diff --git a/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/README.md b/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/README.md
index af7de6df..93e66efb 100644
--- a/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/README.md
+++ b/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/README.md
@@ -29,3 +29,4 @@
| [text-to-python.ipynb](./text-to-code/text-to-python.ipynb) | Text-to-Code | Generate Python code from natural language instructions with validation and evaluation |
| [text-to-python-evol.ipynb](./text-to-code/text-to-python-evol.ipynb) | Text-to-Code | Build advanced Python code generation with evolutionary improvements and iterative refinement |
| [text-to-sql.ipynb](./text-to-code/text-to-sql.ipynb) | Text-to-Code | Create SQL queries from natural language descriptions with validation and testing |
+| [japanese_commonsense_qa_data_generator_nemotron_persona_jp_seed.ipynb](./nemotron-persona-jp/japanese_commonsense_qa_data_generator_nemotron_persona_jp_seed.ipynb) | Text-to-Code | Create SQL queries from natural language descriptions with validation and testing |
diff --git a/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/nemotron-persona-jp/japanese_commonsense_qa_data_generator_nemotron_persona_jp_seed.ipynb b/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/nemotron-persona-jp/japanese_commonsense_qa_data_generator_nemotron_persona_jp_seed.ipynb
index a9c6bc71..1a9b2758 100644
--- a/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/nemotron-persona-jp/japanese_commonsense_qa_data_generator_nemotron_persona_jp_seed.ipynb
+++ b/nemo/NeMo-Data-Designer/self-hosted-tutorials/community-contributions/nemotron-persona-jp/japanese_commonsense_qa_data_generator_nemotron_persona_jp_seed.ipynb
@@ -5,7 +5,7 @@
"id": "header",
"metadata": {},
"source": [
- "# ๐จ NeMo Data Designer: Japanese Commonsense Reasoning Dataset Generation (Improved Version)\n",
+ "# ๐จ NeMo Data Designer: Japanese Commonsense Reasoning Dataset Generation\n",
"\n",
"#### ๐ Overview\n",
"\n",
@@ -88,7 +88,7 @@
},
{
"cell_type": "code",
- "execution_count": 70,
+ "execution_count": 3,
"id": "model_config_code",
"metadata": {},
"outputs": [],
@@ -96,7 +96,7 @@
"MODEL_PROVIDER = \"nvidiabuild\"\n",
"MODEL_ID = \"openai/gpt-oss-120b\"\n",
"MODEL_ALIAS = \"gpt-oss-120b\"\n",
- "SYSTEM_PROMPT = \"\" \n",
+ "SYSTEM_PROMPT = \"\"\n",
"JUDGE_MODEL_ALIAS = \"quality-judge\"\n",
"\n",
"model_configs = [\n",
@@ -1300,7 +1300,7 @@
},
{
"cell_type": "code",
- "execution_count": 71,
+ "execution_count": 20,
"id": "config_with_seed_code",
"metadata": {},
"outputs": [],
@@ -1314,7 +1314,7 @@
},
{
"cell_type": "code",
- "execution_count": 72,
+ "execution_count": 21,
"id": "be7698ab-4dac-4491-aad2-ba2ad67df724",
"metadata": {},
"outputs": [
@@ -1322,13 +1322,13 @@
"name": "stderr",
"output_type": "stream",
"text": [
- "[11:35:31] [INFO] ๐ Uploading seed dataset to datastore\n"
+ "[15:03:51] [INFO] ๐ Uploading seed dataset to datastore\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "2f5dd6f00f0846a88b1597b4b81d14f7",
+ "model_id": "f151f7e25f84424eaa890f36d335fff5",
"version_major": 2,
"version_minor": 0
},
@@ -1350,7 +1350,7 @@
},
{
"cell_type": "code",
- "execution_count": 73,
+ "execution_count": 22,
"id": "f72b780b-1838-4288-b7a2-f96a8276d21c",
"metadata": {},
"outputs": [
@@ -1574,7 +1574,7 @@
")"
]
},
- "execution_count": 73,
+ "execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
@@ -1598,7 +1598,7 @@
},
{
"cell_type": "code",
- "execution_count": 74,
+ "execution_count": 23,
"id": "jcqa_generation_code",
"metadata": {},
"outputs": [
@@ -1669,7 +1669,7 @@
},
{
"cell_type": "code",
- "execution_count": 75,
+ "execution_count": 24,
"id": "quality_metrics_code",
"metadata": {},
"outputs": [
@@ -1761,7 +1761,7 @@
},
{
"cell_type": "code",
- "execution_count": 77,
+ "execution_count": 25,
"id": "preview_with_seed",
"metadata": {
"scrolled": true
@@ -1771,10 +1771,10 @@
"name": "stderr",
"output_type": "stream",
"text": [
- "[11:35:59] [INFO] โ
Validation passed\n",
- "[11:35:59] [INFO] ๐ Starting preview generation\n",
- "[11:35:59] [INFO] โ๏ธ Sorting column configs into a Directed Acyclic Graph\n",
- "[11:35:59] [INFO] ๐ฉบ Running health checks for models...\n"
+ "[15:03:56] [INFO] โ
Validation passed\n",
+ "[15:03:56] [INFO] ๐ Starting preview generation\n",
+ "[15:03:56] [INFO] โ๏ธ Sorting column configs into a Directed Acyclic Graph\n",
+ "[15:03:56] [INFO] ๐ฉบ Running health checks for models...\n"
]
},
{
@@ -1791,17 +1791,17 @@
"name": "stderr",
"output_type": "stream",
"text": [
- "[11:36:00] [INFO] |-- ๐ Checking 'openai/gpt-oss-120b' in provider named 'nvidiabuild' for model alias 'quality-judge'...\n",
