Delete README.md

README.md

@@ -1,89 +0,0 @@
-# CIS 6270 Course Code
-
-![CIS6270_Unified_Generative_Dynamics (1)](https://cdn-uploads.huggingface.co/production/uploads/64cd5b3f0494187a9e8b7c69/myMO3iMdYkquG6D5mhU4S.png)
-
-Code for CIS 6270, Discrete Generative Models, at the University of Pennsylvania. One file per
-framework, written to be read beside the lecture notes: the corruption, the loss and the sampler of
-each method are in that file, in the notes' own symbols, with a comment on almost every line.
-
-Everything runs offline on a laptop CPU in seconds. The one exception is
-`lecture_2/flow_matching_mnist.py`, which downloads MNIST, as its name says; `--dataset toy` runs it
-offline. Lecture 3 goes the other way: its scripts run offline and reach MNIST or ESM-2 only when
-you ask with `--dataset mnist` or `--dataset esm2`.
-
-```bash
-git clone https://huggingface.co/ChatterjeeLab/CIS6270
-cd CIS6270
-python3 -m venv .venv && source .venv/bin/activate
-python -m pip install -r requirements.txt
-
-python lecture_4/mdlm.py          # train masked diffusion on DNA and generate, about 20 seconds
-python -m pytest tests -q         # the notes' worked numbers, checked against the code
-```
-
-On Windows PowerShell, activate with `.venv\Scripts\Activate.ps1`.
-
-## The lectures
-
-| | Lecture | What the code builds | Guide |
-| --- | --- | --- | --- |
-| 1 | Mathematical Foundations | Every worked example of the chapter, printed beside the chapter's own number | [lecture_1](lecture_1/README.md) |
-| 2 | Flow Matching | Velocity fields and the flow map, the conditional path, the objective, MNIST | [lecture_2](lecture_2/README.md) |
-| 3 | Score Matching and Diffusion | The forward SDE, the three score objectives, DDPM, three samplers, guidance, latent diffusion on protein embeddings | [lecture_3](lecture_3/README.md) |
-| 4 | Discrete Diffusion | Jump processes, MDLM, uniform replacement, D3PM, SEDD, block diffusion, guidance, planners, search | [lecture_4](lecture_4/README.md) |
-| 5 | Discrete Flow Matching | The mixture path and its velocity, correctors, kinetic-optimal paths, Edit Flows, Dirichlet, Fisher and Gumbel geometries, guidance, rectification | [lecture_5](lecture_5/README.md) |
-| 6 | Flow Maps | The three axioms, the three residuals, MeanFlow, Shortcut, consistency, the discrete flow maps, stochastic, expanding and posterior maps | [lecture_6](lecture_6/README.md) |
-| 7 | Optimal Transport and Schrödinger Bridges | Kantorovich and duality, Sinkhorn, displacement interpolation, static and dynamic bridges, DSB, DSBM, SF2M, the discrete bridges, branched and entangled | [lecture_7](lecture_7/README.md) |
-
-## How a file is laid out
-
-Every script follows the shape of `lecture_2/flow_matching_mnist.py`: a docstring naming the
-framework and the chapter sections it implements, numbered `# %%` sections in the order the lecture
-develops them, and a `main()` that prints a report.
-
-```
-cis6270/           the scaffolding, read once: data, the four small networks, the training loop,
-                   the sample metrics. None of it is the subject of a lecture.
-lecture_1/ ... 7/  one file per framework, standalone and runnable
-tests/             the chapters' worked numbers, asserted against the code
-```
-
-Nothing in `lecture_N/` imports anything from another `lecture_N/`. To read a method, open its file.
-
-## The data
-
-| Data | Used by | Download |
-| --- | --- | --- |
-| Synthetic DNA, four letters, a planted ACGT motif, a GC label | Lectures 4 to 6 | no |
-| Four 2-D Gaussians, two moons, a checkerboard | Lectures 2, 3, 6, 7 | no |
-| Four-token phrases such as `red circle moves left` | Lecture 6 | no |
-| MNIST | Lectures 2, 3, with `--dataset mnist` | yes |
-| ESM-2 residue embeddings | Lecture 3, with `--dataset esm2` | yes |
-
-The DNA is synthetic and the two guidance objectives are explicit toy properties. They are there so
-that a training run finishes while you watch it, and so that the structure a sampler is supposed to
-recover is one you can check by eye. Nothing here is a benchmark or a biological predictor.
-
-## The shared flags
-
-`--steps`, `--batch-size`, `--lr`, `--width`, `--length`, `--samples`, `--sample-steps`, `--seed`,
-`--device`, `--out`, `--quiet`. Learn them once and they work everywhere. `--out` writes
-`checkpoint.pt`, `config.json`, `report.json`, `losses.json` and `samples.txt`.
-
-The defaults are a short budget, enough to see a method work in well under a minute. A method
-usually looks better with `--steps 1500 --width 128`.
-
-## Tests
-
-```bash
-python -m pytest tests -q
-```
-
-The suite asserts the numbers the notes work by hand: the posterior average of the reverse rates on
-one DNA base, the Sinkhorn iterates at `epsilon = 2`, the cocycle and Jacobian of the Gaussian flow,
-the conditional velocity of the mixture path. A number in the book and a number in the code are the
-same number, and the suite is what keeps them that way.
-
-## License
-
-[MIT](LICENSE). ESM-2 weights come from their original repository under its own terms.