solution check

.gitignore

@@ -131,3 +131,6 @@ dmypy.json
 data/
 pyodideModule2.md
 module2-modernization.md
+task2_1_comments.md
+task2_2_comments.md
+task2_3_comments.md

README.md

@@ -39,11 +39,25 @@ python project/run_tensor.py
 
 ## Tasks
 
-- **Task 2.1**: Implement tensor data structures with indexing and strides
-- **Task 2.2**: Implement tensor broadcasting for operations between different shapes
-- **Task 2.3**: Implement tensor operations (map, zip, reduce) and mathematical functions
-- **Task 2.4**: Extend autodifferentiation to work with tensors and broadcasting
-- **Task 2.5**: Create tensor-based neural network training
+### Task 2.1: Tensor Data - Indexing
+**File to Edit**: `minitorch/tensor_data.py`
+
+### Task 2.2: Tensor Broadcasting
+**File to Edit**: `minitorch/tensor_data.py`
+
+### Task 2.3: Tensor Operations
+**Files to Edit**: `minitorch/tensor_ops.py`, `minitorch/tensor_functions.py`, `minitorch/tensor.py`
+
+### Task 2.4: Extend autodifferentiation to work with tensors and broadcasting
+**Files to Edit**: `minitorch/tensor_functions.py`
+
+### Task 2.5: Tensor-Based Neural Network Training
+**File to Edit**: `project/run_tensor.py`
+
+**Requirements**:
+- Train on all datasets and record results in README
+- Record time per epoch for performance comparison
+- Should match functionality of `project/run_scalar.py` but use tensor operations
 
 ## Testing
 

tests/tensor_strategies.py

@@ -63,6 +63,11 @@ def indices(draw: DrawFn, layout: Tensor) -> UserIndex:
     return tuple((draw(integers(min_value=0, max_value=s - 1)) for s in layout.shape))
 
 
+@composite
+def tensor_data_indices(draw: DrawFn, layout: TensorData) -> UserIndex:
+    return tuple((draw(integers(min_value=0, max_value=s - 1)) for s in layout.shape))
+
+
 @composite
 def tensors(
     draw: DrawFn,

tests/test_tensor.py

@@ -1,7 +1,7 @@
 from typing import Callable, Iterable, List, Tuple
 
 import pytest
-from hypothesis import given
+from hypothesis import assume, given
 from hypothesis.strategies import DataObject, data, lists, permutations
 
 from minitorch import MathTestVariable, Tensor, grad_check, tensor
@@ -106,6 +106,15 @@ def test_two_grad(
 ) -> None:
     name, _, tensor_fn = fn
     t1, t2 = ts
+
+    # Avoid discontinuities for comparison functions
+    if name == "gt2" or name == "lt2":
+        assume(abs((t1.to_numpy().min() + 1.2) - t2.to_numpy().max()) > 1e-3)
+        assume(abs((t1.to_numpy().max() + 1.2) - t2.to_numpy().min()) > 1e-3)
+    elif name == "eq2":
+        assume(abs(t1.to_numpy().min() - (t2.to_numpy().max() + 5.5)) > 1e-3)
+        assume(abs(t1.to_numpy().max() - (t2.to_numpy().min() + 5.5)) > 1e-3)
+
     grad_check(tensor_fn, t1, t2)
 
 
@@ -119,6 +128,15 @@ def test_two_grad_broadcast(
     "Test the grad of a two argument function"
     name, base_fn, tensor_fn = fn
     t1, t2 = ts
+
+    # Avoid discontinuities for comparison functions
+    if name == "gt2" or name == "lt2":
+        assume(abs((t1.to_numpy().min() + 1.2) - t2.to_numpy().max()) > 1e-3)
+        assume(abs((t1.to_numpy().max() + 1.2) - t2.to_numpy().min()) > 1e-3)
+    elif name == "eq2":
+        assume(abs(t1.to_numpy().min() - (t2.to_numpy().max() + 5.5)) > 1e-3)
+        assume(abs(t1.to_numpy().max() - (t2.to_numpy().min() + 5.5)) > 1e-3)
+
     grad_check(tensor_fn, t1, t2)
 
     # broadcast check

tests/test_tensor_data.py

@@ -5,7 +5,7 @@
 import minitorch
 from minitorch import TensorData
 
-from .tensor_strategies import indices, tensor_data
+from .tensor_strategies import tensor_data, tensor_data_indices
 
 # ## Tasks 2.1
 
@@ -81,7 +81,7 @@ def test_index(tensor_data: TensorData) -> None:
 @given(data())
 def test_permute(data: DataObject) -> None:
     td = data.draw(tensor_data())
-    ind = data.draw(indices(td))
+    ind = data.draw(tensor_data_indices(td))
     td_rev = td.permute(*list(reversed(range(td.dims))))
     assert td.index(ind) == td_rev.index(tuple(reversed(ind)))
 
@@ -120,6 +120,30 @@ def test_shape_broadcast() -> None:
     assert c == (2, 5)
 
 
+@pytest.mark.task2_2
+def test_broadcast_index() -> None:
+    big_index = minitorch.array([1, 2, 3])
+    big_shape = minitorch.array([2, 3, 4])
+    shape = minitorch.array([1, 3, 4])
+    out_index = minitorch.array([0, 0, 0])
+
+    minitorch.broadcast_index(big_index, big_shape, shape, out_index)
+    assert out_index[0] == 0  # dimension 1 maps to 0
+    assert out_index[1] == 2  # dimension 3 matches
+    assert out_index[2] == 3  # dimension 4 matches
+
+    # Test with more dimensions of size 1
+    big_index = minitorch.array([1, 0, 2])
+    big_shape = minitorch.array([2, 1, 3])
+    shape = minitorch.array([1, 1, 3])
+    out_index = minitorch.array([0, 0, 0])
+
+    minitorch.broadcast_index(big_index, big_shape, shape, out_index)
+    assert out_index[0] == 0  # dimension 1 maps to 0
+    assert out_index[1] == 0  # dimension 1 maps to 0
+    assert out_index[2] == 2  # dimension 3 matches
+
+
 @given(tensor_data())
 def test_string(tensor_data: TensorData) -> None:
     tensor_data.to_string()