- "[11:36:00] [INFO] |-- โ
Passed!\n",
- "[11:36:00] [INFO] |-- ๐ Checking 'openai/gpt-oss-120b' in provider named 'nvidiabuild' for model alias 'gpt-oss-120b'...\n",
- "[11:36:00] [INFO] |-- โ
Passed!\n",
- "[11:36:01] [INFO] โณ Processing batch 1 of 1\n",
- "[11:36:01] [INFO] ๐ฑ Sampling 1 records from seed dataset\n",
- "[11:36:01] [INFO] |-- seed dataset size: 2000 records\n",
- "[11:36:01] [INFO] |-- sampling strategy: ordered\n",
- "[11:36:01] [INFO] ๐๏ธ Preparing llm-structured column generation\n",
- "[11:36:01] [INFO] |-- column name: 'jcqa_data'\n",
- "[11:36:01] [INFO] |-- model config:\n",
+ "[15:03:57] [INFO] |-- ๐ Checking 'openai/gpt-oss-120b' in provider named 'nvidiabuild' for model alias 'gpt-oss-120b'...\n",
+ "[15:03:57] [INFO] |-- โ
Passed!\n",
+ "[15:03:58] [INFO] |-- ๐ Checking 'openai/gpt-oss-120b' in provider named 'nvidiabuild' for model alias 'quality-judge'...\n",
+ "[15:03:58] [INFO] |-- โ
Passed!\n",
+ "[15:03:58] [INFO] โณ Processing batch 1 of 1\n",
+ "[15:03:58] [INFO] ๐ฑ Sampling 1 records from seed dataset\n",
+ "[15:03:58] [INFO] |-- seed dataset size: 2000 records\n",
+ "[15:03:58] [INFO] |-- sampling strategy: ordered\n",
+ "[15:03:58] [INFO] ๐๏ธ Preparing llm-structured column generation\n",
+ "[15:03:58] [INFO] |-- column name: 'jcqa_data'\n",
+ "[15:03:58] [INFO] |-- model config:\n",
"{\n",
" \"alias\": \"gpt-oss-120b\",\n",
" \"model\": \"openai/gpt-oss-120b\",\n",
@@ -1815,10 +1815,10 @@
" },\n",
" \"provider\": \"nvidiabuild\"\n",
"}\n",
- "[11:36:06] [INFO] ๐ Processing llm-structured column 'jcqa_data' with 8 concurrent workers\n",
- "[11:36:09] [INFO] โ๏ธ Preparing llm-judge column generation\n",
- "[11:36:09] [INFO] |-- column name: 'quality_metrics'\n",
- "[11:36:09] [INFO] |-- model config:\n",
+ "[15:04:03] [INFO] ๐ Processing llm-structured column 'jcqa_data' with 8 concurrent workers\n",
+ "[15:04:05] [INFO] โ๏ธ Preparing llm-judge column generation\n",
+ "[15:04:05] [INFO] |-- column name: 'quality_metrics'\n",
+ "[15:04:05] [INFO] |-- model config:\n",
"{\n",
" \"alias\": \"quality-judge\",\n",
" \"model\": \"openai/gpt-oss-120b\",\n",
@@ -1832,76 +1832,76 @@
" },\n",
" \"provider\": \"nvidiabuild\"\n",
"}\n",
- "[11:36:13] [INFO] ๐ Processing llm-judge column 'quality_metrics' with 4 concurrent workers\n",
- "[11:36:13] [INFO] ๐งฉ Generating column `clarity_score` from expression\n",
- "[11:36:13] [INFO] ๐งฉ Generating column `difficulty` from expression\n",
- "[11:36:13] [INFO] ๐ Model usage summary:\n",
+ "[15:04:10] [INFO] ๐ Processing llm-judge column 'quality_metrics' with 4 concurrent workers\n",
+ "[15:04:10] [INFO] ๐งฉ Generating column `clarity_score` from expression\n",
+ "[15:04:10] [INFO] ๐งฉ Generating column `difficulty` from expression\n",
+ "[15:04:10] [INFO] ๐ Model usage summary:\n",
"{\n",
" \"openai/gpt-oss-120b\": {\n",
" \"token_usage\": {\n",
- " \"prompt_tokens\": 1216,\n",
- " \"completion_tokens\": 934,\n",
- " \"total_tokens\": 2150\n",
+ " \"prompt_tokens\": 1653,\n",
+ " \"completion_tokens\": 563,\n",
+ " \"total_tokens\": 2216\n",
" },\n",
" \"request_usage\": {\n",
" \"successful_requests\": 1,\n",
" \"failed_requests\": 0,\n",
" \"total_requests\": 1\n",
" },\n",
- " \"tokens_per_second\": 168,\n",
- " \"requests_per_minute\": 4\n",
+ " \"tokens_per_second\": 187,\n",
+ " \"requests_per_minute\": 5\n",
" }\n",
"}\n",
- "[11:36:13] [INFO] ๐ Measuring dataset column statistics:\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'uuid'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'professional_persona'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'sports_persona'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'arts_persona'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'travel_persona'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'culinary_persona'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'persona'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'cultural_background'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'skills_and_expertise'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'skills_and_expertise_list'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'hobbies_and_interests'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'hobbies_and_interests_list'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'career_goals_and_ambitions'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'sex'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'age'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'marital_status'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'education