hhhar/Linguist_should_be_smart_2
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: hhhar/Linguist_should_be_smart_2
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-30 03:28:27 +08:00
commit a55ff33ebd
14 changed files with 456672 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
.gitattributes vendored Normal file
View File

@@ -0,0 +1,35 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text
*.ftz filter=lfs diff=lfs merge=lfs -text
*.gz filter=lfs diff=lfs merge=lfs -text
*.h5 filter=lfs diff=lfs merge=lfs -text
*.joblib filter=lfs diff=lfs merge=lfs -text
*.lfs.* filter=lfs diff=lfs merge=lfs -text
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz filter=lfs diff=lfs merge=lfs -text
*.onnx filter=lfs diff=lfs merge=lfs -text
*.ot filter=lfs diff=lfs merge=lfs -text
*.parquet filter=lfs diff=lfs merge=lfs -text
*.pb filter=lfs diff=lfs merge=lfs -text
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl filter=lfs diff=lfs merge=lfs -text
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* filter=lfs diff=lfs merge=lfs -text
*.tar filter=lfs diff=lfs merge=lfs -text
*.tflite filter=lfs diff=lfs merge=lfs -text
*.tgz filter=lfs diff=lfs merge=lfs -text
*.wasm filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zst filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text

202
LICENSE Normal file
View File

@@ -0,0 +1,202 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright 2024 Alibaba Cloud
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

1
README.md Normal file
View File

@@ -0,0 +1 @@
should be updated

35
config.json Normal file
View File

@@ -0,0 +1,35 @@
{
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 151643,
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 5120,
"initializer_range": 0.02,
"intermediate_size": 13824,
"max_position_embeddings": 32768,
"max_window_layers": 70,
"model_type": "qwen2",
"num_attention_heads": 40,
"num_hidden_layers": 48,
"num_key_value_heads": 8,
"quantization_config": {
"bits": 4,
"group_size": 128,
"modules_to_not_convert": null,
"quant_method": "awq",
"version": "gemm",
"zero_point": true
},
"rms_norm_eps": 1e-06,
"rope_theta": 1000000.0,
"sliding_window": 131072,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.41.1",
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 152064
}

14
generation_config.json Normal file
View File

@@ -0,0 +1,14 @@
{
"bos_token_id": 151643,
"do_sample": true,
"eos_token_id": [
151645,
151643
],
"pad_token_id": 151643,
"repetition_penalty": 1.05,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8,
"transformers_version": "4.41.1"
}

151387
merges.txt Normal file
View File

File diff suppressed because it is too large Load Diff

3
model-00001-of-00003.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4e874dc3b1febb5b22fe74a8793066ae430d90cdbc51765d0a4eb44a82a1fbbd
size 3988804408

3
model-00002-of-00003.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b3b25da74cc854cdc726956f8152f1dda8519c7bb7d4724ac12c1312463e61c8
size 3968309440

3
model-00003-of-00003.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:25b97bf28033ed4560387293ce76bd3dc55a22882fea005a10c137b6478dae85
size 2023056736

