def spiral(N: int) -> Graph:
+    """Generate a spiral dataset with N points where the first half of the points are in class 0 and the second half are in class 1.
+
+    Args:
+    ----
+        N: The number of points to generate.
+
+    Returns:
+    -------
+        A Graph object containing the generated data.
+
+    """
+
     def x(t: float) -> float:
+        """Calculate the x-coordinate of a point on a spiral."""
         return t * math.cos(t) / 20.0
 
     def y(t: float) -> float:
+        """Calculate the y-coordinate of a point on a spiral."""
         return t * math.sin(t) / 20.0
 
     X = [

minitorch/module.py

@@ -30,12 +30,29 @@ def modules(self) -> Sequence[Module]:
         return list(m.values())
 
     def train(self) -> None:
-        """Set the mode of this module and all descendent modules to `train`."""
-        raise NotImplementedError("Need to include this file from past assignment.")
+        """Set the `training` flag of this and descendent to true.
+
+        Returns
+        -------
+            None.
+
+        """
+        self.training = True
+        for module in self.modules():
+            module.train()
 
     def eval(self) -> None:
-        """Set the mode of this module and all descendent modules to `eval`."""
-        raise NotImplementedError("Need to include this file from past assignment.")
+        """Set the `training` flag of this and descendent to false.
+
+        Returns
+        -------
+            None.
+
+        """
+        self.training = False
+
+        for module in self.modules():
+            module.training = False
 
     def named_parameters(self) -> Sequence[Tuple[str, Parameter]]:
         """Collect all the parameters of this module and its descendents.
@@ -45,11 +62,26 @@ def named_parameters(self) -> Sequence[Tuple[str, Parameter]]:
             The name and `Parameter` of each ancestor parameter.
 
         """
-        raise NotImplementedError("Need to include this file from past assignment.")
+        parameters = {}
+
+        for key, val in self._parameters.items():
+            parameters[key] = val
+
+        for name, module in self._modules.items():
+            for key, val in module.named_parameters():
+                parameters[f"{name}.{key}"] = val
+
+        return list(parameters.items())
 
     def parameters(self) -> Sequence[Parameter]:
-        """Enumerate over all the parameters of this module and its descendents."""
-        raise NotImplementedError("Need to include this file from past assignment.")
+        """Enumerate over all the parameters of this module and its descendents.
+
+        Returns
+        -------
+            The `Parameter` of this module and its descendents.
+
+        """
+        return [val for _, val in self.named_parameters()]
 
     def add_parameter(self, k: str, v: Any) -> Parameter:
         """Manually add a parameter. Useful helper for scalar parameters.
@@ -85,6 +117,7 @@ def __getattr__(self, key: str) -> Any:
         return None
 
     def __call__(self, *args: Any, **kwargs: Any) -> Any:
+        """Turns the instance into a callable object."""
         return self.forward(*args, **kwargs)
 
     def __repr__(self) -> str: