Correct library versions, add authorship, remove em dashes

README.md

@@ -11,10 +11,10 @@ tags:
 - edge-features
 ---
 
-# GrapHist v2 — Edge-Informed Graph Self-Supervised Learning for Histopathology
+# GrapHist v2: Edge-Informed Graph Self-Supervised Learning for Histopathology
 
-GrapHist v2 encodes a histopathology tile as a **cell graph** — one node per nucleus, edges
-from a Delaunay triangulation pruned at 100 µm — and embeds it with a self-supervised
+GrapHist v2 encodes a histopathology tile as a **cell graph** (one node per nucleus, edges
+from a Delaunay triangulation pruned at 100 µm) and embeds it with a self-supervised
 **ACM-GINEConv** encoder. It extends [GrapHist](https://huggingface.co/ogutsevda/graphist)
 ([arXiv:2603.00143](https://arxiv.org/abs/2603.00143)) by making the **edges carry
 information**: each edge holds a 75-dimensional descriptor of the tissue between two cells,
@@ -26,7 +26,7 @@ objective is GraphMAE-style masked node-feature reconstruction (mask 50 %, scale
 error, α = 3) plus a **VICReg** variance/covariance term on the pooled graph embedding, which
 keeps all embedding dimensions in use (final `pca_1` 0.17, effective dimension 11.9).
 
-7.98 M-parameter encoder · 512-dim embeddings · 0.093 ms/patch inference.
+7.98 M-parameter encoder, 512-dim embeddings, 0.093 ms/patch inference.
 
 ## Results
 
@@ -44,10 +44,10 @@ every artifact here to the result it reproduces.
 ## What's in this repo
 
 ```
-graphist_v2.pt        the model — 126,597,010 B, md5 81a2e6b91cefff0bbbc13c6fd318ee78
+graphist_v2.pt        the model. 126,597,010 B, md5 81a2e6b91cefff0bbbc13c6fd318ee78
 modeling/             code to load and run it, plus helper scripts
 graphs/               cell graphs: TCGA-BRCA (11.1 M, pre-training) · BACH · BRACS ·
-                      BreakHis · SPIDER-breast
+                      BreakHis, SPIDER-breast
 embeddings/           embeddings from this model: slide-level (4 cohorts), cell-level (22 sets)
 baselines/            DINOv2, MAE and GrapHist v1 features for the comparison rows
 labels/               TCGA-BRCA clinical table and slide labels
@@ -56,7 +56,7 @@ upstream_v1/          mirror of the original GrapHist v1 releases
 ```
 
 For inference you need only `graphist_v2.pt` + `modeling/`. Everything else exists to reproduce
-a specific published number — see [`DATA.md`](DATA.md).
+a specific published number. See [`DATA.md`](DATA.md).
 
 > ⚠️ **If you use `graphs/`, run `modeling/rebase_graph_paths.py` first.** The label CSVs store
 > bare filenames and the loader resolves them against the working directory, so without this
@@ -69,8 +69,8 @@ from huggingface_hub import snapshot_download
 path = snapshot_download(repo_id="Ace3Z/graphist-v2")
 ```
 
-Three arguments are **load-critical** — `build_model` reads two with defaults that do not match
-this checkpoint, and the third has no default at all:
+Three arguments are **load-critical**. `build_model` reads two of them with defaults that do
+not match this checkpoint, and the third has no default at all:
 
 | Argument | Must be | If wrong |
 |---|---|---|
@@ -123,13 +123,27 @@ with torch.no_grad():
 ```
 
 Region- and slide-level embeddings are means over node embeddings. To match the published
-numbers, apply the transforms used at training time — `NormalizeData` with the dataset's
-`normalization.json`, then `AddVirtualNode` (both in `modeling/graphist_utils.py`).
+numbers, apply the transforms used at training time: `NormalizeData` with the dataset's
+`normalization.json`, then `AddVirtualNode`. Both are in `modeling/graphist_utils.py`.
 
-Requires `torch >= 2.2`, `torch_geometric >= 2.5`, `numpy`, `pandas`. Verified on torch
-2.10.0+cu128 / PyG 2.7.0; CPU-only inference works.
+### Requirements
 
-> `modeling/models/acm_gin.py` is required even though this model never uses it —
+Inference needs only `torch`, `torch_geometric`, `numpy` and `pandas`.
+
+| | Trained with | Also verified on |
+|---|---|---|
+| Python | 3.10 | 3.10 |
+| torch | 2.2.2 (CUDA 11.8) | 2.10.0 (CUDA 12.8) |
+| torch_geometric | 2.5.2 | 2.7.0 |
+| numpy | 1.26.4 | 2.2.6 |
+| pandas | 2.2.2 | 2.3.3 |
+
+So it runs across torch 2.2 to 2.10 and on both numpy 1.x and 2.x. CPU-only inference works.
+
+Using the released data needs a little more: `h5py` (the embedding files are HDF5),
+`scikit-learn` 1.5+ for the cell-level probe, and `lifelines` 0.30 for the survival analysis.
+
+> `modeling/models/acm_gin.py` is required even though this model never uses it:
 > `edcoder.py` imports it at module level.
 
 ## Graph format
@@ -152,9 +166,9 @@ gives ≈ 0), confirming these weights produced the released embeddings.
 
 ## Licence and citation
 
-Released `cc-by-nc-sa-4.0`. Derived from public cohorts — BACH/ICIAR 2018, BreakHis, BRACS,
+Released `cc-by-nc-sa-4.0`. Derived from public cohorts: BACH/ICIAR 2018, BreakHis, BRACS,
 TCGA-BRCA, PanNuke, NuCLS and [SPIDER-breast](https://huggingface.co/datasets/histai/SPIDER-breast)
-(`cc-by-nc-4.0`, research use only). These are derived cell-level features, not images; the
+(`cc-by-nc-4.0`, research use only). These are derived cell-level features, not images. The
 source cohorts keep their own terms, and `labels/tcga_brca_clinical.tsv` is the open-access GDC
 clinical export. Cite the original cohort papers alongside GrapHist.
 
@@ -172,4 +186,7 @@ clinical export. Cite the original cohort papers alongside GrapHist.
 ```
 
 Built on GraphMAE (Hou et al., 2022), ACM (Luan et al., 2022), GINEConv (Hu et al., 2020) and
-VICReg (Bardes et al., 2022). Developed at [LTS4, EPFL](https://www.epfl.ch/labs/lts4/).
+VICReg (Bardes et al., 2022).
+
+This work was done by **Mahbod Tajdini** and **Tomás Gadea Alcaide** at
+[LTS4, EPFL](https://www.epfl.ch/labs/lts4/).