1258
model.safetensors.index.json Normal file
View File

File diff suppressed because it is too large Load Diff

241
script.py Normal file
View File

@@ -0,0 +1,241 @@
#!/usr/bin/env python
import os, re, json, time, unicodedata
from collections import Counter, defaultdict
T0 = time.time()
TIME_LIMIT = float(os.environ.get("IOL_TIME_LIMIT", "1800"))
SAFETY = float(os.environ.get("IOL_SAFETY", "150"))
DEADLINE = T0 + TIME_LIMIT - SAFETY
TEST_CSV = os.environ.get("IOL_TEST_CSV", "/tmp/data/test.csv")
OUT_CSV = os.environ.get("IOL_OUT_CSV", "submission.csv")
MODEL_ID = os.environ.get("IOL_MODEL", ".")
WANT_EXPL = os.environ.get("IOL_EXPLAIN", "1") == "1"
TOK_PER_S = float(os.environ.get("IOL_TOKS", "30"))
os.environ.setdefault("HF_HUB_OFFLINE", "1")
os.environ.setdefault("TRANSFORMERS_OFFLINE", "1")
os.environ.setdefault("TOKENIZERS_PARALLELISM", "false")
os.environ.setdefault("PYTORCH_CUDA_ALLOC_CONF", "expandable_segments:True")
MAX_NEW = int(os.environ.get("IOL_MAXNEW", "600")) # was 900 — faster passes, answers are short
MAX_SAMPLES= int(os.environ.get("IOL_MAXSAMPLES", "40")) # was 12/24 — more votes
BATCH_SIZE = int(os.environ.get("IOL_BATCH", "4")) # REVERT from 8 (this caused 0.1485)
SAMPLE_TEMP= float(os.environ.get("IOL_TEMP", "0.5")) # keep — value that gave 0.2039
COT = os.environ.get("IOL_COT", "0") == "1" # keep off
def log(m): print(f"[{time.time()-T0:7.1f}s] {m}", flush=True)
def left(): return DEADLINE - time.time()
# ---- item count + source fallback ------------------------------------------
_LINE_NUM=re.compile(r"^[ \t]*(\d{1,3})[.)\]]",re.M); _PAREN_NUM=re.compile(r"\((\d{1,3})\)")
_RANGE=re.compile(r"\(?(\d{1,3})\s*(?:[-–—]|to)\s*(\d{1,3})\)?"); _LINE_LET=re.compile(r"^[ \t]*([A-Z])[.)\]]\s",re.M)
def detect_n_items(query, task_type="", context=""):
q=query or ""
line=[int(m) for m in _LINE_NUM.findall(q)]; par=[int(m) for m in _PAREN_NUM.findall(q)]
rng=0
for a,b in _RANGE.findall(q):
a,b=int(a),int(b)
if 0<b-a<60: rng=max(rng,b-a+1)
cand=max(len(set(line)),len(set(par)))
if rng and cand and rng!=cand: return cand
cand=max(cand,len(set(_LINE_LET.findall(q)))); n=max(rng,cand)
if n>1: return n
lines=[l.strip() for l in q.splitlines() if l.strip()]
if len(lines)>1:
body=lines[1:] if lines[0].endswith((":",".")) else lines
if body: return len(body)
if context:
cn=len(set(int(m) for m in _LINE_NUM.findall(context)))
if cn>1: return cn
cl=len(set(_LINE_LET.findall(context)))
if cl>1: return cl
return max(n,1)
def extract_item_sources(query,n):
q=query or ""; out=[]
for ln in q.splitlines():
s=ln.strip()
if not s: continue
m=re.match(r"^\(?(\d{1,3})\)?[.):\]]\s*(.+)$",s)
if m: out.append(m.group(2).strip())
if not out:
lines=[l.strip() for l in q.splitlines() if l.strip()]
if len(lines)>1 and lines[0].endswith((":",".")): out=lines[1:]
out=[o.split("|")[0].strip() if "|" in o else o for o in out]; out=[o for o in out if o]
while len(out)<n: out.append(out[-1] if out else "?")
return out[:n]
# ---- parsing ----------------------------------------------------------------
_STRIP_PREFIX=re.compile(r"^\s*(?:\(?\d{1,3}\)?[.):\]]\s*|[-*•]\s+)"); _FENCE=re.compile(r"^```[a-zA-Z]*\s*$")
_CHATTY=re.compile(r"^\s*(?:here (?:are|is)\b|answers?\s*:?\s*$|explanation\b|note\b|okay\b|"
r"solution\b|reasoning\b|analysis\b|translations?\s*:?\s*$|the answers?\b|let me\b|first,|so,|therefore\b|thus\b)",re.I)
def clean_line(s):
s=s.strip(); s=_STRIP_PREFIX.sub("",s); s=s.strip().strip("`").strip()
if len(s)>=2 and s[0]==s[-1] and s[0] in "\"'“”": s=s[1:-1].strip()
return s.strip()
def fit_to_n(items,n,fb=None):
items=[i for i in items if i and i.strip()]
if len(items)>n: items=items[-n:]
while len(items)<n:
if fb and len(items)<len(fb): items.append(fb[len(items)])
else: items.append(items[-1] if items else "?")
return items[:n]
