Upload structure model checkpoint at step 2000
Browse files- checkpoint_2000/README.md +46 -0
- checkpoint_2000/config.json +14 -0
- checkpoint_2000/model.npz +3 -0
- checkpoint_2000/training_state.json +126 -0
checkpoint_2000/README.md
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: mlx
|
3 |
+
pipeline_tag: text-generation
|
4 |
+
tags:
|
5 |
+
- music
|
6 |
+
- midi
|
7 |
+
- generation
|
8 |
+
- mlx
|
9 |
+
- autoregressive
|
10 |
+
- structure
|
11 |
+
- musiclang
|
12 |
+
---
|
13 |
+
|
14 |
+
# okai-musiclang-structure v2.0 - Structure Model
|
15 |
+
|
16 |
+
This is an autoregressive structure generation model for music using MLX.
|
17 |
+
|
18 |
+
## Model Details
|
19 |
+
|
20 |
+
- **Model Type**: Structure Generator (Autoregressive)
|
21 |
+
- **Version**: v2.0
|
22 |
+
- **Step**: 2000
|
23 |
+
- **Architecture**: Transformer with causal language modeling
|
24 |
+
- **Vocabulary Size**: 4796
|
25 |
+
- **Model Dimension**: 256
|
26 |
+
- **Layers**: 6
|
27 |
+
- **Max Sequence Length**: 1024
|
28 |
+
|
29 |
+
## Training Configuration
|
30 |
+
|
31 |
+
- **Batch Size**: 16
|
32 |
+
- **Learning Rate**: 1e-05
|
33 |
+
- **Training Steps**: 2000
|
34 |
+
|
35 |
+
## Usage
|
36 |
+
|
37 |
+
This model generates sequential music structure:
|
38 |
+
- Input: Song control tokens (genre, instruments, etc.)
|
39 |
+
- Output: Sequential bar structure with chords and tonality
|
40 |
+
|
41 |
+
Example generation:
|
42 |
+
```
|
43 |
+
GENRE__ROCK SUBGENRE__ALTERNATIVE START BAR__1 CHORD_DEGREE__1 TONALITY_DEGREE__1 BAR__2 CHORD_DEGREE__4 TONALITY_DEGREE__5 ... WILL_END
|
44 |
+
```
|
45 |
+
|
46 |
+
Generated with MLX framework for Apple Silicon.
|
checkpoint_2000/config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"model_name": "okai-musiclang-structure",
|
3 |
+
"model_version": "v2.0",
|
4 |
+
"model_type": "structure",
|
5 |
+
"global_step": 2000,
|
6 |
+
"architecture": "AutoregressiveTransformer",
|
7 |
+
"training_type": "causal_lm",
|
8 |
+
"vocab_size": 4796,
|
9 |
+
"model_dim": 256,
|
10 |
+
"num_heads": 8,
|
11 |
+
"num_layers": 6,
|
12 |
+
"max_sequence_length": 1024,
|
13 |
+
"dropout": 0.1
|
14 |
+
}
|
checkpoint_2000/model.npz
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bfc8fd7ea5dddf6656b21b9823e7b7b7763603bee6f7e2582984d281481507ae
|
3 |
+
size 29839884
|
checkpoint_2000/training_state.json
ADDED
@@ -0,0 +1,126 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"global_step": 2000,
|
3 |
+
"epoch": 1,
|
4 |
+
"trainer_step": 2000,
|
5 |
+
"learning_rate": 9.999999747378752e-06,
|
6 |
+
"epoch_losses": [
|
7 |
+
1.7293955087661743,
|
8 |
+
1.7601858377456665,
|
9 |
+
1.6190868616104126,
|
10 |
+
1.7451980113983154,
|
11 |
+
1.701377511024475,
|
12 |
+
1.6588249206542969,
|
13 |
+
1.6583818197250366,
|
14 |
+
1.6878631114959717,
|
15 |
+
1.6302950382232666,
|
16 |
+
1.7025562524795532,
|
17 |
+
1.8004159927368164,
|
18 |
+
1.7292814254760742,
|
19 |
+
1.616337776184082,
|
20 |
+
1.7280219793319702,
|
21 |
+
1.823448896408081,