_level'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'occupation'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'region'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'area'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'prefecture'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'country'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'age_band'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_all_text'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_core_text'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_core_len'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_attr_key'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_finance'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_safety'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_vocab'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_public'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_tools'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_life'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_geo'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'score_culture'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_geo_text'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_tools_text'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_kw_hits'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'jc_category'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'max_score_any'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_public_bonus'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: '_religion_pen'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'jc_theme'\n",
- "[11:36:13] [INFO] |-- ๐ฑ column: 'topic_category'\n",
- "[11:36:13] [INFO] |-- ๐๏ธ column: 'jcqa_data'\n",
- "[11:36:13] [INFO] |-- โ๏ธ column: 'quality_metrics'\n",
- "[11:36:13] [INFO] |-- ๐งฉ column: 'clarity_score'\n",
- "[11:36:13] [INFO] |-- ๐งฉ column: 'difficulty'\n",
- "[11:36:13] [INFO] ๐ Preview complete!\n"
+ "[15:04:10] [INFO] ๐ Measuring dataset column statistics:\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'uuid'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'professional_persona'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'sports_persona'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'arts_persona'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'travel_persona'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'culinary_persona'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'persona'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'cultural_background'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'skills_and_expertise'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'skills_and_expertise_list'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'hobbies_and_interests'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'hobbies_and_interests_list'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'career_goals_and_ambitions'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'sex'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'age'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'marital_status'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'education_level'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'occupation'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'region'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'area'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'prefecture'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'country'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'age_band'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_all_text'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_core_text'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_core_len'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_attr_key'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_finance'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_safety'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_vocab'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_public'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_tools'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_life'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_geo'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'score_culture'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_geo_text'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_tools_text'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_kw_hits'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'jc_category'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'max_score_any'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_public_bonus'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: '_religion_pen'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'jc_theme'\n",
