File size: 2,039 Bytes
7458034
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
model:
  _component_: models.lora_mmllama3_8b
  lora_attn_modules:
  - q_proj
  - v_proj
  apply_lora_to_mlp: false
  apply_lora_to_output: false
  lora_rank: 8
  lora_alpha: 16
  perception_tokens: 2
tokenizer:
  _component_: models.a2a_tokenizer
  path: checkpoints/Meta-Llama-3-8B-Instruct/tokenizer.model
checkpointer:
  _component_: torchtune.utils.FullModelMetaCheckpointer
  checkpoint_dir: checkpoints/Meta-Llama-3-8B-Instruct/
  checkpoint_files:
  - consolidated.00.pth
  adapter_checkpoint: null
  recipe_checkpoint: null
  output_dir: output
  model_type: LLAMA3
resume_from_checkpoint: false
dataset:
  _component_: ds.EvenBatcher
  dataset:
    _component_: ds.RoundRobinDataset
    datasets:
    - _component_: ds.LlavaInstructDataset
      ib_embed_path: ds/coco_llava_instruct/*.ib_embed.pt
      clip_embed_path: ds/coco_llava_instruct/*.clip_embed.pt
      caption_path: ds/coco_llava_instruct/*.caption.pt
      train_on_input: false
    - _component_: ds.LlavaInstructDataset
      ib_embed_path: ds/vision_flan/*.ib_embed.pt
      clip_embed_path: ds/vision_flan/*.clip_embed.pt
      caption_path: ds/vision_flan/*.caption.pt
      train_on_input: false
    - _component_: ds.CaptionInstructDataset
      ib_embed_path: ds/sam_llava/00.ib_embed.pt
      clip_embed_path: ds/sam_llava/00.clip_embed.pt
      caption_path: ds/sam_llava/00.caption.pt
      train_on_input: false
seed: null
shuffle: true
batch_size: 4
optimizer:
  _component_: torch.optim.AdamW
  weight_decay: 0.01
  lr: 0.0003
lr_scheduler:
  _component_: torchtune.modules.get_cosine_schedule_with_warmup
  num_warmup_steps: 100
loss:
  _component_: torch.nn.CrossEntropyLoss
epochs: 1
max_steps_per_epoch: null
gradient_accumulation_steps: 64
compile: false
output_dir: /tmp/lora_finetune_output
metric_logger:
  _component_: torchtune.utils.metric_logging.DiskLogger
  log_dir: ${output_dir}
log_every_n_steps: null
device: cuda
dtype: bf16
enable_activation_checkpointing: false
profiler:
  _component_: torchtune.utils.profiler
  enabled: false