End of training
Browse files- README.md +3 -1
- trainer_state.json +53 -151
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-4-lr-2e-05
|
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
# musicgen-melody-lora-dnb-colab-gs-4-lr-2e-05
|
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-4-lr-2e-05
|
|
|
16 |
|
17 |
# musicgen-melody-lora-dnb-colab-gs-4-lr-2e-05
|
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
@@ -1,223 +1,125 @@
|
|
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":
|
15 |
-
"loss": 9.
|
16 |
"step": 2
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.5333333333333333,
|
20 |
-
"grad_norm": 8.
|
21 |
-
"learning_rate":
|
22 |
-
"loss": 9.
|
23 |
"step": 4
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.8,
|
27 |
-
"grad_norm": 8.
|
28 |
-
"learning_rate":
|
29 |
-
"loss": 9.
|
30 |
"step": 6
|
31 |
},
|
32 |
{
|
33 |
"epoch": 1.0,
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate":
|
36 |
-
"loss": 7.
|
37 |
"step": 8
|
38 |
},
|
39 |
{
|
40 |
"epoch": 1.2666666666666666,
|
41 |
-
"grad_norm": 8.
|
42 |
-
"learning_rate":
|
43 |
-
"loss": 9.
|
44 |
"step": 10
|
45 |
},
|
46 |
{
|
47 |
"epoch": 1.5333333333333332,
|
48 |
-
"grad_norm":
|
49 |
-
"learning_rate":
|
50 |
-
"loss": 9.
|
51 |
"step": 12
|
52 |
},
|
53 |
{
|
54 |
"epoch": 1.8,
|
55 |
-
"grad_norm": 9.
|
56 |
-
"learning_rate":
|
57 |
-
"loss": 9.
|
58 |
"step": 14
|
59 |
},
|
60 |
{
|
61 |
"epoch": 2.0,
|
62 |
-
"grad_norm":
|
63 |
-
"learning_rate":
|
64 |
-
"loss": 7.
|
65 |
"step": 16
|
66 |
},
|
67 |
{
|
68 |
"epoch": 2.2666666666666666,
|
69 |
-
"grad_norm":
|
70 |
-
"learning_rate": 7.
|
71 |
-
"loss": 9.
|
72 |
"step": 18
|
73 |
},
|
74 |
{
|
75 |
"epoch": 2.533333333333333,
|
76 |
-
"grad_norm":
|
77 |
-
"learning_rate": 6.
|
78 |
-
"loss": 9.
|
79 |
"step": 20
|
80 |
},
|
81 |
{
|
82 |
"epoch": 2.8,
|
83 |
-
"grad_norm":
|
84 |
-
"learning_rate":
|
85 |
-
"loss": 9.
|
86 |
"step": 22
|
87 |
},
|
88 |
{
|
89 |
"epoch": 3.0,
|
90 |
-
"grad_norm":
|
91 |
-
"learning_rate":
|
92 |
-
"loss": 7.
|
93 |
"step": 24
|
94 |
},
|
95 |
{
|
96 |
"epoch": 3.2666666666666666,
|
97 |
-
"grad_norm":
|
98 |
-
"learning_rate":
|
99 |
-
"loss": 9.
|
100 |
"step": 26
|
101 |
},
|
102 |
{
|
103 |
"epoch": 3.533333333333333,
|
104 |
-
"grad_norm":
|
105 |
-
"learning_rate": 5.357142857142857e-06,
|
106 |
-
"loss": 9.5187,
|
107 |
-
"step": 28
|
108 |
-
},
|
109 |
-
{
|
110 |
-
"epoch": 3.8,
|
111 |
-
"grad_norm": 10.997018814086914,
|
112 |
-
"learning_rate": 5e-06,
|
113 |
-
"loss": 9.5172,
|
114 |
-
"step": 30
|
115 |
-
},
|
116 |
-
{
|
117 |
-
"epoch": 4.0,
|
118 |
-
