DevPanda004
commited on
Commit
•
66be604
1
Parent(s):
5fcdbb7
End of training
Browse files- README.md +3 -1
- trainer_state.json +110 -33
README.md
CHANGED
@@ -3,6 +3,8 @@ library_name: peft
|
|
3 |
license: cc-by-nc-4.0
|
4 |
base_model: facebook/musicgen-melody
|
5 |
tags:
|
|
|
|
|
6 |
- generated_from_trainer
|
7 |
model-index:
|
8 |
- name: musicgen-melody-indian
|
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
# musicgen-melody-indian
|
16 |
|
17 |
-
This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on
|
18 |
|
19 |
## Model description
|
20 |
|
|
|
3 |
license: cc-by-nc-4.0
|
4 |
base_model: facebook/musicgen-melody
|
5 |
tags:
|
6 |
+
- text-to-audio
|
7 |
+
- indian
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: musicgen-melody-indian
|
|
|
16 |
|
17 |
# musicgen-melody-indian
|
18 |
|
19 |
+
This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the DevPanda004/indian-music-and-metadata-3 dataset.
|
20 |
|
21 |
## Model description
|
22 |
|
trainer_state.json
CHANGED
@@ -1,53 +1,130 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 1.
|
5 |
-
"eval_steps":
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
-
"epoch": 0.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 0.
|
15 |
-
"loss":
|
16 |
-
"step":
|
17 |
},
|
18 |
{
|
19 |
-
"epoch": 0.
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate": 0.
|
22 |
-
"loss":
|
23 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
},
|
25 |
{
|
26 |
-
"epoch": 1.
|
27 |
-
"grad_norm":
|
28 |
-
"learning_rate":
|
29 |
-
"loss":
|
30 |
-
"step":
|
31 |
},
|
32 |
{
|
33 |
-
"epoch": 1.
|
34 |
-
"grad_norm":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
"learning_rate": 0.0,
|
36 |
-
"loss":
|
37 |
-
"step":
|
38 |
},
|
39 |
{
|
40 |
-
"epoch": 1.
|
41 |
-
"step":
|
42 |
-
"total_flos":
|
43 |
-
"train_loss":
|
44 |
-
"train_runtime":
|
45 |
-
"train_samples_per_second": 0.
|
46 |
-
"train_steps_per_second": 0.
|
47 |
}
|
48 |
],
|
49 |
-
"logging_steps":
|
50 |
-
"max_steps":
|
51 |
"num_input_tokens_seen": 0,
|
52 |
"num_train_epochs": 2,
|
53 |
"save_steps": 500,
|
@@ -63,8 +140,8 @@
|
|
63 |
"attributes": {}
|
64 |
}
|
65 |
},
|
66 |
-
"total_flos":
|
67 |
-
"train_batch_size":
|
68 |
"trial_name": null,
|
69 |
"trial_params": null
|
70 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.96,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 30,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.128,
|
13 |
+
"grad_norm": 13.701993942260742,
|
14 |
+
"learning_rate": 0.0001866666666666667,
|
15 |
+
"loss": 68.1376,
|
16 |
+
"step": 2
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.256,
|
20 |
+
"grad_norm": 14.573441505432129,
|
21 |
+
"learning_rate": 0.00017333333333333334,
|
22 |
+
"loss": 63.6084,
|
23 |
+
"step": 4
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.384,
|
27 |
+
"grad_norm": 12.23367977142334,
|
28 |
+
"learning_rate": 0.00016,
|
29 |
+
"loss": 60.5103,
|
30 |
+
"step": 6
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.512,
|
34 |
+
"grad_norm": 11.008517265319824,
|
35 |
+
"learning_rate": 0.00014666666666666666,
|
36 |
+
"loss": 58.3959,
|
37 |
+
"step": 8
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.64,
|
41 |
+
"grad_norm": 9.491394996643066,
|
42 |
+
"learning_rate": 0.00013333333333333334,
|
43 |
+
"loss": 57.2558,
|
44 |
+
"step": 10
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.768,
|
48 |
+
"grad_norm": 7.618191719055176,
|
49 |
+
"learning_rate": 0.00012,
|
50 |
+
"loss": 55.359,
|
51 |
+
"step": 12
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.896,
|
55 |
+
"grad_norm": 6.041190147399902,
|
56 |
+
"learning_rate": 0.00010666666666666667,
|
57 |
+
"loss": 55.4167,
|
58 |
+
"step": 14
|
59 |
},
|
60 |
{
|
61 |
+
"epoch": 1.064,
|
62 |
+
"grad_norm": 11.27497673034668,
|
63 |
+
"learning_rate": 9.333333333333334e-05,
|
64 |
+
"loss": 70.263,
|
65 |
+
"step": 16
|
66 |
},
|
67 |
{
|
68 |
+
"epoch": 1.192,
|
69 |
+
"grad_norm": 7.6353678703308105,
|
70 |
+
"learning_rate": 8e-05,
|
71 |
+
"loss": 55.4048,
|
72 |
+
"step": 18
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 1.32,
|
76 |
+
"grad_norm": 8.388099670410156,
|
77 |
+
"learning_rate": 6.666666666666667e-05,
|
78 |
+
"loss": 55.5403,
|
79 |
+
"step": 20
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 1.448,
|
83 |
+
"grad_norm": 5.153501987457275,
|
84 |
+
"learning_rate": 5.333333333333333e-05,
|
85 |
+
"loss": 54.9104,
|
86 |
+
"step": 22
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.576,
|
90 |
+
"grad_norm": 6.167812824249268,
|
91 |
+
"learning_rate": 4e-05,
|
92 |
+
"loss": 55.303,
|
93 |
+
"step": 24
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.704,
|
97 |
+
"grad_norm": 5.262896537780762,
|
98 |
+
"learning_rate": 2.6666666666666667e-05,
|
99 |
+
"loss": 54.783,
|
100 |
+
"step": 26
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 1.8319999999999999,
|
104 |
+
"grad_norm": 4.895148277282715,
|
105 |
+
"learning_rate": 1.3333333333333333e-05,
|
106 |
+
"loss": 53.8364,
|
107 |
+
"step": 28
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 1.96,
|
111 |
+
"grad_norm": 9.489398956298828,
|
112 |
"learning_rate": 0.0,
|
113 |
+
"loss": 55.1209,
|
114 |
+
"step": 30
|
115 |
},
|
116 |
{
|
117 |
+
"epoch": 1.96,
|
118 |
+
"step": 30,
|
119 |
+
"total_flos": 470062485745920.0,
|
120 |
+
"train_loss": 58.2563720703125,
|
121 |
+
"train_runtime": 1682.7296,
|
122 |
+
"train_samples_per_second": 0.594,
|
123 |
+
"train_steps_per_second": 0.018
|
124 |
}
|
125 |
],
|
126 |
+
"logging_steps": 2,
|
127 |
+
"max_steps": 30,
|
128 |
"num_input_tokens_seen": 0,
|
129 |
"num_train_epochs": 2,
|
130 |
"save_steps": 500,
|
|
|
140 |
"attributes": {}
|
141 |
}
|
142 |
},
|
143 |
+
"total_flos": 470062485745920.0,
|
144 |
+
"train_batch_size": 4,
|
145 |
"trial_name": null,
|
146 |
"trial_params": null
|
147 |
}
|