Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
54 commits
Select commit Hold shift + click to select a range
ed63499
feat: neutrophils
Adames4 May 5, 2026
0adec3f
refactor: project_name -> ml
Adames4 May 5, 2026
517d919
chore: new mlkit
Adames4 May 5, 2026
e0d214c
feat: labels
Adames4 May 5, 2026
6f73eb9
refactor: labels
Adames4 May 5, 2026
fd22f86
feat: types
Adames4 May 5, 2026
4e12f9c
feat: tiles dataset
Adames4 May 5, 2026
c7d7297
feat: tiles dataframe to hfdataset
Adames4 May 5, 2026
f9969c6
feat: neutrophils ml
Adames4 May 5, 2026
c4fd74d
feat: datasets
Adames4 May 6, 2026
a56b3f4
feat: neutrophil dataset conf
Adames4 May 6, 2026
49d4cff
feat: datasets
Adames4 May 7, 2026
182350f
feat: loss
Adames4 May 7, 2026
44c981a
feat: modeling
Adames4 May 7, 2026
b615d2f
feat: ml
Adames4 May 7, 2026
81e8eb0
feat: postprocessing WIP
Adames4 May 7, 2026
773e4df
feat: ensembling confs
Adames4 May 7, 2026
f34c711
feat: configs
Adames4 May 7, 2026
d9ed16d
feat: postprocessing
Adames4 May 7, 2026
eec5bb1
chore: format
Adames4 May 7, 2026
b653c46
feat: mil
Adames4 May 7, 2026
838125a
feat: ml
Adames4 May 7, 2026
617c720
feat: bags
Adames4 May 7, 2026
34d19a7
feat: qc filtering
Adames4 May 7, 2026
24d11d6
fix: bags
Adames4 May 8, 2026
6c4cea6
feat: more tilings
Adames4 May 8, 2026
a494ced
feat: folds
Adames4 May 8, 2026
e497a93
feat: sampler WIP
Adames4 May 8, 2026
406e83b
feat: sampling
Adames4 May 8, 2026
8932659
feat: datamodule
Adames4 May 8, 2026
53fc65e
feat: ml experiments
Adames4 May 8, 2026
af90f47
feat: callback
Adames4 May 8, 2026
70540f1
feat: ml configs
Adames4 May 8, 2026
3fffc85
feat: postprocessing confs
Adames4 May 8, 2026
8d5235f
feat: uris
Adames4 May 8, 2026
29633f7
feat: artifacts deleted
Adames4 May 8, 2026
bb71d72
feature: scripts
Adames4 May 8, 2026
cc369f2
fix: bugs
Adames4 May 8, 2026
e5bf6d1
fix: ruff
Adames4 May 8, 2026
3548ffc
chore: ruff
Adames4 May 8, 2026
c993466
fix: ruff
Adames4 May 8, 2026
d395ca1
chore: mypy
Adames4 May 8, 2026
fa0d5e7
chore: mypy
Adames4 May 8, 2026
a3165e5
feat: readme
Adames4 May 8, 2026
7aad84f
feat: qc thresholds
Adames4 May 8, 2026
e114530
Experiment/ikem validation (#14)
Adames4 Aug 29, 2026
c1ebb33
feat: update configuration files and add new prediction setup
Adames4 Aug 29, 2026
93bfe9a
fix: update threshold keys in bags, embeddings, and tiles configurations
Adames4 Aug 29, 2026
f4e7a61
fix: remove redundant mode entries from ikem validation configuration…
Adames4 Aug 29, 2026
c9a655f
fix: add missing mode entry in ikem validation configuration files
Adames4 Aug 29, 2026
58ffbc2
fix: correct target dataset for prediction in bags configuration
Adames4 Aug 29, 2026
7c02847
fix: update number of classes in neutrophils configuration to match m…
Adames4 Aug 29, 2026
a2ca1db
fix: add missing checkpoint file extension in nancy_high configuration
Adames4 Aug 29, 2026
e38f750
feat: add prediction scripts and configurations for ensembling and ma…
Adames4 Aug 29, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
2 changes: 1 addition & 1 deletion .ruff.toml
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ extend-ignore = [
"D106", # missing docstring in public nested class
"D107", # missing docstring in __init__
"N812", # lowercase imported as non lowercase
"TCH002", # move third-party into a type-checking block
"TC002", # move third-party into a type-checking block

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The rule code TC002 is invalid in Ruff. The correct rule code for ignoring third-party imports in type-checking blocks is TCH002 (as part of the TCH / flake8-type-checking ruleset). Changing this to TC002 will cause Ruff to fail with a configuration parsing error.

