junhaoxjtu commited on
Commit
97edb2a
1 Parent(s): d79797c

End of training

Browse files
Files changed (5) hide show
  1. README.md +6 -1
  2. all_results.json +15 -0
  3. eval_results.json +9 -0
  4. train_results.json +9 -0
  5. trainer_state.json +98 -0
README.md CHANGED
@@ -2,6 +2,8 @@
2
  license: cc-by-nc-4.0
3
  library_name: peft
4
  tags:
 
 
5
  - generated_from_trainer
6
  base_model: facebook/musicgen-melody
7
  model-index:
@@ -16,7 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
16
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/xianjiaotonguni/huggingface/runs/fb9arrpz)
17
  # musicgen-melody-lora-punk
18
 
19
- This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on an unknown dataset.
 
 
 
20
 
21
  ## Model description
22
 
 
2
  license: cc-by-nc-4.0
3
  library_name: peft
4
  tags:
5
+ - text-to-audio
6
+ - ylacombe/tiny-punk
7
  - generated_from_trainer
8
  base_model: facebook/musicgen-melody
9
  model-index:
 
18
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/xianjiaotonguni/huggingface/runs/fb9arrpz)
19
  # musicgen-melody-lora-punk
20
 
21
+ This model is a fine-tuned version of [facebook/musicgen-melody](https://huggingface.co/facebook/musicgen-melody) on the YLACOMBE/TINY-PUNK - DEFAULT dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 4.7289
24
+ - Clap: 0.2488
25
 
26
  ## Model description
27
 
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.764705882352941,
3
+ "eval_clap": 0.2487790882587433,
4
+ "eval_loss": 4.728930473327637,
5
+ "eval_runtime": 85.623,
6
+ "eval_samples": 5,
7
+ "eval_samples_per_second": 0.058,
8
+ "eval_steps_per_second": 0.058,
9
+ "total_flos": 35328595626816.0,
10
+ "train_loss": 8.988131165504456,
11
+ "train_runtime": 179.4885,
12
+ "train_samples": 33,
13
+ "train_samples_per_second": 0.735,
14
+ "train_steps_per_second": 0.045
15
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.764705882352941,
3
+ "eval_clap": 0.2487790882587433,
4
+ "eval_loss": 4.728930473327637,
5
+ "eval_runtime": 85.623,
6
+ "eval_samples": 5,
7
+ "eval_samples_per_second": 0.058,
8
+ "eval_steps_per_second": 0.058
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.764705882352941,
3
+ "total_flos": 35328595626816.0,
4
+ "train_loss": 8.988131165504456,
5
+ "train_runtime": 179.4885,
6
+ "train_samples": 33,
7
+ "train_samples_per_second": 0.735,
8
+ "train_steps_per_second": 0.045
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.764705882352941,
5
+ "eval_steps": 25,
6
+ "global_step": 8,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.47058823529411764,
13
+ "grad_norm": 1.3445526361465454,
14
+ "learning_rate": 0.000175,
15
+ "loss": 9.5733,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.9411764705882353,
20
+ "grad_norm": 1.2209374904632568,
21
+ "learning_rate": 0.00015000000000000001,
22
+ "loss": 9.4639,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 1.4117647058823528,
27
+ "grad_norm": 1.165825605392456,
28
+ "learning_rate": 0.000125,
29
+ "loss": 9.2536,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 1.8823529411764706,
34
+ "grad_norm": 1.3359602689743042,
35
+ "learning_rate": 0.0001,
36
+ "loss": 9.0183,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 2.3529411764705883,
41
+ "grad_norm": 1.567587971687317,
42
+ "learning_rate": 7.500000000000001e-05,
43
+ "loss": 8.8427,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 2.8235294117647056,
48
+ "grad_norm": 1.6827342510223389,
49
+ "learning_rate": 5e-05,
50
+ "loss": 8.7436,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 3.2941176470588234,
55
+ "grad_norm": 2.0196425914764404,
56
+ "learning_rate": 2.5e-05,
57
+ "loss": 8.527,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 3.764705882352941,
62
+ "grad_norm": 2.155809164047241,
63
+ "learning_rate": 0.0,
64
+ "loss": 8.4827,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 3.764705882352941,
69
+ "step": 8,
70
+ "total_flos": 35328595626816.0,
71
+ "train_loss": 8.988131165504456,
72
+ "train_runtime": 179.4885,
73
+ "train_samples_per_second": 0.735,
74
+ "train_steps_per_second": 0.045
75
+ }
76
+ ],
77
+ "logging_steps": 1,
78
+ "max_steps": 8,
79
+ "num_input_tokens_seen": 0,
80
+ "num_train_epochs": 4,
81
+ "save_steps": 500,
82
+ "stateful_callbacks": {
83
+ "TrainerControl": {
84
+ "args": {
85
+ "should_epoch_stop": false,
86
+ "should_evaluate": false,
87
+ "should_log": false,
88
+ "should_save": false,
89
+ "should_training_stop": false
90
+ },
91
+ "attributes": {}
92
+ }
93
+ },
94
+ "total_flos": 35328595626816.0,
95
+ "train_batch_size": 2,
96
+ "trial_name": null,
97
+ "trial_params": null
98
+ }