Phase 2: Integration Tests (progressive integration)
❌ CompleteMLPipelineStillWorks: End To End Ml Pipeline Stable
Test failed (see output for details)
❌ CompleteMLPipelineStillWorks: Attention And Spatial Integration Stable
Test failed (see output for details)
❌ Module 09AutogradCore: Variable Wrapper Exists
Test failed (see output for details)
❌ Module 09AutogradCore: Gradient Computation
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ Module 09AutogradCore: Computation Graph Building
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ AutogradIntegration: Autograd With Layers
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ AutogradIntegration: Autograd With Spatial Operations
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ AutogradIntegration: Autograd With Attention
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ GradientBasedLearningFoundation: Parameter Gradient Computation
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ GradientBasedLearningFoundation: Loss Function Gradients
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ GradientBasedLearningFoundation: Optimization Readiness
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ Module 09Completion: Autograd Foundation Complete
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ FAILED
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ Integration tests failed (13 failures)
❌ Tests failed for 05_dataloader
💡 Fix the issues and try again
But when I run the test_module() in the notebook. It returns
Originally created by @kkkamur07 on GitHub (Apr 26, 2026).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/1547
Originally assigned to: @profvjreddi on GitHub.
### Area
Book — Volume I
### Location
Module 05 DataLoaders
### Description
When I run
```bash
tito module complete 05
```
it returns
```bash
Phase 2: Integration Tests (progressive integration)
❌ CompleteMLPipelineStillWorks: End To End Ml Pipeline Stable
Test failed (see output for details)
❌ CompleteMLPipelineStillWorks: Attention And Spatial Integration Stable
Test failed (see output for details)
❌ Module 09AutogradCore: Variable Wrapper Exists
Test failed (see output for details)
❌ Module 09AutogradCore: Gradient Computation
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ Module 09AutogradCore: Computation Graph Building
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ AutogradIntegration: Autograd With Layers
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ AutogradIntegration: Autograd With Spatial Operations
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ AutogradIntegration: Autograd With Attention
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ GradientBasedLearningFoundation: Parameter Gradient Computation
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ GradientBasedLearningFoundation: Loss Function Gradients
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ GradientBasedLearningFoundation: Optimization Readiness
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ Module 09Completion: Autograd Foundation Complete
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ FAILED
E ModuleNotFoundError: No module named 'tinytorch.core.spatial'
❌ Integration tests failed (13 failures)
❌ Tests failed for 05_dataloader
💡 Fix the issues and try again
```
But when I run the `test_module()` in the notebook. It returns
```python
🧪 RUNNING MODULE INTEGRATION TEST
==================================================
Running unit tests...
🔬 Unit Test: Dataset Abstract Base Class...
✅ Dataset is properly abstract
✅ Dataset interface works correctly!
🔬 Unit Test: TensorDataset...
Index out of bounds
✅ TensorDataset works correctly!
🔬 Unit Test: DataLoader...
✅ DataLoader works correctly!
🔬 Unit Test: DataLoader Deterministic Shuffling...
✅ Deterministic shuffling works correctly!
🔬 Unit Test: Data Augmentation...
Testing RandomHorizontalFlip...
Testing RandomCrop...
Testing Compose...
Testing Tensor compatibility...
Testing randomness...
✅ Data Augmentation works correctly!
Running integration scenarios...
🔬 Integration Test: Training Workflow...
📊 Dataset splits:
Training: 800 samples, 25 batches
Validation: 200 samples, 7 batches
🏃 Simulated Training Loop:
Batch 1: 32 samples
Batch 2: 32 samples
Batch 3: 32 samples
Total: 25 batches, 800 samples processed
✅ Training integration works correctly!
🔬 Integration Test: Augmentation with DataLoader...
✅ Augmentation + DataLoader integration works!
==================================================
🎉 ALL TESTS PASSED! Module ready for export.
Run: tito module complete 08
🎯 AHA MOMENT: DataLoader Batches Your Data
=============================================
Dataset: 100 samples
Batch size: 32
Number of batches: 4
Batches:
Batch 1: 32 samples, shape (32, 64)
Batch 2: 32 samples, shape (32, 64)
Batch 3: 32 samples, shape (32, 64)
Batch 4: 4 samples, shape (4, 64)
✨ Your DataLoader organizes data for efficient training!