+ "[15:04:10] [INFO] |-- ๐ฑ column: 'topic_category'\n",
+ "[15:04:10] [INFO] |-- ๐๏ธ column: 'jcqa_data'\n",
+ "[15:04:10] [INFO] |-- โ๏ธ column: 'quality_metrics'\n",
+ "[15:04:10] [INFO] |-- ๐งฉ column: 'clarity_score'\n",
+ "[15:04:10] [INFO] |-- ๐งฉ column: 'difficulty'\n",
+ "[15:04:10] [INFO] โ
Preview complete!\n"
]
},
{
@@ -2043,19 +2043,19 @@
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
"โ clarity_score โ ๆ็ขบ โ\n",
"โโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ difficulty โ ๆฎ้ โ\n",
+ "โ difficulty โ ๆใใ โ\n",
"โโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
"โ jcqa_data โ { โ\n",
"โ โ 'answer_index': 0, โ\n",
- "โ โ 'choice0': '็พ้๏ผ็ดๅนฃใป็กฌ่ฒจ๏ผ', โ\n",
- "โ โ 'choice1': 'ใฏใฌใธใใใซใผใ', โ\n",
- "โ โ 'choice2': '้ปๅญใใใผ๏ผSuicaใปICOCA ใชใฉ๏ผ', โ\n",
- "โ โ 'choice3': 'ๅฐๅๆ', โ\n",
- "โ โ 'choice4': 'ในใใผใใใฉใณๆฑบๆธ๏ผPayPay ใชใฉ๏ผ', โ\n",
+ "โ โ 'choice0': '็พ้ใงๆฏๆใ', โ\n",
+ "โ โ 'choice1': 'ใฏใฌใธใใใซใผใใงๆฏๆใ', โ\n",
+ "โ โ 'choice2': 'ใขใใคใซSuicaใงๆฏๆใ', โ\n",
+ "โ โ 'choice3': 'ไบค้็ณปICใซใผใใฎใใฃใผใธใ่กใ', โ\n",
+ "โ โ 'choice4': '้ปๅญใใใผ๏ผๆฅฝๅคฉใใค๏ผใงๆฏๆใ', โ\n",
"โ โ 'question': โ\n",
- "โ โ 'ไธ้็ใฎๅ
ฌๅ
ฑๅณๆธ้คจใงๆฌใฎ่ฟๅดๆ้ใ้ใใ้ใฎๅปถๆป้ใๆฏๆใใจใใๅไปใงไธ่ฌ็ใซๅใไปใใฆใใ โฆ โ\n",
+ "โ โ 'ไธ้็ไผๅข้ง
ใฎ็ชๅฃใงใICใซใผใใฎๆฎ้ซใไธ่ถณใใฆใใใใไน่ปๅธใ่ณผๅ
ฅใใใใจใใฆใใใ็ชๅฃใงใฎ โฆ โ\n",
"โ โ 'reasoning': โ\n",
- "โ โ 'ๆฅๆฌใฎๅคใใฎๅ
ฌๅ
ฑๅณๆธ้คจใงใฏใๅปถๆป้ใฎๆฏๆใใฏ็ชๅฃใงใฎ็พ้ใฎใฟใๆจๆบ็ใซๅใไปใใใใฆใใพใใ โฆ โ\n",
+ "โ โ '้ง
ใฎ็ชๅฃใงไน่ปๅธใ่ณผๅ
ฅใใ้ใฏใๆใไธ่ฌ็ใใค็ขบๅฎใซๅใไปใใใใๆฏๆใๆนๆณใฏ็พ้ใงใใใใฏ โฆ โ\n",
"โ โ } โ\n",
"โโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
@@ -2064,9 +2064,9 @@
"โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
"โ difficulty โ question_clarity โ\n",
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ score: ๆฎ้ โ score: ๆ็ขบ โ\n",
+ "โ score: ๆใใ โ score: ๆ็ขบ โ\n",
"โ reasoning: โ reasoning: โ\n",
- "โ ใใฎๅ้กใฏๆฅๆฌใฎๅ
ฌๅ
ฑๅณๆธ้คจใฎไธ่ฌ็ใช้ๅถๅฎๅใซ้ขใใ โฆ โ ่ณชๅใฏใไธ้็ใฎๅ
ฌๅ
ฑๅณๆธ้คจใงๆฌใฎ่ฟๅดๆ้ใ้ใใ้ใฎ โฆ โ\n",
+ "โ ๆฅๆฌใฎ้ง
็ชๅฃใงใฎๆฏๆใๆนๆณใฏๅบใ็ฅใใใๅธธ่ญใงใใใ โฆ โ ่ณชๅใฏๅ
ทไฝ็ใชใทใใฅใจใผใทใงใณ๏ผไผๅข้ง
ใฎ็ชๅฃใงICใซใผ โฆ โ\n",
"โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
" [index: 0] \n",
@@ -2201,19 +2201,19 @@
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
"โ clarity_score โ ๆ็ขบ โ\n",
"โโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ difficulty โ ๆฎ้ โ\n",
+ "โ difficulty โ ๆใใ โ\n",
"โโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
"โ jcqa_data โ \u001b[1m{\u001b[0m โ\n",
"โ โ \u001b[32m'answer_index'\u001b[0m: \u001b[1;36m0\u001b[0m, โ\n",
- "โ โ \u001b[32m'choice0'\u001b[0m: \u001b[32m'็พ้๏ผ็ดๅนฃใป็กฌ่ฒจ๏ผ'\u001b[0m, โ\n",
- "โ โ \u001b[32m'choice1'\u001b[0m: \u001b[32m'ใฏใฌใธใใใซใผใ'\u001b[0m, โ\n",
- "โ โ \u001b[32m'choice2'\u001b[0m: \u001b[32m'้ปๅญใใใผ๏ผSuicaใปICOCA ใชใฉ๏ผ'\u001b[0m, โ\n",
- "โ โ \u001b[32m'choice3'\u001b[0m: \u001b[32m'ๅฐๅๆ'\u001b[0m, โ\n",
- "โ โ \u001b[32m'choice4'\u001b[0m: \u001b[32m'ในใใผใใใฉใณๆฑบๆธ๏ผPayPay ใชใฉ๏ผ'\u001b[0m, โ\n",
+ "โ โ \u001b[32m'choice0'\u001b[0m: \u001b[32m'็พ้ใงๆฏๆใ'\u001b[0m, โ\n",
+ "โ โ \u001b[32m'choice1'\u001b[0m: \u001b[32m'ใฏใฌใธใใใซใผใใงๆฏๆใ'\u001b[0m, โ\n",
+ "โ โ \u001b[32m'choice2'\u001b[0m: \u001b[32m'ใขใใคใซSuicaใงๆฏๆใ'\u001b[0m, โ\n",
+ "โ โ \u001b[32m'choice3'\u001b[0m: \u001b[32m'ไบค้็ณปICใซใผใใฎใใฃใผใธใ่กใ'\u001b[0m, โ\n",
+ "โ โ \u001b[32m'choice4'\u001b[0m: \u001b[32m'้ปๅญใใใผ๏ผๆฅฝๅคฉใใค๏ผใงๆฏๆใ'\u001b[0m, โ\n",
"โ โ \u001b[32m'question'\u001b[0m: โ\n",
- "โ โ \u001b[32m'ไธ้็ใฎๅ
ฌๅ
ฑๅณๆธ้คจใงๆฌใฎ่ฟๅดๆ้ใ้ใใ้ใฎๅปถๆป้ใๆฏๆใใจใใๅไปใงไธ่ฌ็ใซๅใไปใใฆใใ โฆ\u001b[0m โ\n",
+ "โ โ \u001b[32m'ไธ้็ไผๅข้ง
ใฎ็ชๅฃใงใICใซใผใใฎๆฎ้ซใไธ่ถณใใฆใใใใไน่ปๅธใ่ณผๅ
ฅใใใใจใใฆใใใ็ชๅฃใงใฎ โฆ\u001b[0m โ\n",
"โ โ \u001b[32m'reasoning'\u001b[0m: โ\n",
- "โ โ \u001b[32m'ๆฅๆฌใฎๅคใใฎๅ
ฌๅ
ฑๅณๆธ้คจใงใฏใๅปถๆป้ใฎๆฏๆใใฏ็ชๅฃใงใฎ็พ้ใฎใฟใๆจๆบ็ใซๅใไปใใใใฆใใพใใ โฆ\u001b[0m โ\n",
+ "โ โ \u001b[32m'้ง
ใฎ็ชๅฃใงไน่ปๅธใ่ณผๅ
ฅใใ้ใฏใๆใไธ่ฌ็ใใค็ขบๅฎใซๅใไปใใใใๆฏๆใๆนๆณใฏ็พ้ใงใใใใฏ โฆ\u001b[0m โ\n",
"โ โ \u001b[1m}\u001b[0m โ\n",
"โโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
@@ -2222,9 +2222,9 @@
"โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
"โ\u001b[1m \u001b[0m\u001b[1mdifficulty \u001b[0m\u001b[1m \u001b[0mโ\u001b[1m \u001b[0m\u001b[1mquestion_clarity \u001b[0m\u001b[1m \u001b[0mโ\n",
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ score: ๆฎ้ โ score: ๆ็ขบ โ\n",
+ "โ score: ๆใใ โ score: ๆ็ขบ โ\n",
