End of training
Browse files- README.md +3 -1
- trainer_state.json +210 -0
README.md
CHANGED
@@ -3,6 +3,8 @@ base_model: facebook/musicgen-melody
|
|
3 |
library_name: peft
|
4 |
license: cc-by-nc-4.0
|
5 |
tags:
|
|
|
|
|
6 |
- generated_from_trainer
|
7 |
model-index:
|
8 |
- name: musicgen-melody-lora-acid_techno-colab
|
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
# musicgen-melody-lora-acid_techno-colab
|
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 |
library_name: peft
|
4 |
license: cc-by-nc-4.0
|
5 |
tags:
|
6 |
+
- text-to-audio
|
7 |
+
- acid_data
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: musicgen-melody-lora-acid_techno-colab
|
|
|
16 |
|
17 |
# musicgen-melody-lora-acid_techno-colab
|
18 |
|
19 |
+
This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the ylacombe/acid_data dataset.
|
20 |
|
21 |
## Model description
|
22 |
|
trainer_state.json
ADDED
@@ -0,0 +1,210 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 12.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 48,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.64,
|
13 |
+
"grad_norm": 0.9444175362586975,
|
14 |
+
"learning_rate": 0.00019166666666666667,
|
15 |
+
"loss": 9.6275,
|
16 |
+
"step": 2
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 1.0,
|
20 |
+
"grad_norm": 0.27594709396362305,
|
21 |
+
"learning_rate": 0.00018333333333333334,
|
22 |
+
"loss": 5.2513,
|
23 |
+
"step": 4
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 1.6400000000000001,
|
27 |
+
"grad_norm": 2.447077989578247,
|
28 |
+
"learning_rate": 0.000175,
|
29 |
+
"loss": 8.6347,
|
30 |
+
"step": 6
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 2.0,
|
34 |
+
"grad_norm": 0.26538360118865967,
|
35 |
+
"learning_rate": 0.0001666666666666667,
|
36 |
+
"loss": 4.5485,
|
37 |
+
"step": 8
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 2.64,
|
41 |
+
"grad_norm": 1.1157130002975464,
|
42 |
+
"learning_rate": 0.00015833333333333332,
|
43 |
+
"loss": 7.6164,
|
44 |
+
"step": 10
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 3.0,
|
48 |
+
"grad_norm": 0.20757710933685303,
|
49 |
+
"learning_rate": 0.00015000000000000001,
|
50 |
+
"loss": 4.2397,
|
51 |
+
"step": 12
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 3.64,
|
55 |
+
"grad_norm": 0.958597719669342,
|
56 |
+
"learning_rate": 0.00014166666666666668,
|
57 |
+
"loss": 7.3819,
|
58 |
+
"step": 14
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 4.0,
|
62 |
+
"grad_norm": 0.285988450050354,
|
63 |
+
"learning_rate": 0.00013333333333333334,
|
64 |
+
"loss": 4.158,
|
65 |
+
"step": 16
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 4.64,
|
69 |
+
"grad_norm": 0.6224765181541443,
|
70 |
+
"learning_rate": 0.000125,
|
71 |
+
"loss": 7.2957,
|
72 |
+
"step": 18
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 5.0,
|
76 |
+
"grad_norm": 0.18170130252838135,
|
77 |
+
"learning_rate": 0.00011666666666666668,
|
78 |
+
"loss": 4.0762,
|
79 |
+
"step": 20
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 5.64,
|
83 |
+
"grad_norm": 0.4508182108402252,
|
84 |
+
"learning_rate": 0.00010833333333333333,
|
85 |
+
"loss": 7.219,
|
86 |
+
"step": 22
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 6.0,
|
90 |
+
"grad_norm": 0.17579227685928345,
|
91 |
+
"learning_rate": 0.0001,
|
92 |
+