def raw_lines(text,n,fb=None):
"""Champion base parse: every non-empty line, forced to N, never blank."""
lines=[ln.strip() for ln in (text or "").splitlines() if ln.strip()]
return fit_to_n(lines,n,fb)
def parse_answers(text,n,fb=None):
"""Robust parse for CoT / samples: slice after ANSWERS:, else salvage tail."""
if not text: return list(fb[:n]) if fb else ["?"]*n
m=None
for m2 in re.finditer(r"(?:^|\n)\s*(?:final\s+)?answers?\s*:\s*\n?",text,re.I): m=m2
body=text[m.end():] if m else text
numbered,raw=[],[]
for ln in body.splitlines():
if _FENCE.match(ln): continue
mm=re.match(r"^\s*\(?(\d{1,3})\)?[.):\]]\s*(.+)$",ln.strip())
if mm:
v=clean_line(mm.group(2))
if v and not _CHATTY.match(v): numbered.append((int(mm.group(1)),v))
c=clean_line(ln)
if c and not _CHATTY.match(c): raw.append(c)
if len(numbered)>=n:
by={}; [by.__setitem__(l,v) for l,v in numbered]
labs=sorted(by)
if len(labs)>=n: return [by[l] for l in labs[:n]]
return fit_to_n(raw,n,fb)
def norm(s):
s=unicodedata.normalize("NFC",(s or "").strip().lower()); s=re.sub(r"\s+"," ",s)
return s.strip(" .!?;:,")
def vote(cands,anchor=None):
cands=[c for c in cands if c and c.strip()]
if anchor is None: anchor=cands[0] if cands else "?"
if len(cands)<3: return anchor
groups=defaultdict(list)
for c in cands: groups[norm(c)].append(c)
a_sup=len(groups.get(norm(anchor),[])); bk,bn=None,0
for k,v in groups.items():
if len(v)>bn: bk,bn=k,len(v)
if bk is not None and bn>=2 and bn>a_sup: return Counter(groups[bk]).most_common(1)[0][0]
return anchor
# ---- prompts ----------------------------------------------------------------
SYS_TRIVIAL=("You solve International Linguistics Olympiad problems. Answer every numbered "
"item. Put each answer on its own line, in order, with no numbering and no extra text.")
SYS_COT=("You solve International Linguistics Olympiad problems. Everything needed is in the "
"problem. Briefly analyse: segment the words, find the recurring morphemes and the rules "
"that order them, and check them against every given example. Keep the analysis short. "
"Then write a line containing exactly ANSWERS: and, below it, one answer per item in the "
"order asked — no numbering, no commentary, no blank lines. Never leave an item blank.")
def write_submission(path,ids,preds,expl=None):
import pandas as pd
rows=[]
for i in ids:
rec={"id":i,"pred":json.dumps(preds[i],ensure_ascii=False)}
if expl is not None: rec["explanation"]=expl.get(i,"")
rows.append(rec)
pd.DataFrame(rows).to_csv(path,index=False)
def main():
import pandas as pd, torch
from transformers import (AutoTokenizer,AutoModelForCausalLM,StoppingCriteria,StoppingCriteriaList)
torch.backends.cuda.matmul.allow_tf32=True; torch.backends.cudnn.allow_tf32=True
log(f"MODE: {'CoT' if COT else 'trivial'} | rep=1.0 | max_samples={MAX_SAMPLES} temp={SAMPLE_TEMP}")
df=pd.read_csv(TEST_CSV,dtype=str).fillna("")
ids=[str(x) for x in df["id"].tolist()]
ns=[detect_n_items(r.get("query",""),r.get("task_type",""),r.get("context","")) for _,r in df.iterrows()]
log(f"loaded {len(df)} problems, {sum(ns)} items")
srcs={i:extract_item_sources(r.get("query",""),n) for i,(_,r),n in zip(ids,df.iterrows(),ns)}
preds={i:list(srcs[i]) for i in ids}; expl={i:"" for i in ids} if WANT_EXPL else None
write_submission(OUT_CSV,ids,preds,expl); log(f"placeholder written ({len(ids)} rows)")
class Deadline(StoppingCriteria):
def __init__(self,t): self.t=t
def __call__(self,i,s,**k): return time.time()>self.t
log("loading model ...")
tok=AutoTokenizer.from_pretrained(MODEL_ID,trust_remote_code=True)
if tok.pad_token is None: tok.pad_token=tok.eos_token
tok.padding_side="left"
def _load(dm):
try: return AutoModelForCausalLM.from_pretrained(MODEL_ID,torch_dtype=torch.float16,device_map=dm,trust_remote_code=True).eval()
