submit first 200 no-record PPU models for zhoushasha

README.md

@@ -2,12 +2,12 @@
 
 批量向 ModelHub XC 平台提交模型验证任务的策略服务，之后保持 HTTP 服务存活供平台探活。
 
-当前批次仅向 `zhoushasha` 账号提交 `main.py` 中列出的 12 个模型，每个模型只提交 `ppu_zw_810e` 一张卡，共 12 个任务。
+当前批次从 `no_record_ppu_zw_810e_2026-10-10.txt` 按原顺序取前 200 个模型，向 `zhoushasha` 账号提交；每个模型只提交 `ppu_zw_810e` 一张卡。原始候选清单保持不变。
 
 ## 功能
 
 - 使用 `AUTH_TOKEN` 环境变量（未提供时使用 `main.py` 中的预设 Bearer Token）向 `zhoushasha` 账号提交任务
-- 从 `main.py` 的 `PPU_MODELS` 读取本批次模型，批量提交验证任务
+- 从 `model_ids/ppu_zw_810e_2026-10-10_batch200.txt` 读取本批次模型，批量提交验证任务
 - 账号额度暂满时，每 30 分钟重试尚未提交的模型
 - 提交结果写入 `submitted_validation_tasks.txt`
 - 暴露 `/health` 和 `/status` 接口满足平台运行时契约
@@ -17,7 +17,7 @@
 ```
 .
 ├── main.py            # 主入口：HTTP 服务 + 提交逻辑
-├── model_ids/         # 历史批次模型 ID 清单，本轮未使用
+├── model_ids/         # 本轮 200 个模型及历史批次清单
 ├── Dockerfile         # 平台镜像构建配置
 ├── requirements.txt   # Python 依赖
 └── submitted_validation_tasks.txt   # 运行后自动生成，记录提交结果

main.py

@@ -2,7 +2,7 @@
 xc_validation_strategy — 主入口
 
 启动后针对 GPU_JOBS 中配置的 GPU 卡分别批量提交各自筛选出的模型验证任务。
-本轮仅向 zhoushasha 提交指定的 12 个模型在 ppu_zw_810e 上的验证任务；
+本轮仅向 zhoushasha 提交候选清单前 200 个模型在 ppu_zw_810e 上的验证任务；
 其他 GPU 的 config_content 模板仍保留，但未列入 GPU_JOBS。
 （/adminApi/async/task/create-contest-task，
 Bearer Token 认证），之后保持 HTTP 服务存活。
@@ -76,22 +76,9 @@ HYGON_MODELS = load_model_ids("hygon_k100-ai_2026-09-24.txt")
 KUNLUNXIN_MODELS = [
 ]
 
-PPU_MODELS = [
-    "raalr/Qwen2.5-1.5-uld-gemma-27b-3",
-    "gemasphi/real-setfit-ss-distiluse-base-multilingual-cased-v2",
-    "intfloat/e5-small",
-    "Nhat1904/32-shot-twitter-2classes-head-body",
-    "Nhat1904/16-shot-twitter-2classes-head-body",
-    "Nhat1904/16-shot-twitter-2classes-new-1",
-    "Nhat1904/16-shot-twitter-2classes-new",
-    "Nhat1904/16-shot-twitter-2classes",
-    "Nhat1904/32-shot-twitter-10epochs",
-    "Nhat1904/32-shot-twitter",
-    "Nhat1904/16-shot-twitter",
-    "kulibinai/cadreasoner",
-]
+PPU_MODELS = load_model_ids("ppu_zw_810e_2026-10-10_batch200.txt")
 
-# 本轮仅提交用户指定的 12 个模型，每个模型只提交 ppu_zw_810e。
+# 本轮仅提交候选清单前 200 个模型，每个模型只提交 ppu_zw_810e。
 GPU_JOBS: List[Tuple[str, List[str]]] = [
     ("ppu_zw_810e",          PPU_MODELS),
 ]

model_ids/ppu_zw_810e_2026-10-10_batch200.txt

@@ -0,0 +1,200 @@
+mitra-mir/setfit-model-Miscellaneous-Misinformation
+mitra-mir/setfit-model-Misinformation-on-Organizations-GoFundMe-WEF
+mitra-mir/setfit-model-Misinformation-on-Media-Traditional-Social
+mitra-mir/setfit-model-Misinformation-on-Govt
+mitra-mir/setfit-model-Misinformation-on-Mandates-Public-Health
+fawazatvetted/fine_tuned_model_mpnet_qa
