nl2sql: sql_prompt.py 모듈 분리 + eval/training/verifier 파이프라인

scripts/run-vllm-eval-model.sh

@@ -0,0 +1,46 @@
+#!/usr/bin/env bash
+# Phase 0 eval 대상 모델 서빙 (소형 dense 등). 35B 전용 플래그(instanttensor/MTP) 없음.
+# 사용: bash scripts/run-vllm-eval-model.sh <hf_id_or_path> <served-name> [gpu_util] [max_len] [port]
+set -euo pipefail
+
+MODEL="${1:?model HF id or path}"
+NAME="${2:?served-model-name}"
+UTIL="${3:-0.30}"
+MAXLEN="${4:-32768}"
+PORT="${5:-8001}"
+CNAME="vllm_eval"
+
+docker rm -f "$CNAME" 2>/dev/null || true
+
+docker run -d --name "$CNAME" \
+  --gpus all --network host --ipc host \
+  --ulimit memlock=-1 --ulimit stack=67108864 \
+  -v /home/windpacer/.cache/huggingface:/root/.cache/huggingface \
+  -v /home/windpacer/.cache/vllm:/root/.cache/vllm \
+  -v /home/windpacer/ai-models:/root/ai-models \
+  --entrypoint "" \
+  vllm-node-tf5 \
+  bash -c "
+exec vllm serve ${MODEL} \
+  --served-model-name ${NAME} \
+  --max-model-len ${MAXLEN} \
+  --max-num-seqs 8 \
+  --gpu-memory-utilization ${UTIL} \
+  --port ${PORT} --host 0.0.0.0 \
+  --enable-chunked-prefill \
+  --enable-auto-tool-choice --tool-call-parser hermes \
+  --trust-remote-code \
+  --kv-cache-dtype fp8 \
+  -tp 1
+"
+
+echo "Waiting for ${NAME} on :${PORT} ..."
+for i in $(seq 1 48); do
+  if curl -sf "http://localhost:${PORT}/v1/models" >/dev/null 2>&1; then
+    echo "✓ Ready: ${NAME}"
+    curl -s "http://localhost:${PORT}/v1/models" | python3 -m json.tool 2>/dev/null | grep '"id"' || true
+    exit 0
+  fi
+  sleep 5
+done
+echo "✗ Timed out — docker logs ${CNAME}"; exit 1

scripts/templates/qwen3-nothink.jinja

@@ -0,0 +1,90 @@
+{%- if enable_thinking is not defined %}{%- set enable_thinking = false %}{%- endif %}
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}