"grad_norm": 5.385960578918457,
|
119 |
-
"learning_rate": 4.642857142857144e-06,
|
120 |
-
"loss": 7.0662,
|
121 |
-
"step": 32
|
122 |
-
},
|
123 |
-
{
|
124 |
-
"epoch": 4.266666666666667,
|
125 |
-
"grad_norm": 10.675312042236328,
|
126 |
-
"learning_rate": 4.2857142857142855e-06,
|
127 |
-
"loss": 9.4752,
|
128 |
-
"step": 34
|
129 |
-
},
|
130 |
-
{
|
131 |
-
"epoch": 4.533333333333333,
|
132 |
-
"grad_norm": 10.457250595092773,
|
133 |
-
"learning_rate": 3.928571428571429e-06,
|
134 |
-
"loss": 9.4481,
|
135 |
-
"step": 36
|
136 |
-
},
|
137 |
-
{
|
138 |
-
"epoch": 4.8,
|
139 |
-
"grad_norm": 10.809959411621094,
|
140 |
-
"learning_rate": 3.5714285714285718e-06,
|
141 |
-
"loss": 9.4466,
|
142 |
-
"step": 38
|
143 |
-
},
|
144 |
-
{
|
145 |
-
"epoch": 5.0,
|
146 |
-
"grad_norm": 6.051748752593994,
|
147 |
-
"learning_rate": 3.2142857142857147e-06,
|
148 |
-
"loss": 7.0524,
|
149 |
-
"step": 40
|
150 |
-
},
|
151 |
-
{
|
152 |
-
"epoch": 5.266666666666667,
|
153 |
-
"grad_norm": 13.849763870239258,
|
154 |
-
"learning_rate": 2.8571428571428573e-06,
|
155 |
-
"loss": 9.4191,
|
156 |
-
"step": 42
|
157 |
-
},
|
158 |
-
{
|
159 |
-
"epoch": 5.533333333333333,
|
160 |
-
"grad_norm": 11.320974349975586,
|
161 |
-
"learning_rate": 2.5e-06,
|
162 |
-
"loss": 9.4206,
|
163 |
-
"step": 44
|
164 |
-
},
|
165 |
-
{
|
166 |
-
"epoch": 5.8,
|
167 |
-
"grad_norm": 13.425915718078613,
|
168 |
"learning_rate": 2.1428571428571427e-06,
|
169 |
-
"loss": 9.
|
170 |
-
"step":
|
171 |
-
},
|
172 |
-
{
|
173 |
-
"epoch": 6.0,
|
174 |
-
"grad_norm": 6.168519020080566,
|
175 |
-
"learning_rate": 1.7857142857142859e-06,
|
176 |
-
"loss": 7.044,
|
177 |
-
"step": 48
|
178 |
-
},
|
179 |
-
{
|
180 |
-
"epoch": 6.266666666666667,
|
181 |
-
"grad_norm": 13.479830741882324,
|
182 |
-
"learning_rate": 1.4285714285714286e-06,
|
183 |
-
"loss": 9.3804,
|
184 |
-
"step": 50
|
185 |
-
},
|
186 |
-
{
|
187 |
-
"epoch": 6.533333333333333,
|
188 |
-
"grad_norm": 11.913396835327148,
|
189 |
-
"learning_rate": 1.0714285714285714e-06,
|
190 |
-
"loss": 9.3622,
|
191 |
-
"step": 52
|
192 |
-
},
|
193 |
-
{
|
194 |
-
"epoch": 6.8,
|
195 |
-
"grad_norm": 12.051567077636719,
|
196 |
-
"learning_rate": 7.142857142857143e-07,
|
197 |
-
"loss": 9.37,
|
198 |
-
"step": 54
|
199 |
-
},
|
200 |
-
{
|
201 |
-
"epoch": 7.0,
|
202 |
-
"grad_norm": 6.3397536277771,
|
203 |
-
"learning_rate": 3.5714285714285716e-07,
|
204 |
-
"loss": 7.0465,
|
205 |
-
"step": 56
|
206 |
},
|
207 |
{
|
208 |
-
"epoch":
|
209 |
-
"step":
|
210 |
-
"total_flos":
|
211 |
-
"train_loss":
|
212 |
-
"train_runtime":
|
213 |
-
"train_samples_per_second": 0.
