End of training
Browse files- README.md +3 -1
- trainer_state.json +42 -70
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-lora-dnb-colab-gs-3-lr-0.0001
|
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
# musicgen-melody-lora-dnb-colab-gs-3-lr-0.0001
|
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 |
+
- dnb
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: musicgen-melody-lora-dnb-colab-gs-3-lr-0.0001
|
|
|
16 |
|
17 |
# musicgen-melody-lora-dnb-colab-gs-3-lr-0.0001
|
18 |
|
19 |
+
This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the istvanj/ncs-dnb dataset.
|
20 |
|
21 |
## Model description
|
22 |
|
trainer_state.json
CHANGED
@@ -1,125 +1,97 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
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": 0.26666666666666666,
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 9.
|
15 |
-
"loss": 9.
|
16 |
"step": 2
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.5333333333333333,
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate": 8.
|
22 |
-
"loss": 9.
|
23 |
"step": 4
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.8,
|
27 |
-
"grad_norm":
|
28 |
-
"learning_rate": 7.
|
29 |
-
"loss": 9.
|
30 |
"step": 6
|
31 |
},
|
32 |
{
|
33 |
"epoch": 1.0,
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate":
|
36 |
-
"loss":
|
37 |
"step": 8
|
38 |
},
|
39 |
{
|
40 |
"epoch": 1.2666666666666666,
|
41 |
-
"grad_norm":
|
42 |
-
"learning_rate":
|
43 |
-
"loss":
|
44 |
"step": 10
|
45 |
},
|
46 |
{
|
47 |
"epoch": 1.5333333333333332,
|
48 |
-
"grad_norm":
|
49 |
-
"learning_rate":
|
50 |
-
"loss":
|
51 |
"step": 12
|
52 |
},
|
53 |
{
|
54 |
"epoch": 1.8,
|
55 |
-
"grad_norm":
|
56 |
-
"learning_rate":
|
57 |
-
"loss":
|
58 |
"step": 14
|
59 |
},
|
60 |
{
|
61 |
"epoch": 2.0,
|
62 |
-
"grad_norm":
|
63 |
-
"learning_rate":
|
64 |
-
"loss":
|
65 |
"step": 16
|
66 |
},
|
67 |
{
|
68 |
"epoch": 2.2666666666666666,
|
69 |
-
"grad_norm":
|
70 |
-
"learning_rate":
|
71 |
-
"loss":
|
72 |
"step": 18
|
73 |
},
|
74 |
{
|
75 |
"epoch": 2.533333333333333,
|
76 |
-
"grad_norm":
|
77 |
-
"learning_rate":
|
78 |
-
"loss":
|
79 |
"step": 20
|
80 |
},
|
81 |
{
|
82 |
-
"epoch": 2.
|
83 |
-
"
|
84 |
-
"
|
85 |
-
"
|
86 |
-
"
|
87 |
-
|
88 |
-
|
89 |
-
"epoch": 3.0,
|
90 |
-
"grad_norm": 5.326416969299316,
|
91 |
-
"learning_rate": 1.4285714285714286e-06,
|
92 |
-
"loss": 7.1593,
|
93 |
-
"step": 24
|
94 |
-
},
|
95 |
-
{
|
96 |
-
"epoch": 3.2666666666666666,
|
97 |
-
"grad_norm": 9.835756301879883,
|
98 |
-
"learning_rate": 7.142857142857143e-07,
|
99 |
-
"loss": 9.5721,
|
100 |
-
"step": 26
|
101 |
-
},
|
102 |
-
{
|
103 |
-
"epoch": 3.533333333333333,
|
104 |
-
"grad_norm": 9.394722938537598,
|
105 |
-
"learning_rate": 0.0,
|
106 |
-
"loss": 9.5758,
|
107 |
-
"step": 28
|
108 |
-
},
|
109 |
-
{
|
110 |
-
"epoch": 3.533333333333333,
|
111 |
-
"step": 28,
|
112 |
-
"total_flos": 108812813622552.0,
|
113 |
-
"train_loss": 9.103422130857195,
|
114 |
-
"train_runtime": 670.6646,
|
115 |
-
"train_samples_per_second": 0.716,
|
116 |
-
"train_steps_per_second": 0.042
|
117 |
}
|