|
22 |
+
1.7120184898376465,
|
23 |
+
1.639157772064209,
|
24 |
+
1.6284191608428955,
|
25 |
+
1.6839994192123413,
|
26 |
+
1.7949339151382446,
|
27 |
+
1.5744112730026245,
|
28 |
+
1.6716513633728027,
|
29 |
+
1.717328429222107,
|
30 |
+
1.7006312608718872,
|
31 |
+
1.859540581703186,
|
32 |
+
1.7632871866226196,
|
33 |
+
1.7942699193954468,
|
34 |
+
1.7329679727554321,
|
35 |
+
1.7112141847610474,
|
36 |
+
1.6165447235107422,
|
37 |
+
1.7208964824676514,
|
38 |
+
1.6240524053573608,
|
39 |
+
1.6348881721496582,
|
40 |
+
1.7209837436676025,
|
41 |
+
1.7384777069091797,
|
42 |
+
1.7252110242843628,
|
43 |
+
1.6180779933929443,
|
44 |
+
1.5901944637298584,
|
45 |
+
1.6428749561309814,
|
46 |
+
1.6507099866867065,
|
47 |
+
1.7259920835494995,
|
48 |
+
1.7451446056365967,
|
49 |
+
1.7164634466171265,
|
50 |
+
1.7105237245559692,
|
51 |
+
1.6539174318313599,
|
52 |
+
1.7445276975631714,
|
53 |
+
1.6990998983383179,
|
54 |
+
1.6163480281829834,
|
55 |
+
1.6803194284439087,
|
56 |
+
1.6707617044448853,
|
57 |
+
1.6227998733520508,
|
58 |
+
1.7251840829849243,
|
59 |
+
1.751766324043274,
|
60 |
+
1.7954998016357422,
|
61 |
+
1.5760107040405273,
|
62 |
+
1.7212122678756714,
|
63 |
+
1.6583250761032104,
|
64 |
+
1.582102656364441,
|
65 |
+
1.8238333463668823,
|
66 |
+
1.692293405532837,
|
67 |
+
1.7631574869155884,
|
68 |
+
1.641127586364746,
|
69 |
+
1.6412501335144043,
|
70 |
+
1.6396362781524658,
|
71 |
+
1.751018762588501,
|
72 |
+
1.6705764532089233,
|
73 |
+
1.765311360359192,
|
74 |
+
1.7680552005767822,
|
75 |
+
1.7106271982192993,
|
76 |
+
1.7239460945129395,
|
77 |
+
1.7113215923309326,
|
78 |
+
1.6063220500946045,
|
79 |
+
1.6960153579711914,
|
80 |
+
1.6591867208480835,
|
81 |
+
1.7286903858184814,
|
82 |
+
1.636628270149231,
|
83 |
+
1.7734651565551758,
|
84 |
+
1.6685333251953125,
|
85 |
+
1.6820003986358643,
|
86 |
+
1.7647629976272583,
|
87 |
+
1.6755094528198242,
|
88 |
+
1.6541610956192017,
|
89 |
+
1.6375068426132202,
|
90 |
+
1.6230283975601196,
|
91 |
+
1.6839603185653687,
|
92 |
+
1.677659273147583,
|
93 |
+
1.7746293544769287,
|
94 |
+
1.804154872894287,
|
95 |
+
1.6957547664642334,
|
96 |
+
1.6372662782669067,
|
97 |
+
1.608318567276001,
|
98 |
+
1.614088773727417,
|
99 |
+
1.6746459007263184,
|
100 |
+
1.6934666633605957,
|
101 |
+
1.6467796564102173,
|
102 |
+
1.6046024560928345,
|
103 |
+
1.6138876676559448
|
104 |
+
],
|
105 |
+
"training_config": {
|
106 |
+
"batch_size": 16,
|
107 |
+
"num_epochs": 4,
|
108 |
+
"learning_rate": 1e-05,
|
109 |
+
"weight_decay": 0.01,
|
110 |
+
"warmup_steps": 1000,
|
111 |
+
"max_grad_norm": 1.0,
|
112 |
+
"eval_steps": 500,
|
113 |
+
"save_steps": 500
|
114 |
+
},
|
115 |
+
"model_config": {
|
116 |
+
"vocab_size": 4796,
|
117 |
+
"model_dim": 256,
|
118 |
+
"num_heads": 8,
|
119 |
+
"num_layers": 6,
|
120 |
+
"max_sequence_length": 1024,
|
121 |
+
"dropout": 0.1
|
122 |
+
},
|
123 |
+
"model_name": "okai-musiclang-structure",
|
124 |
+
"model_version": "v2.0",
|
125 |
+
"model_type": "structure"
|
126 |
+
}
|