Suggested change
"TC002", # move third-party into a type-checking block
"TCH002", # move third-party into a type-checking block

"F722", # jaxtyping
]

Expand Down
149 changes: 148 additions & 1 deletion README.md
Original file line number Diff line number Diff line change
@@ -1 +1,148 @@
# Machine Learning Template
# Automated Grading of Ulcerative Colitis from H&E Whole Slide Images

This repository contains the machine learning pipeline developed for automated histological grading of ulcerative colitis (UC) biopsies using the Nancy Histological Index (NHI). The approach frames NHI grading as a hierarchical classification problem: a binary neutrophil-detection task routes each whole-slide image through either a low-grade (NHI 0–2) or a high-grade (NHI 2–4) classifier. All models are based on pretrained pathology foundation models and use attention-based multiple instance learning (MIL) over tile embeddings, making them applicable to gigapixel slides without requiring tile-level annotations. Experiments evaluate multiple foundation models (prov-gigapath, UNI, UNI2, Virchow, Virchow2), MIL against tile-level approaches, ordered regression, and ensembling/confidence strategies over a multi-institution cohort (IKEM, FTN, KNL).

> **Note: This code is not runnable as-is.** The underlying H&E whole-slide image data and derived embeddings are sensitive patient data from clinical institutions and are not included in this repository. The code is provided for reference and reproducibility of the methodology only.

---

## Repository Structure

```
.
├── configs/ # Hydra configuration tree
│ ├── dataset/ # Dataset split URIs (embeddings, slides, tiles)
│ ├── checkpoints/ # Checkpoint paths for test/inference runs
│ ├── predictions/ # MLflow prediction artifact URIs (per institution/fold)
│ └── experiment/
│ ├── ml/ # ML training & test experiment configs (Exp I–VI + final)
│ └── postprocessing/ # Postprocessing experiment configs (Exp VII–VIII + final)
├── ml/ # PyTorch Lightning modules & entry point
│ ├── __main__.py # Entry point: `python -m ml`
│ ├── base.py # BaseModule (tile-level training loop)
│ ├── classification.py # Tile-level softmax classification
│ ├── ordered_regression.py # Cumulative link ordered regression
│ ├── mil.py # Attention MIL (bag-level)
│ ├── neutrophils.py # Binary neutrophil detection (tile-level)
│ └── modeling/ # Building blocks (heads, normalization)
├── preprocessing/ # Data preparation pipeline
│ ├── tiling.py # Tile extraction from WSIs
│ ├── tissue_masks.py # Tissue segmentation
│ ├── quality_control.py # Tile quality filtering
│ ├── embeddings.py # Foundation model feature extraction
│ ├── neutrophils.py # Neutrophil pseudo-label extraction
│ ├── create_dataset.py # Assemble HuggingFace dataset
│ └── split_dataset.py # Train/val/test split assignment
├── postprocessing/ # Slide-level aggregation & confidence
│ ├── ensembling.py # Soft majority vote + hierarchical ensembling
│ ├── markov_chain_confidence.py # Markov chain absorption confidence scores
│ ├── ensembling_predict.py # Ensembling for unlabeled cohorts (no metrics)
│ └── markov_chain_confidence_predict.py # Confidence scores for unlabeled cohorts
├── scripts/ # Kubernetes job submission scripts
│ ├── ml/ # train.py, test.py, neutrophils.py
│ ├── preprocessing/ # Preprocessing job scripts
│ └── postprocessing/ # Postprocessing job scripts
├── pyproject.toml
└── thesis.pdf # Thesis manuscript
```

---

## Setup

Requires Python 3.12 and [`uv`](https://github.com/astral-sh/uv).

```bash
uv sync --frozen
```

---

## Pipeline Overview

The full pipeline runs in four stages:

