Transformers
PyTorch
TensorBoard
ONNX
Safetensors
midi_model
music
midi
music generation
midi generation
Instructions to use Library-Mutsumi/midi-model-tv2o-medium with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Library-Mutsumi/midi-model-tv2o-medium with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Library-Mutsumi/midi-model-tv2o-medium", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload folder using huggingface_hub
Browse files- README.md +59 -0
- config.json +105 -0
- generation_config.json +4 -0
- logs/version_0/events.out.tfevents.1727438892.autodl-container-14ba49b3e1-5960be9d.1410.0 +3 -0
- logs/version_1/events.out.tfevents.1727446966.autodl-container-14ba49b3e1-5960be9d.8721.0 +3 -0
- logs/version_2/events.out.tfevents.1727455013.autodl-container-14ba49b3e1-5960be9d.11379.0 +3 -0
- logs/version_3/events.out.tfevents.1727487870.autodl-container-14ba49b3e1-5960be9d.19440.0 +3 -0
- logs/version_4/events.out.tfevents.1727492956.autodl-container-14ba49b3e1-5960be9d.1026.0 +3 -0
- logs/version_5/events.out.tfevents.1727511849.autodl-container-14ba49b3e1-5960be9d.6787.0 +3 -0
- logs/version_6/events.out.tfevents.1727559621.autodl-container-14ba49b3e1-5960be9d.21247.0 +3 -0
- logs/version_7/events.out.tfevents.1727561353.autodl-container-14ba49b3e1-5960be9d.1073.0 +3 -0
- logs/version_8/events.out.tfevents.1727576759.autodl-container-14ba49b3e1-5960be9d.5180.0 +3 -0
- logs/version_9/events.out.tfevents.1727610496.autodl-container-14ba49b3e1-5960be9d.13841.0 +3 -0
- model.safetensors +3 -0
- onnx/model_base.onnx +3 -0
- onnx/model_token.onnx +3 -0
- pytorch_model.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- projectlosangeles/Los-Angeles-MIDI-Dataset
|
| 5 |
+
- projectlosangeles/Monster-MIDI-Dataset
|
| 6 |
+
library_name: transformers
|
| 7 |
+
tags:
|
| 8 |
+
- music
|
| 9 |
+
- midi
|
| 10 |
+
- music generation
|
| 11 |
+
- midi generation
|
| 12 |
+
---
|
| 13 |
+
# Model Card for midi-model-tv2o-medium
|
| 14 |
+
|
| 15 |
+
Midi event transformer for music generation.
|
| 16 |
+
|
| 17 |
+
### Model Updates
|
| 18 |
+
|
| 19 |
+
- new tv2o-medium model: "tvo" means MidiTokenizerV2 optimised.
|
| 20 |
+
|
| 21 |
+
## Model Details
|
| 22 |
+
|
| 23 |
+
### Model Description
|
| 24 |
+
|
| 25 |
+
- **Developed by:** SkyTNT
|
| 26 |
+
- **Trained by:** SkyTNT
|
| 27 |
+
- **Model type:** Transformer
|
| 28 |
+
- **License:** apache-2.0
|
| 29 |
+
|
| 30 |
+
### Model Sources
|
| 31 |
+
|
| 32 |
+
- **Repository:** https://github.com/SkyTNT/midi-model
|
| 33 |
+
- **Demo:** https://huggingface.co/spaces/skytnt/midi-composer
|
| 34 |
+
|
| 35 |
+
|
| 36 |
+
## Training Details
|
| 37 |
+
|
| 38 |
+
### Training Data
|
| 39 |
+
|
| 40 |
+
- [projectlosangeles/Los-Angeles-MIDI-Dataset](https://huggingface.co/datasets/projectlosangeles/Los-Angeles-MIDI-Dataset)
|
| 41 |
+
- [projectlosangeles/Monster-MIDI-Dataset](https://huggingface.co/datasets/projectlosangeles/Monster-MIDI-Dataset)
|
| 42 |
+
- [SymphonyNet MIDI Dataset](https://symphonynet.github.io/)