except TypeError: return AutoModelForCausalLM.from_pretrained(MODEL_ID,dtype=torch.float16,device_map=dm,trust_remote_code=True).eval()
try: model=_load({"":0} if torch.cuda.is_available() else "auto")
except Exception as e: log(f"pinned load failed ({e}); auto"); model=_load("auto")
log(f"model ready ({left():.0f}s left)")
sysmsg=SYS_COT if COT else SYS_TRIVIAL
base_parse=parse_answers if COT else raw_lines # <-- the champion distinction
prompts=[tok.apply_chat_template(
[{"role":"system","content":sysmsg},
{"role":"user","content":f"{r['context'].strip()}\n\n{r['query'].strip()}"}],
tokenize=False,add_generation_prompt=True) for _,r in df.iterrows()]
cb=BATCH_SIZE
def generate(texts,max_new,sample,temp=0.7):
nonlocal cb
out=[""]*len(texts); order=sorted(range(len(texts)),key=lambda i:len(texts[i])); i=0
while i<len(order):
if left()<25: break
idx=order[i:i+cb]; chunk=[texts[j] for j in idx]
try:
enc=tok(chunk,return_tensors="pt",padding=True,truncation=True,max_length=6144).to(model.device)
kw=dict(max_new_tokens=max_new,pad_token_id=tok.pad_token_id,repetition_penalty=1.0,
stopping_criteria=StoppingCriteriaList([Deadline(DEADLINE-10)]))
kw.update(dict(do_sample=True,temperature=temp,top_p=0.95) if sample else dict(do_sample=False))
with torch.no_grad(): o=model.generate(**enc,**kw)
for k,j in enumerate(idx): out[j]=tok.decode(o[k][enc["input_ids"].shape[1]:],skip_special_tokens=True)
i+=cb
except torch.cuda.OutOfMemoryError:
torch.cuda.empty_cache()
if cb==1: i+=1
else: cb=max(1,cb//2)
except Exception: i+=cb
return out
cap=max(MAX_NEW,1200) if COT else MAX_NEW # reasoning needs headroom
adaptive=int(0.40*max(1.0,left())*TOK_PER_S/max(1,len(df)))
max_new=max(192,min(cap,adaptive))
log(f"Pass 1 greedy, {max_new} tok/item")
t=time.time(); texts=generate(prompts,max_new=max_new,sample=False); c1=time.time()-t
samples={i:[] for i in ids}
for i,n,txt in zip(ids,ns,texts):
a=base_parse(txt,n,srcs[i]); preds[i]=a; samples[i].append(a)
write_submission(OUT_CSV,ids,preds,expl); log(f"Pass 1 done in {c1:.0f}s")
reserve=min(300.0,0.25*c1+60) if WANT_EXPL else 30.0
ne=0
while left()-reserve>c1*1.25 and ne<MAX_SAMPLES:
ne+=1; log(f"sample pass {ne} ({left():.0f}s left)")
texts=generate(prompts,max_new=max_new,sample=True,temp=SAMPLE_TEMP)
for i,n,txt in zip(ids,ns,texts):
if txt: samples[i].append(parse_answers(txt,n,srcs[i]))
for i,n in zip(ids,ns):
if len(samples[i])>=3:
g=samples[i][0]
preds[i]=fit_to_n([vote([s[k] for s in samples[i] if k<len(s)],
anchor=g[k] if k<len(g) else None) for k in range(n)],n,srcs[i])
write_submission(OUT_CSV,ids,preds,expl); log(f"voted over {ne+1} samples")
log(f"self-consistency: {ne} sample pass(es) completed")
if WANT_EXPL and left()>60:
log(f"explanations ({left():.0f}s left)")
ex_sys=("You explain International Linguistics Olympiad solutions to a human judge. State "
"the key rules of the language: morphemes, word order, sound changes. Concise (2-4 sentences).")
ep=[tok.apply_chat_template([{"role":"system","content":ex_sys},
{"role":"user","content":f"{r['context'].strip()}\n\n{r['query'].strip()}\n\nAnswers given:\n"
+"\n".join(f"- {a}" for a in preds[str(r['id'])])+"\n\nBriefly explain the linguistic rules."}],
tokenize=False,add_generation_prompt=True) for _,r in df.iterrows()]
for i,e in zip(ids,generate(ep,max_new=200,sample=False)):
e=re.sub(r"\s+"," ",(e or "").strip())
if e: expl[i]=e[:1200]
write_submission(OUT_CSV,ids,preds,expl); log("explanations written")
bad=[i for i,n in zip(ids,ns) if len(preds[i])!=n or any(not str(x).strip() for x in preds[i])]
if bad:
for i,n in zip(ids,ns): preds[i]=fit_to_n([x for x in preds[i] if str(x).strip()],n,srcs[i])
write_submission(OUT_CSV,ids,preds,expl)
log(f"DONE. {len(ids)} rows, {time.time()-T0:.0f}s elapsed.")
if __name__=="__main__":
main()