"โ reasoning: โ reasoning: โ\n",
- "โ ใใฎๅ้กใฏๆฅๆฌใฎๅ
ฌๅ
ฑๅณๆธ้คจใฎไธ่ฌ็ใช้ๅถๅฎๅใซ้ขใใ โฆ โ ่ณชๅใฏใไธ้็ใฎๅ
ฌๅ
ฑๅณๆธ้คจใงๆฌใฎ่ฟๅดๆ้ใ้ใใ้ใฎ โฆ โ\n",
+ "โ ๆฅๆฌใฎ้ง
็ชๅฃใงใฎๆฏๆใๆนๆณใฏๅบใ็ฅใใใๅธธ่ญใงใใใ โฆ โ ่ณชๅใฏๅ
ทไฝ็ใชใทใใฅใจใผใทใงใณ๏ผไผๅข้ง
ใฎ็ชๅฃใงICใซใผ โฆ โ\n",
"โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
" [index: 0] \n"
@@ -2251,7 +2251,7 @@
},
{
"cell_type": "code",
- "execution_count": 78,
+ "execution_count": 26,
"id": "preview_analysis",
"metadata": {
"scrolled": true
@@ -2378,7 +2378,7 @@
"โ โ โ โ prompt tokens โ completion tokens โ\n",
"โ column name โ data type โ number unique values โ per record โ per record โ\n",
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ jcqa_data โ dict โ 1 (100.0%) โ 1319.0 +/- 0.0 โ 396.0 +/- nan โ\n",
+ "โ jcqa_data โ dict โ 1 (100.0%) โ 1319.0 +/- 0.0 โ 361.0 +/- nan โ\n",
"โโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
" \n",
@@ -2387,7 +2387,7 @@
"โ โ โ โ prompt tokens โ completion tokens โ\n",
"โ column name โ data type โ number unique values โ per record โ per record โ\n",
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ quality_metrics โ dict โ 1 (100.0%) โ 1746.0 +/- 0.0 โ 272.0 +/- nan โ\n",
+ "โ quality_metrics โ dict โ 1 (100.0%) โ 1714.0 +/- 0.0 โ 282.0 +/- nan โ\n",
"โโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
" \n",
@@ -2522,7 +2522,7 @@
"โ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m prompt tokens\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m completion tokens\u001b[0m\u001b[1;2m \u001b[0mโ\n",
"โ\u001b[1;2m \u001b[0m\u001b[1;2mcolumn name \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m data type\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m number unique values\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\n",
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ jcqa_data โ dict โ 1 (100.0%) โ 1319.0 +/- 0.0 โ 396.0 +/- nan โ\n",
+ "โ jcqa_data โ dict โ 1 (100.0%) โ 1319.0 +/- 0.0 โ 361.0 +/- nan โ\n",
"โโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
" \n",
@@ -2531,7 +2531,7 @@
"โ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m prompt tokens\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m completion tokens\u001b[0m\u001b[1;2m \u001b[0mโ\n",
"โ\u001b[1;2m \u001b[0m\u001b[1;2mcolumn name \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m data type\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m number unique values\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\n",
"โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ quality_metrics โ dict โ 1 (100.0%) โ 1746.0 +/- 0.0 โ 272.0 +/- nan โ\n",
+ "โ quality_metrics โ dict โ 1 (100.0%) โ 1714.0 +/- 0.0 โ 282.0 +/- nan โ\n",
"โโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโ\n",
" \n",
" \n",
@@ -2567,7 +2567,7 @@
},
{
"cell_type": "code",
- "execution_count": 79,
+ "execution_count": 27,
"id": "preview_dataset",
"metadata": {
"scrolled": true
@@ -2613,16 +2613,16 @@
"0 None C_ๆฏๆใใปใ้ ๅ
ฌๅ
ฑใฎๅ ด \n",
"\n",
" jcqa_data \\\n",
- "0 {'answer_index': 0, 'choice0': '็พ้๏ผ็ดๅนฃใป็กฌ่ฒจ๏ผ', 'c... \n",
+ "0 {'answer_index': 0, 'choice0': '็พ้ใงๆฏๆใ', 'choi... \n",
"\n",
" jcqa_data__reasoning_trace \\\n",
"0 We need to output JSON with fields: question, ... \n",
"\n",
" quality_metrics \\\n",
- "0 {'difficulty': {'reasoning': 'ใใฎๅ้กใฏๆฅๆฌใฎๅ
ฌๅ
ฑๅณๆธ้คจใฎไธ่ฌ... \n",
+ "0 {'difficulty': {'reasoning': 'ๆฅๆฌใฎ้ง
็ชๅฃใงใฎๆฏๆใๆนๆณใฏๅบใ... \n",
"\n",
" quality_metrics__reasoning_trace clarity_score difficulty \n",
- "0 We need to evaluate the generated data's quali... ๆ็ขบ ๆฎ้ \n",
+ "0 We need to evaluate the generated data's quali... ๆ็ขบ ๆใใ \n",
"\n",
"[1 rows x 50 columns]\n"
]
@@ -2681,314 +2681,10 @@
},
{
"cell_type": "code",
- "execution_count": 30,
+ "execution_count": null,
"id": "analysis_with_seed",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\n",
- "================================================================================\n",
- "Seedใใผใฟใใ็ใฎๅๆ\n",
- "================================================================================\n"
- ]
- },
- {
- "data": {
- "text/html": [
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ ๐จ Data Designer Dataset Profile โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "\n",
- " \n",