|
| 43 |
+
|
| 44 |
+
#### Training Hyperparameters
|
| 45 |
+
|
| 46 |
+
- config: tv2o-medium
|
| 47 |
+
- max-len: 2048 then 4096
|
| 48 |
+
- lr: 1e-4 then 2e-4
|
| 49 |
+
- weight-decay: 0.01
|
| 50 |
+
- batch: 2x2x2
|
| 51 |
+
- bfloat16 precision
|
| 52 |
+
|
| 53 |
+
#### Loss
|
| 54 |
+
- val loss: 0.2159
|
| 55 |
+
|
| 56 |
+
## Files
|
| 57 |
+
|
| 58 |
+
- model.ckpt: latest model checkpoint
|
| 59 |
+
- onnx/*.onnx: onnx format model
|
config.json
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MIDIModel"
|
| 4 |
+
],
|
| 5 |
+
"model_type": "midi_model",
|
| 6 |
+
"n_embd": 1024,
|
| 7 |
+
"net_config": {
|
| 8 |
+
"hidden_size": 1024,
|
| 9 |
+
"intermediate_size": 4096,
|
| 10 |
+
"max_position_embeddings": 4096,
|
| 11 |
+
"model_type": "llama",
|
| 12 |
+
"num_attention_heads": 16,
|
| 13 |
+
"num_hidden_layers": 12,
|
| 14 |
+
"num_key_value_heads": 16,
|
| 15 |
+
"pad_token_id": 0,
|
| 16 |
+
"use_cache": false,
|
| 17 |
+
"vocab_size": 3406
|
| 18 |
+
},
|
| 19 |
+
"net_token_config": {
|
| 20 |
+
"hidden_size": 1024,
|
| 21 |
+
"intermediate_size": 1024,
|
| 22 |
+
"max_position_embeddings": 4096,
|
| 23 |
+
"model_type": "llama",
|
| 24 |
+
"num_attention_heads": 4,
|
| 25 |
+
"num_hidden_layers": 3,
|
| 26 |
+
"num_key_value_heads": 4,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 3406
|
| 30 |
+
},
|
| 31 |
+
"tokenizer": {
|
| 32 |
+
"bos_id": 1,
|
| 33 |
+
"eos_id": 2,
|
| 34 |
+
"event_parameters": {
|
| 35 |
+
"bpm": 384,
|
| 36 |
+
"channel": 16,
|
| 37 |
+
"controller": 128,
|
| 38 |
+
"dd": 4,
|
| 39 |
+
"duration": 2048,
|
| 40 |
+
"mi": 2,
|
| 41 |
+
"nn": 16,
|
| 42 |
+
"patch": 128,
|
| 43 |
+
"pitch": 128,
|
| 44 |
+
"sf": 15,
|
| 45 |
+
"time1": 128,
|
| 46 |
+
"time2": 16,
|
| 47 |
+
"track": 128,
|
| 48 |
+
"value": 128,
|
| 49 |
+
"velocity": 128
|
| 50 |
+
},
|
| 51 |
+
"events": {
|
| 52 |
+
"control_change": [
|
| 53 |
+
"time1",
|
| 54 |
+
"time2",
|
| 55 |
+
"track",
|
| 56 |
+
"channel",
|
| 57 |
+
"controller",
|
| 58 |
+
"value"
|
| 59 |
+
],
|
| 60 |
+
"key_signature": [
|
| 61 |
+
"time1",
|
| 62 |
+
"time2",
|
| 63 |
+
"track",
|
| 64 |
+
"sf",
|
| 65 |
+
"mi"
|
| 66 |
+
],
|
| 67 |
+
"note": [
|
| 68 |
+
"time1",
|
| 69 |
+
"time2",
|
| 70 |
+
"track",
|
| 71 |
+
"channel",
|
| 72 |
+
"pitch",
|
| 73 |
+
"velocity",
|
| 74 |
+
"duration"
|
| 75 |
+
],
|
| 76 |
+
"patch_change": [
|
| 77 |
+
"time1",
|
| 78 |
+
"time2",
|
| 79 |
+
"track",
|
| 80 |
+
"channel",
|
| 81 |
+
"patch"
|
| 82 |
+
],
|
| 83 |
+
"set_tempo": [
|
| 84 |
+
"time1",
|
| 85 |
+
"time2",
|
| 86 |
+
"track",
|
| 87 |
+
"bpm"
|
| 88 |
+
],
|
| 89 |
+
"time_signature": [
|
| 90 |
+
"time1",
|
| 91 |
+
"time2",
|
| 92 |
+
"track",
|
| 93 |
+
"nn",
|
| 94 |
+
"dd"
|
| 95 |
+
]
|
| 96 |
+
},
|
| 97 |
+
"max_token_seq": 8,
|
| 98 |
+
"optimise_midi": true,
|
| 99 |
+
"pad_id": 0,
|
| 100 |
+
"version": "v2",
|
| 101 |
+
"vocab_size": 3406
|
| 102 |
+
},
|
| 103 |
+
"torch_dtype": "bfloat16",
|
| 104 |
+
