Hugo Flores Garcia commited on
Commit
74cced7
1 Parent(s): e9fd215

update readme

Browse files
Files changed (1) hide show
  1. README.md +12 -23
README.md CHANGED
@@ -4,20 +4,6 @@ This repository contains recipes for training generative music models on top of
4
 
5
  # Setting up
6
 
7
- install AudioTools
8
-
9
- ```bash
10
- git clone https://github.com/descriptinc/audiotools.git
11
- pip install -e ./audiotools
12
- ```
13
-
14
- install the Descript Audio Codec.
15
-
16
- ```bash
17
- git clone https://github.com/descriptinc/descript-audio-codec.git
18
- pip install -e ./descript-audio-codec
19
- ```
20
-
21
  install VampNet
22
 
23
  ```bash
@@ -40,20 +26,27 @@ First, you'll want to set up your environment
40
  source ./env/env.sh
41
  ```
42
 
43
- ## Staging a Run
44
-
45
- Staging a run makes a copy of all the git-tracked files in the codebase and saves them to a folder for reproducibility. You can then run the training script from the staged folder.
46
 
 
 
47
  ```
48
- stage --name my_run --run_dir /path/to/staging/folder
49
- ```
50
 
51
  ## Training a model
52
 
 
 
53
  ```bash
54
  python scripts/exp/train.py --args.load conf/vampnet.yml --save_path /path/to/checkpoints
55
  ```
56
 
 
 
 
 
57
  See `python scripts/exp/train.py -h` for a list of options.
58
 
59
  ## Fine-tuning
@@ -86,7 +79,3 @@ python demo.py --args.load conf/generated/<fine_tune_name>/interface.yml
86
  ```
87
 
88
 
89
- ## Launching the Gradio Interface
90
- ```bash
91
- python demo.py --args.load conf/interface/spotdl.yml --Interface.device cuda
92
- ```
 
4
 
5
  # Setting up
6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  install VampNet
8
 
9
  ```bash
 
26
  source ./env/env.sh
27
  ```
28
 
29
+ ## Launching the Gradio Interface
30
+ You can launch a gradio UI to play with vampnet.
 
31
 
32
+ ```bash
33
+ python demo.py --args.load conf/interface/spotdl.yml --Interface.device cuda
34
  ```
35
+
36
+ # Training / Fine-tuning
37
 
38
  ## Training a model
39
 
40
+ To train a model, run the following script:
41
+
42
  ```bash
43
  python scripts/exp/train.py --args.load conf/vampnet.yml --save_path /path/to/checkpoints
44
  ```
45
 
46
+ You can edit `conf/vampnet.yml` to change the dataset paths or any training hyperparameters.
47
+
48
+ For coarse2fine models, you can use `conf/c2f.yml` as a starting configuration.
49
+
50
  See `python scripts/exp/train.py -h` for a list of options.
51
 
52
  ## Fine-tuning
 
79
  ```
80
 
81