From 087d94e8d5f61f2216997116a78d0395336ead87 Mon Sep 17 00:00:00 2001
From: Alain Riou <36546630+aRI0U@users.noreply.github.com>
Date: Wed, 17 Jan 2024 19:53:06 +0100
Subject: [PATCH 1/2] Update README.md

---
 README.md | 25 -------------------------
 1 file changed, 25 deletions(-)

diff --git a/README.md b/README.md
index fa53f13..5ab324f 100644
--- a/README.md
+++ b/README.md
@@ -183,31 +183,6 @@ If you want to use this work, please cite:
 ```
 
 
-## TODO
-
-### Research
-
-- Implement confidence score
-- Handle continuous changes in time
-- Handle velocity
-- Smaller model
-- train on MUSDB to make it robust to background music
-- Optimal Transport
-- circular distributions to avoid shitty initialization
-
-
-### Implementation
-
-- Viterbi smoothing
-
-
-### Refactoring
-
-- simplify `datamodules`
-- split `augmentations/cqt` in two files + implement data augmentations directly in this repo
-- general evaluation loop: use same codebase for evaluating all baselines
-
-
 ### Pretrained models
 
 - Train good models, evaluate them and release them

From 616cc9bfc53c94ef3f1df02a8bc449e30a529e92 Mon Sep 17 00:00:00 2001
From: Alain Riou <36546630+aRI0U@users.noreply.github.com>
Date: Wed, 17 Jan 2024 19:53:36 +0100
Subject: [PATCH 2/2] Update README.md

---
 README.md | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/README.md b/README.md
index 5ab324f..927ed8c 100644
--- a/README.md
+++ b/README.md
@@ -182,7 +182,3 @@ If you want to use this work, please cite:
     doi={10.1109/TASLP.2022.3200547}}
 ```
 
-
-### Pretrained models
-
-- Train good models, evaluate them and release them