"transformers_version": "4.44.2"
|
| 105 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.44.2"
|
| 4 |
+
}
|
logs/version_0/events.out.tfevents.1727438892.autodl-container-14ba49b3e1-5960be9d.1410.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4500b6c32dcea8a385f188f4a7f5aabe394cd467a7a47b70a0001cd5a2e5b9e4
|
| 3 |
+
size 650173
|
logs/version_1/events.out.tfevents.1727446966.autodl-container-14ba49b3e1-5960be9d.8721.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9c9bb134cee6d6e7fc818631bc48087a9e87ade171c73ed18f075936e21d6e2
|
| 3 |
+
size 1533662
|
logs/version_2/events.out.tfevents.1727455013.autodl-container-14ba49b3e1-5960be9d.11379.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f638df8693bca62a0b72f4c8f06434fb43ec5165edf0dd08d3866a7c0b169fbe
|
| 3 |
+
size 8579818
|
logs/version_3/events.out.tfevents.1727487870.autodl-container-14ba49b3e1-5960be9d.19440.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c5db26c660c13a267ea600b72089203c3bd66277c3acdeaa65a314a9f0af8b7
|
| 3 |
+
size 1259604
|
logs/version_4/events.out.tfevents.1727492956.autodl-container-14ba49b3e1-5960be9d.1026.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7e7a14fe3356b57efbbd41aaa91d5814cef92df2fd4ed8580938d1f2436372f
|
| 3 |
+
size 4943632
|
logs/version_5/events.out.tfevents.1727511849.autodl-container-14ba49b3e1-5960be9d.6787.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7be351e35cc7516f82f6f6e4e14047fb37dc514e10b0c2e25e83568bbdc80a4
|
| 3 |
+
size 8861276
|
logs/version_6/events.out.tfevents.1727559621.autodl-container-14ba49b3e1-5960be9d.21247.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28549a476acb52ab7eced0d8c0bb45da1da5ffe8fdd9fbffb24f29a3309082be
|
| 3 |
+
size 467102
|
logs/version_7/events.out.tfevents.1727561353.autodl-container-14ba49b3e1-5960be9d.1073.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4102cd846f6a1c39afddda58842fa1a56650945a3a4c98a319dd48b8c6ef5f1
|
| 3 |
+
size 5446371
|
logs/version_8/events.out.tfevents.1727576759.autodl-container-14ba49b3e1-5960be9d.5180.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3bd8a475be99d1d0ae4fc33a7396fb651e42c67636c8ca6dc6a573305bcba37c
|
| 3 |
+
size 12285950
|
logs/version_9/events.out.tfevents.1727610496.autodl-container-14ba49b3e1-5960be9d.13841.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e983c154355d4e5ac68662567f99b5bbee156a455f98a44b2fdfd6bc17c9297
|
| 3 |
+
size 3384064
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82ac8b2217f8f66f79737e444fe60c686d3cbfee54b0c8ef717f701213bbbb83
|
| 3 |
+
size 467701064
|
onnx/model_base.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1af68e16a0509936caeb70d322ab30619d784c7af72df019f3a59276c67c1bc4
|
| 3 |
+
size 821713887
|
onnx/model_token.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cce9376abe8f5c8d51d1b110580cc8501d38679af55ebd5a332e757e62ab9911
|
| 3 |
+
size 116661381
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88fb164517d86c3c1433a1ab315b25c0d271443b81d73a101bee3c46f09779e8
|
| 3 |
+
size 467731642
|