"loss": 4.0777,
|
93 |
+
"step": 24
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 6.64,
|
97 |
+
"grad_norm": 0.5887883305549622,
|
98 |
+
"learning_rate": 9.166666666666667e-05,
|
99 |
+
"loss": 7.2008,
|
100 |
+
"step": 26
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 7.0,
|
104 |
+
"grad_norm": 0.16587680578231812,
|
105 |
+
"learning_rate": 8.333333333333334e-05,
|
106 |
+
"loss": 4.0601,
|
107 |
+
"step": 28
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 7.64,
|
111 |
+
"grad_norm": 0.8699614405632019,
|
112 |
+
"learning_rate": 7.500000000000001e-05,
|
113 |
+
"loss": 7.1846,
|
114 |
+
"step": 30
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 8.0,
|
118 |
+
"grad_norm": 0.12891338765621185,
|
119 |
+
"learning_rate": 6.666666666666667e-05,
|
120 |
+
"loss": 4.0341,
|
121 |
+
"step": 32
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 8.64,
|
125 |
+
"grad_norm": 0.6497959494590759,
|
126 |
+
"learning_rate": 5.833333333333334e-05,
|
127 |
+
"loss": 7.1684,
|
128 |
+
"step": 34
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 9.0,
|
132 |
+
"grad_norm": 0.21780814230442047,
|
133 |
+
"learning_rate": 5e-05,
|
134 |
+
"loss": 4.0268,
|
135 |
+
"step": 36
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 9.64,
|
139 |
+
"grad_norm": 0.5005463361740112,
|
140 |
+
"learning_rate": 4.166666666666667e-05,
|
141 |
+
"loss": 7.1632,
|
142 |
+
"step": 38
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 10.0,
|
146 |
+
"grad_norm": 0.1379019170999527,
|
147 |
+
"learning_rate": 3.3333333333333335e-05,
|
148 |
+
"loss": 4.0141,
|
149 |
+
"step": 40
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 10.64,
|
153 |
+
"grad_norm": 0.3742949962615967,
|
154 |
+
"learning_rate": 2.5e-05,
|
155 |
+
"loss": 7.1528,
|
156 |
+
"step": 42
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 11.0,
|
160 |
+
"grad_norm": 0.1629418581724167,
|
161 |
+
"learning_rate": 1.6666666666666667e-05,
|
162 |
+
"loss": 4.0143,
|
163 |
+
"step": 44
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 11.64,
|
167 |
+
"grad_norm": 0.3627496659755707,
|
168 |
+
"learning_rate": 8.333333333333334e-06,
|
169 |
+
"loss": 7.147,
|
170 |
+
"step": 46
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 12.0,
|
174 |
+
"grad_norm": 0.13871623575687408,
|
175 |
+
"learning_rate": 0.0,
|
176 |
+
"loss": 4.012,
|
177 |
+
"step": 48
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 12.0,
|
181 |
+
"step": 48,
|
182 |
+
"total_flos": 147726005249160.0,
|
183 |
+
"train_loss": 5.887702425320943,
|
184 |
+
"train_runtime": 183.8612,
|
185 |
+
"train_samples_per_second": 4.351,
|
186 |
+
"train_steps_per_second": 0.261
|
187 |
+
}
|
188 |
+
],
|
189 |
+
"logging_steps": 2,
|
190 |
+
"max_steps": 48,
|
191 |
+
"num_input_tokens_seen": 0,
|
192 |
+
"num_train_epochs": 16,
|
193 |
+
"save_steps": 500,
|
194 |
+
"stateful_callbacks": {
|
195 |
+
"TrainerControl": {
|
196 |
+
"args": {
|
197 |
+
"should_epoch_stop": false,
|
198 |
+
"should_evaluate": false,
|
199 |
+
"should_log": false,
|
200 |
+
"should_save": true,
|
201 |
+
"should_training_stop": true
|
202 |
+
},
|
203 |
+
"attributes": {}
|
204 |
+
}
|
205 |
+
},
|
206 |
+
"total_flos": 147726005249160.0,
|
207 |
+
"train_batch_size": 2,
|
208 |
+
"trial_name": null,
|
209 |
+
"trial_params": null
|
210 |
+
}
|