```
Please help me through this.
### Expected Behavior
_No response_
### Environment (TinyTorch bugs only)
_No response_
Ah, I just fixed quite a few of these. The labs are the last part that I really need to iron out now. So thank you for sharing these, I just need to push my changes up and publish the live version! Thanks @kkkamur07
<!-- gh-comment-id:4322014952 -->
@profvjreddi commented on GitHub (Apr 26, 2026):
Ah, I just fixed quite a few of these. The labs are the last part that I really need to iron out now. So thank you for sharing these, I just need to push my changes up and publish the live version! Thanks @kkkamur07
Hey @kkkamur07, thanks for the detailed bug report — the full test trace was the key to figuring this out.
Good news: this isn't a code bug in current dev, it's a stale-checkout issue. The class names in your trace (Module 09AutogradCore, CompleteMLPipelineStillWorks, AutogradIntegration, GradientBasedLearningFoundation) don't exist anywhere in the current branch — they got removed when we did a big module renumber back in January 2026. In the new layout, Module 06 is autograd and Module 09 is convolutions, and the test scaffolding around tinytorch.core.spatial looks completely different.
Could you pull the latest and try again?
git pull origin dev
tito dev export
tito module complete05
That should pick up the post-renumber module layout and the failures should disappear. If anything still breaks after that, please reopen with the fresh trace and I'll dig back in.
Really appreciate you taking the time to file this with full output — it made the diagnosis straightforward.
<!-- gh-comment-id:4322104936 -->
@profvjreddi commented on GitHub (Apr 26, 2026):
Hey @kkkamur07, thanks for the detailed bug report — the full test trace was the key to figuring this out.
Good news: this isn't a code bug in current dev, it's a stale-checkout issue. The class names in your trace (`Module 09AutogradCore`, `CompleteMLPipelineStillWorks`, `AutogradIntegration`, `GradientBasedLearningFoundation`) don't exist anywhere in the current branch — they got removed when we did a big module renumber back in January 2026. In the new layout, Module 06 is `autograd` and Module 09 is `convolutions`, and the test scaffolding around `tinytorch.core.spatial` looks completely different.
Could you pull the latest and try again?
```bash
git pull origin dev
tito dev export
tito module complete 05
```
That should pick up the post-renumber module layout and the failures should disappear. If anything still breaks after that, please reopen with the fresh trace and I'll dig back in.
Really appreciate you taking the time to file this with full output — it made the diagnosis straightforward.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @kkkamur07 on GitHub (Apr 26, 2026).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/1547
Originally assigned to: @profvjreddi on GitHub.
Area
Book — Volume I
Location
Module 05 DataLoaders
Description
When I run
it returns
But when I run the
test_module()in the notebook. It returnsPlease help me through this.
Expected Behavior
No response
Environment (TinyTorch bugs only)
No response
@profvjreddi commented on GitHub (Apr 26, 2026):
Ah, I just fixed quite a few of these. The labs are the last part that I really need to iron out now. So thank you for sharing these, I just need to push my changes up and publish the live version! Thanks @kkkamur07
@profvjreddi commented on GitHub (Apr 26, 2026):
Hey @kkkamur07, thanks for the detailed bug report — the full test trace was the key to figuring this out.
Good news: this isn't a code bug in current dev, it's a stale-checkout issue. The class names in your trace (
Module 09AutogradCore,CompleteMLPipelineStillWorks,AutogradIntegration,GradientBasedLearningFoundation) don't exist anywhere in the current branch — they got removed when we did a big module renumber back in January 2026. In the new layout, Module 06 isautogradand Module 09 isconvolutions, and the test scaffolding aroundtinytorch.core.spatiallooks completely different.Could you pull the latest and try again?
That should pick up the post-renumber module layout and the failures should disappear. If anything still breaks after that, please reopen with the fresh trace and I'll dig back in.
Really appreciate you taking the time to file this with full output — it made the diagnosis straightforward.