- " Dataset Overview \n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ number of records โ number of columns โ percent complete records โ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ 7,992 โ 48 โ 99.9% โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- " ๐ฑ Seed-Dataset Columns \n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ column name โ data type โ number unique values โ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ uuid โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ professional_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ sports_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ arts_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ travel_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ culinary_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ cultural_background โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ skills_and_expertise โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ skills_and_expertise_list โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ hobbies_and_interests โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ hobbies_and_interests_list โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ career_goals_and_ambitions โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ sex โ string โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ age โ int โ 82 (1.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ marital_status โ string โ 8 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ education_level โ string โ 10 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ occupation โ string โ 615 (7.7%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ region โ string โ 8 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ area โ string โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ prefecture โ string โ 47 (0.6%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ country โ string โ 1 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ age_band โ None โ 0 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _all_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _core_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _core_len โ int โ 130 (1.6%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _attr_key โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_finance โ int โ 4 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_safety โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_vocab โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_public โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_tools โ int โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_life โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_geo โ int โ 7 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_culture โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _geo_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _tools_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _kw_hits โ int โ 7 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ jc_category โ string โ 8 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ max_score_any โ float โ 6 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _public_bonus โ None โ 0 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _religion_pen โ None โ 0 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ jc_theme โ string โ 6 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ topic_category โ string โ 7 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- " ๐๏ธ LLM-Structured Columns \n",
- "โโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ โ โ โ prompt tokens โ completion tokens โ\n",
- "โ column name โ data type โ number unique values โ per record โ per record โ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ jcqa_data โ dict โ 7992 (100.0%) โ 1322.0 +/- 7.7 โ 318.0 +/- 64.9 โ\n",
- "โโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- " โ๏ธ LLM-Judge Columns \n",
- "โโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ โ โ โ prompt tokens โ completion tokens โ\n",
- "โ column name โ data type โ number unique values โ per record โ per record โ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ quality_metrics โ dict โ 7992 (100.0%) โ 1667.0 +/- 66.5 โ 281.0 +/- 42.2 โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- " ๐งฉ Expression Columns \n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ column name โ data type โ number unique values โ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ clarity_score โ string โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ difficulty โ string โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- "โญโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ Table Notes โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฎ\n",
- "โ โ\n",
- "โ 1. All token statistics are based on a sample of max(1000, len(dataset)) records. โ\n",
- "โ 2. Tokens are calculated using tiktoken's cl100k_base tokenizer. โ\n",
- "โ โ\n",
- "โฐโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฏ\n",
- " \n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "
\n"
- ],
- "text/plain": [
- "\u001b[1;38;2;118;185;0mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ \u001b[0m๐จ Data Designer Dataset Profile\u001b[1;38;2;118;185;0m โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m\n",
- "\n",
- " \n",
- "\u001b[1;38;2;118;185;0m Dataset Overview \u001b[0m\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ\u001b[1;2m \u001b[0m\u001b[1;2mnumber of records \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2mnumber of columns \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2mpercent complete records \u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ 7,992 โ 48 โ 99.9% โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- "\u001b[1;38;2;118;185;0m ๐ฑ Seed-Dataset Columns \u001b[0m\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ\u001b[1;2m \u001b[0m\u001b[1;2mcolumn name \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m data type\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m number unique values\u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ uuid โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ professional_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ sports_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ arts_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ travel_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ culinary_persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ persona โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ cultural_background โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ skills_and_expertise โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ skills_and_expertise_list โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ hobbies_and_interests โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ hobbies_and_interests_list โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ career_goals_and_ambitions โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ sex โ string โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ age โ int โ 82 (1.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ marital_status โ string โ 8 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ education_level โ string โ 10 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ occupation โ string โ 615 (7.7%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ region โ string โ 8 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ area โ string โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ prefecture โ string โ 47 (0.6%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ country โ string โ 1 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ age_band โ \u001b[3;35mNone\u001b[0m โ 0 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _all_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _core_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _core_len โ int โ 130 (1.6%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _attr_key โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_finance โ int โ 4 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_safety โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_vocab โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_public โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_tools โ int โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_life โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_geo โ int โ 7 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ score_culture โ int โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _geo_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _tools_text โ string โ 2000 (25.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _kw_hits โ int โ 7 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ jc_category โ string โ 8 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ max_score_any โ float โ 6 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _public_bonus โ \u001b[3;35mNone\u001b[0m โ 0 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ _religion_pen โ \u001b[3;35mNone\u001b[0m โ 0 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ jc_theme โ string โ 6 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ topic_category โ string โ 7 (0.1%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- "\u001b[1;38;2;118;185;0m ๐๏ธ LLM-Structured Columns \u001b[0m\n",
- "โโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m prompt tokens\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m completion tokens\u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โ\u001b[1;2m \u001b[0m\u001b[1;2mcolumn name \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m data type\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m number unique values\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ jcqa_data โ dict โ 7992 (100.0%) โ 1322.0 +/- 7.7 โ 318.0 +/- 64.9 โ\n",