303282
tokenizer.json Normal file
View File

File diff suppressed because it is too large Load Diff

207
tokenizer_config.json Normal file
View File

@@ -0,0 +1,207 @@
{
"add_bos_token": false,
"add_prefix_space": false,
"added_tokens_decoder": {
"151643": {
"content": "<|endoftext|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151644": {
"content": "<|im_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151645": {
"content": "<|im_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151646": {
"content": "<|object_ref_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151647": {
"content": "<|object_ref_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151648": {
"content": "<|box_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151649": {
"content": "<|box_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151650": {
"content": "<|quad_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151651": {
"content": "<|quad_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151652": {
"content": "<|vision_start|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151653": {
"content": "<|vision_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151654": {
"content": "<|vision_pad|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151655": {
"content": "<|image_pad|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151656": {
"content": "<|video_pad|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151657": {
"content": "<tool_call>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151658": {
"content": "</tool_call>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151659": {
"content": "<|fim_prefix|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151660": {
"content": "<|fim_middle|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151661": {
"content": "<|fim_suffix|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151662": {
"content": "<|fim_pad|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151663": {
"content": "<|repo_name|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151664": {
"content": "<|file_sep|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
}
},
"additional_special_tokens": [
"<|im_start|>",
"<|im_end|>",
"<|object_ref_start|>",
"<|object_ref_end|>",
"<|box_start|>",
"<|box_end|>",
"<|quad_start|>",
"<|quad_end|>",
"<|vision_start|>",
"<|vision_end|>",
"<|vision_pad|>",
"<|image_pad|>",
"<|video_pad|>"
],
"bos_token": null,
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
"clean_up_tokenization_spaces": false,
"eos_token": "<|im_end|>",
"errors": "replace",
"model_max_length": 131072,
"pad_token": "<|endoftext|>",
"split_special_tokens": false,
"tokenizer_class": "Qwen2Tokenizer",
"unk_token": null
}

1
vocab.json Normal file
View File

File diff suppressed because one or more lines are too long