+fpadovani/isl-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed455
+Hoax0930/pseudo_paraphrase-multilingual-MiniLM-L12-v2
+nkkbr/Mini-K3-1H-mamba2-n64-g8-v1_C
+Hoax0930/sbert
+Mayhem50/sgpt-bloom-560m-nli-v2
+menbom/test-setfit-model
+RinTrin/climate-tech-0.8316
+ietz/token-paraphrase-MiniLM-L6-v2-baseline
+Mayhem50/sgpt-bloom-560M-nli
+afschowdhury/semantic-xlmr-bn
+testCelUR/all-mpnet-base-v2-celanese
+RinTrin/climate-tech-0.8605
+RinTrin/climate-tech-test
+simplexico/legalevalrr
+hxssgaa/dstc_setfit_deberta
+abhijitt/bert_st_qa_allminiLML6V2
+hxssgaa/dstc_setfit
+devon7y/Clod
+intfloat/e5-large-unsupervised
+shrikritisingh/my-setfit-model
+shrikritisingh/my-setfit
+scaperex/SetFit-all-data
+scaperex/SetFitTest
+Ramos-Ramos/xlm-roberta-base-en-tl
+HasinMDG/SetFit_Labse_Sentiment_Towards_Topic
+umass/mpnet-base-mimics-query-facet-encoder
+Omega-Factory/OLMoE-0125-Instruct-Compressed-Experimental
+jamiehudson/600-STmodel-brand-rem
+hmehta92/finetuned-model
+maryumraza/setfit-model-category-v3.4
+maryumraza/setfit-model-category-v3.3
+abhijitt/bert_ft_sentence_abhijitt
+johnt/bert_ft_sentence
+MergekitCloud/mergekit-64
+MergekitCloud/mergekit-65
+maryumraza/setfit-model-category-v3.0
+trtd56/suicide-tweet-setfit-model
+MergekitCloud/mergekit-66
+WatsonNT/Ling-3.0-tiny
+MergekitCloud/mergekit-67
+lewispons/Email-classifier-v2
+ramdane/jurimodel
+AISE-TUDelft/pharo-responsibilities-classifier
+AISE-TUDelft/pharo-example-classifier
+AISE-TUDelft/java-ownership-classifier
+AISE-TUDelft/pharo-classreferences-classifier
+AISE-TUDelft/pharo-collaborators-classifier
+AISE-TUDelft/java-rational-classifier
+AISE-TUDelft/pharo-keymessages-classifier
+AISE-TUDelft/pharo-keyimplementationpoints-classifier
+AISE-TUDelft/java-expand-classifier
+singtan/solvrays-llm-pdf
+AISE-TUDelft/java-usage-classifier
+AISE-TUDelft/java-pointer-classifier
+AISE-TUDelft/java-summary-classifier
+eekay/gemma-2b-it-noised-np0.25-attn-emb
+Elshan/distilroberta-base-sentence-transformer
+rahmanfadhil/indobert-finetuned-indonli
+Jordancole21/SEC_QA_mnr2_multi_QA_10_E_data_5
+alicekwak/test-sbert
+taldarim/setfit-symptoms-v1
+stjiris/bert-large-portuguese-cased-legal-tsdae-mkd-nli-sts-v1
+stjiris/bert-large-portuguese-cased-legal-tsdae-mkd-nli-sts-v0
+stjiris/bert-large-portuguese-cased-legal-mlm-mkd-nli-sts-v1
+stjiris/bert-large-portuguese-cased-legal-mlm-mkd-nli-sts-v0
+taldarim/setfit-pc-multilabel-example
+kmseong/qwen2_5_32b_instruct_SSFT-lr5e-5
+dinait/test-setfit-model
+danielsaggau/setfit_bregman_mimic_less_training
+danielsaggau/setfit_simcse_mimic_lesstraining
+danielsaggau/setfit_simcse_mimic