|
214 |
"train_steps_per_second": 0.037
|
215 |
}
|
216 |
],
|
217 |
"logging_steps": 2,
|
218 |
-
"max_steps":
|
219 |
"num_input_tokens_seen": 0,
|
220 |
-
"num_train_epochs":
|
221 |
"save_steps": 500,
|
222 |
"stateful_callbacks": {
|
223 |
"TrainerControl": {
|
@@ -231,7 +133,7 @@
|
|
231 |
"attributes": {}
|
232 |
}
|
233 |
},
|
234 |
-
"total_flos":
|
235 |
"train_batch_size": 2,
|
236 |
"trial_name": null,
|
237 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.533333333333333,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 28,
|
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": 7.757550239562988,
|
14 |
+
"learning_rate": 1.8571428571428575e-05,
|
15 |
+
"loss": 9.6987,
|
16 |
"step": 2
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.5333333333333333,
|
20 |
+
"grad_norm": 8.458471298217773,
|
21 |
+
"learning_rate": 1.7142857142857142e-05,
|
22 |
+
"loss": 9.6678,
|
23 |
"step": 4
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.8,
|
27 |
+
"grad_norm": 8.552592277526855,
|
28 |
+
"learning_rate": 1.5714285714285715e-05,
|
29 |
+
"loss": 9.6529,
|
30 |
"step": 6
|
31 |
},
|
32 |
{
|
33 |
"epoch": 1.0,
|
34 |
+
"grad_norm": NaN,
|
35 |
+
"learning_rate": 1.5000000000000002e-05,
|
36 |
+
"loss": 7.1937,
|
37 |
"step": 8
|
38 |
},
|
39 |
{
|
40 |
"epoch": 1.2666666666666666,
|
41 |
+
"grad_norm": 8.743433952331543,
|
42 |
+
"learning_rate": 1.3571428571428574e-05,
|
43 |
+
"loss": 9.6023,
|
44 |
"step": 10
|
45 |
},
|
46 |
{
|
47 |
"epoch": 1.5333333333333332,
|
48 |
+
"grad_norm": 9.371781349182129,
|
49 |
+
"learning_rate": 1.2142857142857142e-05,
|
50 |
+
"loss": 9.5618,
|
51 |
"step": 12
|
52 |
},
|
53 |
{
|
54 |
"epoch": 1.8,
|
55 |
+
"grad_norm": 9.993501663208008,
|
56 |
+
"learning_rate": 1.0714285714285714e-05,
|
57 |
+
"loss": 9.4908,
|
58 |
"step": 14
|
59 |
},
|
60 |
{
|
61 |
"epoch": 2.0,
|
62 |
+
"grad_norm": 5.5162153244018555,
|
63 |
+
"learning_rate": 9.285714285714288e-06,
|
64 |
+
"loss": 7.1442,
|
65 |
"step": 16
|
66 |
},
|
67 |
{
|
68 |
"epoch": 2.2666666666666666,
|
69 |
+
"grad_norm": 9.81357479095459,
|
70 |
+
"learning_rate": 7.857142857142858e-06,
|
71 |
+
"loss": 9.4899,
|
72 |
"step": 18
|
73 |
},
|
74 |
{
|
75 |
"epoch": 2.533333333333333,
|
76 |
+
"grad_norm": 10.446647644042969,
|
77 |
+
"learning_rate": 6.4285714285714295e-06,
|
78 |
+
"loss": 9.4298,
|
79 |
"step": 20
|
80 |
},
|
81 |
{
|
82 |
"epoch": 2.8,
|
83 |
+
"grad_norm": 10.67019271850586,
|
84 |
+
"learning_rate": 5e-06,
|
85 |
+
"loss": 9.4395,
|
86 |
"step": 22
|
87 |
},
|
88 |
{
|
89 |
"epoch": 3.0,
|
90 |
+
"grad_norm": 6.129962921142578,
|
91 |
+
"learning_rate": 3.5714285714285718e-06,
|
92 |
+
"loss": 7.0353,
|
93 |
"step": 24
|
94 |
},
|
95 |
{
|
96 |
"epoch": 3.2666666666666666,
|
97 |
+
"grad_norm": NaN,
|
98 |
+
"learning_rate": 3.5714285714285718e-06,
|
99 |
+
"loss": 9.4083,
|
100 |
"step": 26
|
101 |
},
|
102 |
{
|
103 |
"epoch": 3.533333333333333,
|
104 |
+
"grad_norm": 11.658792495727539,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
105 |
"learning_rate": 2.1428571428571427e-06,
|
106 |
+
"loss": 9.4101,
|
107 |
+
"step": 28
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
108 |
},
|
109 |
{
|
110 |
+
"epoch": 3.533333333333333,
|
111 |
+
"step": 28,
|
112 |
+
"total_flos": 108812813622552.0,
|
113 |
+
"train_loss": 9.016076462609428,
|
114 |
+
"train_runtime": 765.9489,
|
115 |
+
"train_samples_per_second": 0.627,
|
116 |
"train_steps_per_second": 0.037
|
117 |
}
|
118 |
],
|
119 |
"logging_steps": 2,
|
120 |
+
"max_steps": 28,
|
121 |
"num_input_tokens_seen": 0,
|
122 |
+
"num_train_epochs": 4,
|
123 |
"save_steps": 500,
|
124 |
"stateful_callbacks": {
|
125 |
"TrainerControl": {
|
|
|
133 |
"attributes": {}
|
134 |
}
|
135 |
},
|
136 |
+
"total_flos": 108812813622552.0,
|
137 |
"train_batch_size": 2,
|
138 |
"trial_name": null,
|
139 |
"trial_params": null
|