118 |
],
|
119 |
"logging_steps": 2,
|
120 |
-
"max_steps":
|
121 |
"num_input_tokens_seen": 0,
|
122 |
-
"num_train_epochs":
|
123 |
"save_steps": 500,
|
124 |
"stateful_callbacks": {
|
125 |
"TrainerControl": {
|
@@ -133,7 +105,7 @@
|
|
133 |
"attributes": {}
|
134 |
}
|
135 |
},
|
136 |
-
"total_flos":
|
137 |
"train_batch_size": 2,
|
138 |
"trial_name": null,
|
139 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.6666666666666665,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 21,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.26666666666666666,
|
13 |
+
"grad_norm": 8.077447891235352,
|
14 |
+
"learning_rate": 9.047619047619048e-05,
|
15 |
+
"loss": 9.6711,
|
16 |
"step": 2
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.5333333333333333,
|
20 |
+
"grad_norm": 10.438846588134766,
|
21 |
+
"learning_rate": 8.095238095238096e-05,
|
22 |
+
"loss": 9.4926,
|
23 |
"step": 4
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.8,
|
27 |
+
"grad_norm": 11.99673080444336,
|
28 |
+
"learning_rate": 7.142857142857143e-05,
|
29 |
+
"loss": 9.2985,
|
30 |
"step": 6
|
31 |
},
|
32 |
{
|
33 |
"epoch": 1.0,
|
34 |
+
"grad_norm": 8.44992733001709,
|
35 |
+
"learning_rate": 6.19047619047619e-05,
|
36 |
+
"loss": 6.8146,
|
37 |
"step": 8
|
38 |
},
|
39 |
{
|
40 |
"epoch": 1.2666666666666666,
|
41 |
+
"grad_norm": 15.83218002319336,
|
42 |
+
"learning_rate": 5.2380952380952384e-05,
|
43 |
+
"loss": 8.8302,
|
44 |
"step": 10
|
45 |
},
|
46 |
{
|
47 |
"epoch": 1.5333333333333332,
|
48 |
+
"grad_norm": 16.710369110107422,
|
49 |
+
"learning_rate": 4.761904761904762e-05,
|
50 |
+
"loss": 8.5759,
|
51 |
"step": 12
|
52 |
},
|
53 |
{
|
54 |
"epoch": 1.8,
|
55 |
+
"grad_norm": 19.962966918945312,
|
56 |
+
"learning_rate": 3.809523809523809e-05,
|
57 |
+
"loss": 8.3464,
|
58 |
"step": 14
|
59 |
},
|
60 |
{
|
61 |
"epoch": 2.0,
|
62 |
+
"grad_norm": 9.256256103515625,
|
63 |
+
"learning_rate": 2.857142857142857e-05,
|
64 |
+
"loss": 6.1951,
|
65 |
"step": 16
|
66 |
},
|
67 |
{
|
68 |
"epoch": 2.2666666666666666,
|
69 |
+
"grad_norm": 18.415756225585938,
|
70 |
+
"learning_rate": 1.9047619047619046e-05,
|
71 |
+
"loss": 8.0719,
|
72 |
"step": 18
|
73 |
},
|
74 |
{
|
75 |
"epoch": 2.533333333333333,
|
76 |
+
"grad_norm": 14.65276050567627,
|
77 |
+
"learning_rate": 9.523809523809523e-06,
|
78 |
+
"loss": 7.9604,
|
79 |
"step": 20
|
80 |
},
|
81 |
{
|
82 |
+
"epoch": 2.6666666666666665,
|
83 |
+
"step": 21,
|
84 |
+
"total_flos": 82187025809664.0,
|
85 |
+
"train_loss": 8.305864515758696,
|
86 |
+
"train_runtime": 507.9977,
|
87 |
+
"train_samples_per_second": 0.709,
|
88 |
+
"train_steps_per_second": 0.041
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
}
|
90 |
],
|
91 |
"logging_steps": 2,
|
92 |
+
"max_steps": 21,
|
93 |
"num_input_tokens_seen": 0,
|
94 |
+
"num_train_epochs": 3,
|
95 |
"save_steps": 500,
|
96 |
"stateful_callbacks": {
|
97 |
"TrainerControl": {
|
|
|
105 |
"attributes": {}
|
106 |
}
|
107 |
},
|
108 |
+
"total_flos": 82187025809664.0,
|
109 |
"train_batch_size": 2,
|
110 |
"trial_name": null,
|
111 |
"trial_params": null
|