### 1. Preprocessing

Tiles are extracted from WSIs, tissue-masked, quality-filtered, and encoded by a foundation model into fixed-dimensional embeddings. Datasets are assembled as HuggingFace Datasets and registered in MLflow.

```bash
# Tissue segmentation masks
uv run --active -m preprocessing.tissue_masks +dataset=processed/...

# Tile extraction from WSIs
uv run --active -m preprocessing.tiling +dataset=processed_w_masks/... +experiment=preprocessing/tiling/...

# Tile quality filtering
uv run -m preprocessing.quality_control +dataset=processed/...

# Foundation model feature extraction
uv run --active -m preprocessing.embeddings +dataset=tiled/...

# Assemble HuggingFace dataset
uv run -m preprocessing.create_dataset +dataset=raw/...

# Train/val/test split assignment
uv run -m preprocessing.split_dataset +experiment=preprocessing/split_dataset/...
```

### 2. ML Training

Models are trained via Hydra experiment configs. The entry point is `python -m ml` and an experiment config must be selected:

```bash
# Example: train Experiment IV (MIL) fold 1
uv run -m ml +experiment=ml/experiment_iv_mil_and_hierarchical_modeling/nancy_high/train/fold_1

# Example: train final model (all institutions combined)
uv run -m ml +experiment=ml/final/nancy_high/train
```

### 3. ML Testing / Inference

```bash
# Example: test Experiment IV fold 1
uv run -m ml +experiment=ml/experiment_iv_mil_and_hierarchical_modeling/nancy_high/test/fold_1

# Neutrophil detection (produces binary slide-level predictions)
uv run -m ml.neutrophils +experiment=ml/experiment_vi_neutrophil_detection/run
```

### 4. Postprocessing

Slide-level predictions from the three hierarchical tasks are combined into final NHI grades and confidence scores:

```bash
# Ensembling (soft majority vote + hierarchical routing)
uv run -m postprocessing.ensembling +postprocessing=ensembling

# Markov chain absorption confidence
uv run -m postprocessing.markov_chain_confidence +postprocessing=markov_chain_confidence
```

For unlabeled cohorts (no ground-truth NHI grades, e.g. external validation), use the
`_predict` variants, which skip label loading and metric computation and only emit
predictions / confidence scores:

```bash
uv run -m postprocessing.ensembling_predict +postprocessing=ensembling_predict
uv run -m postprocessing.markov_chain_confidence_predict +postprocessing=markov_chain_confidence_predict
```

---

## Experiments

| # | Name | Config path | Description |
|---|------|-------------|-------------|
| I | Baseline Classification | `experiment/ml/experiment_i_baseline_classification` | VGG16 tile-level classifier trained on raw pixel tiles |
| II | Ordered Regression | `experiment/ml/experiment_ii_ordered_regression` | Cumulative link loss for NHI ordinal structure |
| III | Pathology Foundation Models | `experiment/ml/experiment_iii_pathology_foundation_model` | Linear probe on frozen prov-gigapath embeddings |
| IV | MIL & Hierarchical Modeling | `experiment/ml/experiment_iv_mil_and_hierarchical_modeling` | Attention MIL; three-task hierarchical decomposition |
| V | More Foundation Models | `experiment/ml/experiment_v_more_foundation_models` | MIL sweep over all foundation models |
| VI | Neutrophil Detection | `experiment/ml/experiment_vi_neutrophil_detection` | Binary tile-level neutrophil classifier |
| VII | Ensembling | `experiment/postprocessing/experiment_vii_ensembling` | Soft majority vote vs. hierarchical routing |
| VIII | Markov Chain Confidence | `experiment/postprocessing/experiment_viii_markov_chain_model_aggregation` | Absorption distribution confidence (entropy / herfindahl / std) |
| — | Final Model | `experiment/ml/final` | Attention MIL (virchow2) trained on all three institutions |
Comment on lines +138 to +146

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The configuration paths listed in the table are missing the configs/ prefix. According to the repository structure, the experiment configurations are located under configs/experiment/.... Updating these paths will make it easier for users to locate the files in the repository.

