InternLM/InternBootcamp
1
0
Fork 0
mirror of https://github.com/InternLM/InternBootcamp.git synced 2026-04-27 17:23:17 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

refactor(data_pipeline): optimize data generation pipeline; add multiple preset configurations for data generation

Browse source
This commit is contained in:
chenyongkang 2025-08-26 18:40:21 +08:00
parent 1a8477c8d8
commit 8d493b35a0
2160 changed files with 69199 additions and 154 deletions
Download patch file Download diff file Expand all files Collapse all files

41
examples/pipelines/puzzle_configs/korOperationUnicode2605_train.json
View file

@ -1,44 +1,17 @@
[
{
"compute_prob": 0.9,
"equation_type2_prob": 0.0,
"min_val": -5,
"max_val": 5
"compute_prob": 0.5,
"min_val": -5000,
"max_val": 5000
},
{
"compute_prob": 0.5,
"equation_type2_prob": 0.2,
"min_val": -5,
"max_val": 5
},
{
"compute_prob": 0.7,
"equation_type2_prob": 0.3,
"min_val": -10,
"max_val": 10
},
{
"compute_prob": 0.8,
"equation_type2_prob": 0.1,
"min_val": -5,
"max_val": 5
},
{
"compute_prob": 0.1,
"equation_type2_prob": 0.9,
"min_val": -3,
"max_val": 3
"min_val": -500,
"max_val": 500
},
{
"compute_prob": 0.5,
"equation_type2_prob": 0.5,
"min_val": -2,
"max_val": 2
},
{
"compute_prob": 0.4,
"equation_type2_prob": 0.1,
"min_val": 0,
"max_val": 5
"min_val": -50,
"max_val": 50
}
]