- "โโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- "\u001b[1;38;2;118;185;0m โ๏ธ LLM-Judge Columns \u001b[0m\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m prompt tokens\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m completion tokens\u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โ\u001b[1;2m \u001b[0m\u001b[1;2mcolumn name \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m data type\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m number unique values\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m per record\u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ quality_metrics โ dict โ 7992 (100.0%) โ 1667.0 +/- 66.5 โ 281.0 +/- 42.2 โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- "\u001b[1;38;2;118;185;0m ๐งฉ Expression Columns \u001b[0m\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโณโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- "โ\u001b[1;2m \u001b[0m\u001b[1;2mcolumn name \u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m data type\u001b[0m\u001b[1;2m \u001b[0mโ\u001b[1;2m \u001b[0m\u001b[1;2m number unique values\u001b[0m\u001b[1;2m \u001b[0mโ\n",
- "โกโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฉ\n",
- "โ clarity_score โ string โ 2 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค\n",
- "โ difficulty โ string โ 3 (0.0%) โ\n",
- "โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\n",
- " \n",
- " \n",
- "\u001b[2mโญโ\u001b[0m\u001b[2mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m\u001b[2m Table Notes \u001b[0m\u001b[2mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m\u001b[2mโโฎ\u001b[0m\n",
- "\u001b[2mโ\u001b[0m \u001b[2mโ\u001b[0m\n",
- "\u001b[2mโ\u001b[0m 1. All token statistics are based on a sample of max(1000, len(dataset)) records. \u001b[2mโ\u001b[0m\n",
- "\u001b[2mโ\u001b[0m 2. Tokens are calculated using tiktoken's cl100k_base tokenizer. \u001b[2mโ\u001b[0m\n",
- "\u001b[2mโ\u001b[0m \u001b[2mโ\u001b[0m\n",
- "\u001b[2mโฐโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฏ\u001b[0m\n",
- " \n",
- "\u001b[1;38;2;118;185;0mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m\n"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"# Seedใใ็ใฎๅๆ\n",
"print(\"\\n\" + \"=\"*80)\n",
@@ -3014,30 +2710,7 @@
"execution_count": null,
"id": "269b2e97-4244-462b-824a-21a231eaac96",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Seedใใ็ใฎใใผใฟๆฐ: 7992\n",
- "\n",
- "Seedใใ็ใฎๅ่ณชในใณใขๅๅธ:\n",
- "\n",
- "clarity_score:\n",
- "clarity_score\n",
- "ๆ็ขบ 7974\n",
- "ใใไธๆ็ขบ 18\n",
- "Name: count, dtype: int64[pyarrow]\n",
- "\n",
- "difficulty:\n",
- "difficulty\n",
- "ๆใใ 6984\n",
- "ๆฎ้ 1007\n",
- "้ฃใใ 1\n",
- "Name: count, dtype: int64[pyarrow]\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"# ใใผใฟใฎ่ชญใฟ่พผใฟ\n",
"df_with_seed = job_with_seed.load_dataset()\n",
@@ -3073,30 +2746,10 @@
},
{
"cell_type": "code",
- "execution_count": 34,
+ "execution_count": null,
"id": "4d1023ea-dd88-4301-85ed-40b60e481024",
"metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "[17:52:08] [INFO] ๐บ Downloading artifacts from Job with ID 'job-2qxnztmlrfrkdkjzuxv2kd'\n",
- "[17:52:10] [INFO] โ
Artifacts downloaded to jcommonsenseqa_data_output_filter_jcommonsenseqa_seed_adjust_2000_temperature_0_9_remake_metric_8000_blog_check_en/with_seed_data\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\n",
- "ใใผใฟใ 'jcommonsenseqa_data_output_filter_jcommonsenseqa_seed_adjust_2000_temperature_0_9_remake_metric_8000_blog_check_en' ใใฃใฌใฏใใชใซไฟๅญใใพใใใ\n",
- "\n",
- "ไฟๅญใใใใใกใคใซ:\n",
- " - with_seed_data.jsonl (103554.34 KB)\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"OUTPUT_DIR = \"jcommonsenseqa_8000_filter_jcommonsenseqa_seed_2000_temperature_0_9\"\n",
"import os\n",
@@ -3138,11 +2791,19 @@
"### What We Did\n",
"1. โ
Correctly configured nvidia/Nemotron-Personas-Japan as seed data\n",
"2. โ
Generated data by directly referencing seed data columns\n",
- "3. โ
Generated synthetic data for jcommonsenseqa and commonsensemoralja\n",
- "4. โ
Created 2 versions: with and without seed data\n",
+ "3. โ
Generated synthetic data for jcommonsenseqa\n",
+ "4. โ
Created with seed data\n",
"5. โ
Quality evaluation using LLM-as-a-Judge\n",
"6. โ
Generated quality comparison report\n"
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "4e1564da-84a0-45d7-9424-6e08faca3ca1",
+ "metadata": {},
+ "outputs": [],
+ "source": []
}
],
"metadata": {