Downloads · 30 days
0
XiaoShengpeng/AegisEER
AegisEER is a machine learning model from XiaoShengpeng. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
赛题定制版 EEG 跨被试情绪识别冲榜工程。v5 在 v4 的 MAE + Contrastive + 多教师 + DE/PSD classical ensemble 基础上,新增:
Downloads · 30 days
0
Access
Public
Updated Jul 24, 2026
Repo size
144 GB
Likes
0
Public
Click a slice to open those files.
.pt119 GB · 92%
From the Hugging Face model README
赛题定制版 EEG 跨被试情绪识别冲榜工程。v5 在 v4 的 MAE + Contrastive + 多教师 + DE/PSD classical ensemble 基础上,新增:
data_raw/
train/
DEP/DEP1003timedata.mat
HC/HC1003timedata.mat
test/
P_test1.mat
...
template.xlsx
bash scripts/build.sh aegis-eer-sota:v5
bash scripts/run_docker.sh aegis-eer-sota:v5
bash scripts/full_sota_pipeline.sh configs/sota_4x4090.yaml data_raw/train data_raw/test data_raw/template.xlsx
最终文件:
outputs/submission_ensemble_v5.xlsx
outputs/prob_ensemble_v5.csv
outputs/diagnosis_v5.csv
bash scripts/resume_from_pretrain_stage_v5.sh configs/sota_4x4090.yaml data_raw/template.xlsx
checkpoints/mae_base.pt
checkpoints/pars_base.pt
checkpoints/pretrain_base.pt
checkpoints/teacher_a_fold0_seed3407.pt
checkpoints/teacher_b_fold1_seed2025.pt
checkpoints/teacher_c_fold2_seed777.pt
checkpoints/teacher_d_fold3_seed2605.pt
checkpoints/student_distilled_v5.pt
checkpoints/final_all_subjects_v5.pt
outputs/submission_ensemble_v5.xlsx
DENSE_STRIDE=625。如需更快,可临时使用 DENSE_STRIDE=1250。trial_top4_bacc / trial_auc 比普通 window-level ACC 更接近线上 top-4 提交流程。best.pt 为准。