Suggested change
| I | Baseline Classification | `experiment/ml/experiment_i_baseline_classification` | VGG16 tile-level classifier trained on raw pixel tiles |
| II | Ordered Regression | `experiment/ml/experiment_ii_ordered_regression` | Cumulative link loss for NHI ordinal structure |
| III | Pathology Foundation Models | `experiment/ml/experiment_iii_pathology_foundation_model` | Linear probe on frozen prov-gigapath embeddings |
| IV | MIL & Hierarchical Modeling | `experiment/ml/experiment_iv_mil_and_hierarchical_modeling` | Attention MIL; three-task hierarchical decomposition |
| V | More Foundation Models | `experiment/ml/experiment_v_more_foundation_models` | MIL sweep over all foundation models |
| VI | Neutrophil Detection | `experiment/ml/experiment_vi_neutrophil_detection` | Binary tile-level neutrophil classifier |
| VII | Ensembling | `experiment/postprocessing/experiment_vii_ensembling` | Soft majority vote vs. hierarchical routing |
| VIII | Markov Chain Confidence | `experiment/postprocessing/experiment_viii_markov_chain_model_aggregation` | Absorption distribution confidence (entropy / herfindahl / std) |
| — | Final Model | `experiment/ml/final` | Attention MIL (virchow2) trained on all three institutions |
| I | Baseline Classification | `configs/experiment/ml/experiment_i_baseline_classification` | VGG16 tile-level classifier trained on raw pixel tiles |
| II | Ordered Regression | `configs/experiment/ml/experiment_ii_ordered_regression` | Cumulative link loss for NHI ordinal structure |
| III | Pathology Foundation Models | `configs/experiment/ml/experiment_iii_pathology_foundation_model` | Linear probe on frozen prov-gigapath embeddings |
| IV | MIL & Hierarchical Modeling | `configs/experiment/ml/experiment_iv_mil_and_hierarchical_modeling` | Attention MIL; three-task hierarchical decomposition |
| V | More Foundation Models | `configs/experiment/ml/experiment_v_more_foundation_models` | MIL sweep over all foundation models |
| VI | Neutrophil Detection | `configs/experiment/ml/experiment_vi_neutrophil_detection` | Binary tile-level neutrophil classifier |
| VII | Ensembling | `configs/experiment/postprocessing/experiment_vii_ensembling` | Soft majority vote vs. hierarchical routing |
| VIII | Markov Chain Confidence | `configs/experiment/postprocessing/experiment_viii_markov_chain_model_aggregation` | Absorption distribution confidence (entropy / herfindahl / std) |
| — | Final Model | `configs/experiment/ml/final` | Attention MIL (virchow2) trained on all three institutions |


All cross-validation experiments run 5 folds over the IKEM institution. Final evaluation uses the held-out test split across IKEM, FTN, and KNL.
2 changes: 2 additions & 0 deletions configs/base.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,5 +3,7 @@ defaults:
- logger: mlflow
- _self_

project_dir: /mnt/projects/inflammatory_bowel_disease/ulcerative_colitis

metadata:
experiment_name: Ulcerative Colitis
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
checkpoint: ???
3 changes: 3 additions & 0 deletions configs/checkpoints/final/nancy_high.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# @package _global_

checkpoint: mlflow-artifacts:/86/ac693e2ee17e4254a30786c0668e7af4/artifacts/checkpoints/epoch=299-step=176100/checkpoint.ckpt
3 changes: 3 additions & 0 deletions configs/checkpoints/final/nancy_low.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# @package _global_

checkpoint: mlflow-artifacts:/86/452e6f1bf18c45d3bc1d2ae9b44154ae/artifacts/checkpoints/epoch=299-step=176100/checkpoint.ckpt
3 changes: 3 additions & 0 deletions configs/checkpoints/final/neutrophils.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# @package _global_

checkpoint: mlflow-artifacts:/86/5f9d50d8dccf45e6b8a80d8d3b0f773c/artifacts/checkpoints/epoch=70-step=41677/checkpoint.ckpt
9 changes: 9 additions & 0 deletions configs/dataset/embeddings/ftn/1_prov-gigapath.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
defaults:
- /dataset/tiled/ftn/1_224@_here_
- _self_

mlflow_uris:
embeddings:
train: ???
test_preliminary: ???
test_final: ???
9 changes: 9 additions & 0 deletions configs/dataset/embeddings/ftn/1_uni2.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
defaults:
- /dataset/tiled/ftn/1_224@_here_
- _self_

mlflow_uris:
embeddings:
train: ???
test_preliminary: ???
test_final: ???
9 changes: 9 additions & 0 deletions configs/dataset/embeddings/ftn/1_virchow2.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
defaults:
- /dataset/tiled/ftn/1_224@_here_
- _self_

mlflow_uris:
embeddings:
train: ???
test_preliminary: ???
test_final: ???
Loading
Loading