# core.solver.TransformationModel { #mlsysim.core.solver.TransformationModel }

```python
core.solver.TransformationModel()
```

Analyzes the 'Transformation Wall' — CPU preprocessing and augmentation bottlenecks.

This solver models the compute required to decode and augment data
(e.g., JPEG decoding, image resizing, tokenization) before it hits the GPU.

Literature Source:
1. Mohan et al. (2022), "Analyzing and Mitigating Data Bottlenecks in Deep Learning Training."
2. Markidis et al. (2021), "The Data Pipeline is the Bottleneck."