+danielsaggau/setfit_modded_mimic
+danielsaggau/bregman_10kflat_moreiterations_ECTHR
+danielsaggau/bregman_base_ecthr
+danielsaggau/modded_ECTHR
+brunokreiner/lyrics-bert
+RIOLITE/sentence_transformers
+neilthematic/setfit-ethos-multilabel-example
+aparna1595/bert-tl-first1000samples
+bongsoo/albert-small-kor-sbert-v1.1
+oliviamga2/multi-qa-mpnet-base-dot-v1_legal_finetune
+textgain/allnli-GroNLP-bert-base-dutch-cased
+huihui-ai/Huihui-Qwen-AgentWorld-35B-A3B-abliterated
+meichen91/review_multiclass_20230116
+deetsml/deetsml
+Jordancole21/SEC_QA_mnr2_roberta_20_E_data_1
+Linco/test-model
+deutsche-telekom/gbert-large-paraphrase-euclidean
+deutsche-telekom/gbert-large-paraphrase-cosine
+bongsoo/kpf-sbert-v1.1
+bongsoo/kpf-sbert-v1
+bongsoo/klue-sbert-v1
+and-effect/musterdatenkatalog_clf
+czesty/ea-setfit-v1-classifier
+bongsoo/albert-small-kor-sbert-v1
+PM-AI/paraphrase-distilroberta-base-v2_de-en
+oshizo/sbert-jsnli-luke-japanese-base-lite
+fpadovani/zho-hans-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407
+fpadovani/zho-hans-100mb-10mb_seed3407
+Jordancole21/SEC_QA_mnr2_roberta_5_E
+ilhkn/mount2_model
+gart-labor/eng-distilBERT-se
+yafsin/test121212
+fpadovani/tur-latn-100mb-ppt-shuff-dyck-100mb_seed10
+Shobhank-iiitdwd/BERT-BU-nli-sts-sent
+gart-labor/paraphrase-MiniLM-L6-v2-eclass
+gart-labor/all-mpnet-base-v2-eclass
+fpadovani/tur-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10
+stjiris/bert-large-portuguese-cased-legal-mlm-gpl-nli-sts-v1
+stjiris/bert-large-portuguese-cased-legal-mlm-gpl-nli-sts-v0
+stjiris/bert-large-portuguese-cased-legal-mlm-nli-sts-v1
+fpadovani/tur-latn-100mb-10mb_seed10
+stjiris/bert-large-portuguese-cased-legal-mlm-nli-sts-v0
+stjiris/bert-large-portuguese-cased-legal-mlm-sts-v1
+stjiris/bert-large-portuguese-cased-legal-mlm-sts-v0
+fpadovani/swa-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407
+nikcheerla/nooks-amd-detection-realtime
+hroth01/psais-paraphrase-multilingual-MiniLM-L12-v2-50shot
+fpadovani/rus-cyrl-10mb-ppt-shuff-dyck-100mb_seed10
+hroth01/psais-LaBSE-1shot
+fpadovani/rus-cyrl-100mb-ppt-shuff-dyck-100mb_seed10
+Kalzaik/setfit-minilm-distilled
+nikcheerla/nooks-amd-detection
+hroth01/psais-LaBSE-50shot
+fpadovani/rus-cyrl-100mb-ppt-Dp-100mb_seed10
+fpadovani/jpn-jpan-10mb-ppt-shuff-dyck-100mb_seed10
+stjiris/bert-large-portuguese-cased-legal-tsdae-gpl-nli-sts-v1
+stjiris/bert-large-portuguese-cased-legal-tsdae-gpl-nli-sts-v0
+stjiris/bert-large-portuguese-cased-legal-tsdae-nli-sts-v1
+fpadovani/jpn-jpan-10mb-ppt-Dp-10mb_seed10
+stjiris/bert-large-portuguese-cased-legal-tsdae-nli-sts-v0
