colin
/
Witllm
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Init status.

This commit is contained in:
Colin 2025-08-10 15:25:14 +08:00
parent b56bbb2952
commit a70e19df5d
1 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
wit/train.py
View File

@ -7,7 +7,7 @@ from model.modeling_rwkv7 import RWKVLMHeadModel
from logger import TBLogger from logger import TBLogger
import configuration import configuration
import dataset.dataset as ds import meaning as m
import numpy as np import numpy as np
if __name__ == "__main__": if __name__ == "__main__":
@ -29,10 +29,13 @@ if __name__ == "__main__":
conf.seed = 42 conf.seed = 42
conf.dataloader_works = 2 conf.dataloader_works = 2
conf.dataset.meaning.start = 800 conf.dataset.meaning.start = 10000
conf.dataset.meaning.size = 200000 conf.dataset.meaning.end = 200000
conf.dataset.meaning.size = None
conf.dataset.meaning.min_subitem = 2 conf.dataset.meaning.min_subitem = 2
conf.dataset.meaning.max_subitem = 4 conf.dataset.meaning.max_subitem = 6
conf.dataset.meaning.stride = 1
conf.dataset.meaning.with_tree = False
conf.dataset.meaning.val_mask_level = [0, 1, 2] conf.dataset.meaning.val_mask_level = [0, 1, 2]
conf.dataset.meaning.val_mask_idx = [0, 0, -1] conf.dataset.meaning.val_mask_idx = [0, 0, -1]
@ -50,7 +53,7 @@ if __name__ == "__main__":
# model = RWKVLMHeadModel(conf.model_config) # model = RWKVLMHeadModel(conf.model_config)
qwen = LightModule(conf, model) qwen = LightModule(conf, model)
train_dataloader, val_dataloader = ds.InitDataset(conf) train_dataloader, val_dataloader = m.InitDataset(conf)
# for i in range(len(train_dataloader)): # for i in range(len(train_dataloader)):
# print(train_dataloader.print_mapping(i)) # print(train_dataloader.print_mapping(i))