mirror of
https://github.com/wassname/prob_jsonformer.git
synced 2026-09-10 12:37:35 +08:00
add (1) better logging (2) better string generation: dont rely on model for starting " in string (3) support T5ForConditionalGeneration
This commit is contained in:
1rgs
+147
-11
@@ -37,13 +37,37 @@
|
||||
"execution_count": 2,
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Generating...\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Generating...\n",
|
||||
"{\n",
|
||||
" temperature: \u001b[32m15.0\u001b[0m\n",
|
||||
" name: \u001b[32m\"Mickey Mouse\"\u001b[0m,\n",
|
||||
" age: \u001b[32m30.0\u001b[0m,\n",
|
||||
" description: \u001b[32m\"A mouse like object\"\u001b[0m\n",
|
||||
"}\n"
|
||||
]
|
||||
}
|
||||
@@ -52,18 +76,21 @@
|
||||
"from jsonformer.format import highlight_values\n",
|
||||
"from jsonformer.main import Jsonformer\n",
|
||||
"\n",
|
||||
"weather_schema = {\n",
|
||||
"mickey_mouse_schema = {\n",
|
||||
" \"type\": \"object\",\n",
|
||||
" \"properties\": {\n",
|
||||
" \"temperature\": {\"type\": \"number\"},\n",
|
||||
" \"name\": {\"type\": \"string\"},\n",
|
||||
" \"age\": {\"type\": \"number\"},\n",
|
||||
" \"description\": {\"type\": \"string\"},\n",
|
||||
" },\n",
|
||||
"}\n",
|
||||
"\n",
|
||||
"builder = Jsonformer(\n",
|
||||
" model=model,\n",
|
||||
" tokenizer=tokenizer,\n",
|
||||
" json_schema=weather_schema,\n",
|
||||
" prompt=\"today's temperature in celsius is\",\n",
|
||||
" json_schema=mickey_mouse_schema,\n",
|
||||
" prompt=\"write a description about mickey mouse\",\n",
|
||||
" max_string_token_length=100,\n",
|
||||
")\n",
|
||||
"\n",
|
||||
"print(\"Generating...\")\n",
|
||||
@@ -77,14 +104,38 @@
|
||||
"execution_count": 3,
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n",
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Generating...\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Generating...\n",
|
||||
"{\n",
|
||||
" make: \u001b[32m\"audi\"\u001b[0m,\n",
|
||||
" model: \u001b[32m\"model a4\"\u001b[0m,\n",
|
||||
" model: \u001b[32m\"a4\"\u001b[0m,\n",
|
||||
" year: \u001b[32m2016.0\u001b[0m,\n",
|
||||
" colors_available: [\n",
|
||||
" \u001b[32m\"blue\"\u001b[0m\n",
|
||||
@@ -122,10 +173,95 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 4,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Generating...\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n",
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n",
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n",
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n",
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n",
|
||||
"The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.\n",
|
||||
"Setting `pad_token_id` to `eos_token_id`:0 for open-end generation.\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"{\n",
|
||||
" car: {\n",
|
||||
" make: \u001b[32m\"Rolls Royce\"\u001b[0m,\n",
|
||||
" model: \u001b[32m\"Phantom\"\u001b[0m,\n",
|
||||
" year: \u001b[32m2016.0\u001b[0m,\n",
|
||||
" colors: [\n",
|
||||
" \u001b[32m\"Gold\"\u001b[0m\n",
|
||||
" ],\n",
|
||||
" features: {\n",
|
||||
" audio: {\n",
|
||||
" brand: \u001b[32m\"Mercedes\"\u001b[0m,\n",
|
||||
" speakers: \u001b[32m2.0\u001b[0m,\n",
|
||||
" hasBluetooth: \u001b[32mTrue\u001b[0m\n",
|
||||
" },\n",
|
||||
" safety: {\n",
|
||||
" airbags: \u001b[32m2.0\u001b[0m,\n",
|
||||
" parkingSensors: \u001b[32mTrue\u001b[0m,\n",
|
||||
" laneAssist: \u001b[32mTrue\u001b[0m\n",
|
||||
" },\n",
|
||||
" performance: {\n",
|
||||
" engine: \u001b[32m\"Mercedes-Benz OM615\"\u001b[0m,\n",
|
||||
" horsepower: \u001b[32m350.0\u001b[0m,\n",
|
||||
" topSpeed: \u001b[32m220.0\u001b[0m\n",
|
||||
" }\n",
|
||||
" }\n",
|
||||
" },\n",
|
||||
" owner: {\n",
|
||||
" firstName: \u001b[32m\"John\"\u001b[0m,\n",
|
||||
" lastName: \u001b[32m\"Doe\"\u001b[0m,\n",
|
||||
" age: \u001b[32m40.0\u001b[0m\n",
|
||||
" }\n",
|
||||
"}\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"complex_car = {\"type\": \"object\", \"properties\": {\"car\": {\"type\": \"object\", \"properties\": {\"make\": {\"type\": \"string\"}, \"model\": {\"type\": \"string\"}, \"year\": {\"type\": \"number\"}, \"colors\": {\"type\": \"array\", \"items\": {\"type\": \"string\"}}, \"features\": {\"type\": \"object\", \"properties\": {\"audio\": {\"type\": \"object\", \"properties\": {\"brand\": {\"type\": \"string\"}, \"speakers\": {\"type\": \"number\"}, \"hasBluetooth\": {\"type\": \"boolean\"}}}, \"safety\": {\"type\": \"object\", \"properties\": {\"airbags\": {\"type\": \"number\"}, \"parkingSensors\": {\"type\": \"boolean\"}, \"laneAssist\": {\"type\": \"boolean\"}}}, \"performance\": {\"type\": \"object\", \"properties\": {\"engine\": {\"type\": \"string\"}, \"horsepower\": {\"type\": \"number\"}, \"topSpeed\": {\"type\": \"number\"}}}}}}}, \"owner\": {\"type\": \"object\", \"properties\": {\"firstName\": {\"type\": \"string\"}, \"lastName\": {\"type\": \"string\"}, \"age\": {\"type\": \"number\"}}}}}\n",
|
||||
"builder = Jsonformer(\n",
|
||||
" model=model,\n",
|
||||
" tokenizer=tokenizer,\n",
|
||||
" json_schema=complex_car,\n",
|
||||
" prompt=\"generate an example Rolls Royce Phantom\",\n",
|
||||
")\n",
|
||||
"\n",
|
||||
"print(\"Generating...\")\n",
|
||||
"output = builder()\n",
|
||||
"\n",
|
||||
"highlight_values(output)\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
|
||||
Reference in New Issue
Block a user