+stjiris/bert-large-portuguese-cased-legal-tsdae-sts-v1
+stjiris/bert-large-portuguese-cased-legal-tsdae-sts-v0
+fpadovani/jpn-jpan-10mb-after-ppt-Dp-100mb-ckpt500_seed10
+fpadovani/jpn-jpan-100mb-ppt-shuff-dyck-100mb_seed10
+stjiris/bert-large-portuguese-cased-legal-mlm
+hroth01/psais-LaBSE-20shot
+hroth01/psais-multi-qa-mpnet-base-dot-v1-50shot
+fpadovani/jpn-jpan-100mb-ppt-Dp-10mb_seed10
+fpadovani/jpn-jpan-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed10
+fpadovani/jpn-jpan-100mb-after-ppt-Dp-100mb-ckpt500_seed10
+hroth01/psais-multi-qa-mpnet-base-dot-v1-20shot
+hroth01/psais-all-mpnet-base-v2-50shot
+hroth01/psais-all-mpnet-base-v2-20shot
+hroth01/psais-all-MiniLM-L6-v2-50shot
+hroth01/psais-all-MiniLM-L6-v2-20shot
+fpadovani/eng-latn-10mb-ppt-shuff-dyck-10mb_seed10
+fpadovani/eng-latn-10mb-ppt-shuff-dyck-100mb_seed10
+fpadovani/eng-latn-10mb-ppt-Dp-100mb_seed10
+hroth/psais-paraphrase-multilingual-MiniLM-L12-v2-20shot
+hroth/psais-LaBSE-10shot
+fpadovani/eng-latn-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed10
+hroth/psais-LaBSE-8shot
+hroth/psais-multi-qa-mpnet-base-dot-v1-10shot
+trl-internal-testing/tiny-FalconMambaForCausalLM
+hroth/psais-multi-qa-mpnet-base-dot-v1-8shot
+fpadovani/eng-latn-100mb-ppt-shuff-dyck-10mb_seed10
+hroth/psais-all-mpnet-base-v2-10shot
+jessietextstan/setfit_v0
+fpadovani/eng-latn-100mb-ppt-Dp-100mb_seed10
+hroth/psais-all-mpnet-base-v2-8shot
+hroth/psais-all-MiniLM-L6-v2-10shot
+hroth/psais-all-MiniLM-L6-v2-8shot
+theyur/dhamma-parrot-v01
+hroth/psais-paraphrase-multilingual-MiniLM-L12-v2-10shot
+hroth/psais-paraphrase-multilingual-MiniLM-L12-v2-8shot
+hroth/psais-multi-qa-mpnet-base-dot-v1-5shot
+flywheel-ai/legal-intake
+hroth/psais-multi-qa-mpnet-base-dot-v1-1shot
+flywheel-ai/fitness
+flywheel-ai/automotive
+hroth/psais-all-mpnet-base-v2-1shot
+hroth/psais-all-MiniLM-L6-v2-5shot
+hroth/psais-all-MiniLM-L6-v2-1shot
+hroth/psais-paraphrase-multilingual-MiniLM-L12-v2-5shot
+hroth/psais-paraphrase-multilingual-MiniLM-L12-v2-1shot
+stjiris/bert-large-portuguese-cased-legal-tsdae
+JoBeer/sentence-t5-base-eclass
+fawazatvetted/fine_tuned_mpnetv4
+lewtun/setfit-new-model-card
+Shobhank-iiitdwd/DistilRoBERTa-QQP
+Kuaaangwen/Setfit-few-shot-classifier
+chiragme/setfit-minilm-distilled-ledgar
+Nischal2015/models_sbert_v6
+mann2107/setfitrrcrececnr
+mann2107/setfitbcmpii
+Shobhank-iiitdwd/BERT-snli-mnli-stsb
+api19750904/tipo_campanya_ong_v3
+fawazatvetted/fine_tuned_mpnetv3
+Chadeh/setfit-minilm-distilled
+MarioMartone/my-awesome-setfit-model2
+cjos/my-awesome-setfit-model
+Chadeh/my-awesome-setfit-model
+MU-Kindai/SBERT-JSNLI-large