Improve spelling

datasets/anomaly/README.md

@@ -1,4 +1,4 @@
-## Anomolies
+## Anomalies
 
-This dataset is used for classifying area summaries as anomolies vs classifying
+This dataset is used for classifying area summaries as anomalies vs classifying
 as operating normally

datasets/assist-mini/README.md

@@ -1,13 +1,13 @@
 # Mini Evaluation dataset for Home Assistant assist actions
 
-This is a variation on the `assist` dataset that tests far fewer entites, devices,
+This is a variation on the `assist` dataset that tests far fewer entities, devices,
 and areas at once. This is designed given poor performance of local models on the
 `assist` baseline to help with even smaller tasks to get right.
 
 This is a dataset for the Home Assistant LLM API ([blog post](https://developers.home-assistant.io/blog/2024/05/20/llm-api/)).
 
 See the `home-assistant-datasets assist` command for more details on how to
-run evalations.
+run evaluations.
 
 See [../assist/README.md](../assist/README.md) for details on how the dataset
 is configured including the inventory fixtures, and the eval tasks.

datasets/assist-mini/home1-us-vacuum/vacuum.yaml

@@ -2,7 +2,7 @@
 category: vacuum
 tests:
   - sentences:
-      - Start vaccuum in the living room
+      - Start vacuum in the living room
       - Start Roborock Downstairs vacuum
     setup:
       vacuum.roborock_downstairs:
@@ -12,7 +12,7 @@ tests:
         state: cleaning
   - sentences:
       - Return Roborock Downstairs to base
-      - Return vaccuum in the living room to base
+      - Return vacuum in the living room to base
     setup:
       vacuum.roborock_downstairs:
         state: cleaning

datasets/assist/README.md

@@ -3,7 +3,7 @@
 This is a dataset for the Home Assistant LLM API ([blog post](https://developers.home-assistant.io/blog/2024/05/20/llm-api/)).
 
 See the `home-assistant-datasets assist` command for more details on how to
-run evalations.
+run evaluations.
 
 ## Dataset details
 
@@ -116,13 +116,14 @@ $ pip3 install -e /workspaces/core
 ```
 
 You will need the [synthetic-home custom component](https://github.com/allenporter/synthetic-home)
-and you can either install it in a separate direcotry like this:
+and you can either install it in a separate directory like this:
+
 ```bash
 $ export PYTHONPATH="${PYTHONPATH}:/workspaces/home-assistant-synthetic-home/"
 ```
 
 Or using a `custom_components` directory in the local directory if you have multiple
-custom components you wnat to evaluate:
+custom components you want to evaluate:
 
 ```bash
 $ export PYTHONPATH="${PYTHONPATH}:${PWD}"

datasets/intents/README.md

@@ -3,7 +3,7 @@
 This is a dataset for the Home Assistant LLM API ([blog post](https://developers.home-assistant.io/blog/2024/05/20/llm-api/)) imported from the home assistant intents repo.
 
 See the `home-assistant-datasets assist` command for more details on how to
-run evalations.
+run evaluations.
 
 ## Dataset details
 

generation/README.md

@@ -3,6 +3,7 @@
 This directory contains notebooks and seeds for generating the datasets.
 
 The phases of generating data are currently:
+
 - Generate a description of a Home ([notebook](homes.ipynb))
 - Generate the Areas of the Home useful for automation ([notebook](areas.ipynb))
 - Generate the devices in each area ([notebook](devices.ipynb))
@@ -20,17 +21,17 @@ The data is generated in a format that can be consumed by https://github.com/all
 
 The next phase needed is to generate useful state data for triggering those actions.
 
-### Notebook Development enviroment
+### Notebook Development environment
 
 This is an example of how to prepare the python venv for use with the jupyet notebooks:
 
 1. Create the virtual environment
 
-    ```bash
-    $ python3 -m venv venv
-    $ source venv/bin/activate
-    $ pip install --upgrade pip
-    $ pip3 install -r requirements_dev.txt
-    ```
+   ```bash
+   $ python3 -m venv venv
+   $ source venv/bin/activate
+   $ pip install --upgrade pip
+   $ pip3 install -r requirements_dev.txt
+   ```
 
 1. Open the notebook and select the venv interpreter and kernel

generation/archive/area-summary.ipynb

@@ -172,7 +172,7 @@
     "which will be fed as input into other agents. The user will feed in details about\n",
     "areas and devices in the home, and you will respond with a summary of the status of the area.\n",
     "\n",
-    "Your summaries are succint, and do not mention boring details or things that seem\n",
+    "Your summaries are succinct, and do not mention boring details or things that seem\n",
     "very mundane or minor. A one sentence summary is best.\n",
     "\n",
     "\n",
@@ -240,7 +240,7 @@
     "        if unit_of_measurement := attributes.get('unit_of_measurement'):\n",
     "            state_value = f\"{state_value} {unit_of_measurement}\"\n",
     "\n",
-    "        data.append(f\"    {friendly_name} ({entity_name}): {state_value}\")    \n",
+    "        data.append(f\"    {friendly_name} ({entity_name}): {state_value}\")\n",
     "    return data\n",
     "\n",
     "for area in home.areas:\n",
@@ -283,7 +283,7 @@
    "source": [
     "OVERALL_PROMPT = f\"\"\"\n",
     "You are an agent running in Home Assistant that is the overall house summary\n",
-    "agent. Other agents have summarized invidivual areas, and you are responsible\n",
+    "agent. Other agents have summarized individual areas, and you are responsible\n",
     "for picking out the important details, ignoring the mundane. You write succinct,\n",
     "1 sentence summaries of the entire house that can be displayed on a dashboard.\n",
     "You don't need to comment on every area, and can omit any that are mundane.\n",
@@ -323,7 +323,7 @@
     "- Attic: The attic temperature is high.\n",
     "- Front Door The front door is being used and the porch is dimmed.\n",
     "\n",
-    "Summary: \n",
+    "Summary:\n",
     "\"\"\"\n",
     "\n",
     "response = model.complete(OVERALL_PROMPT, message)\n",

generation/devices.ipynb

@@ -170,7 +170,7 @@
       "- Balcony\n",
       "\n",
       "Output:\n",
-      "# thought: I will genearate a set of devices commonly found in a New York City Apartment\n",
+      "# thought: I will generate a set of devices commonly found in a New York City Apartment\n",
       "# thought: Given the compact nature of the living space, the devices may focus on essential functionalities and convenience for the resident\n",
       "Living Room:\n",
       "- name: Lamp\n",
@@ -310,15 +310,15 @@
     "You are an expert in smart home automation and are generating data used to\n",
     "evaluate the performance of a smart home system Home Assistant on tasks like\n",
     "summarization, performing actions, or being an independent agent managing\n",
-    "automations and maintenace tasks.\n",
+    "automations and maintenance tasks.\n",
     "\n",
     "You use your knowledge about the world to generate details about homes that\n",
     "can be used for synthetic smart home automation data. For example, an apartment\n",
     "may have a smart thermostat, a house may have a smart garage door opener or\n",
     "smart lock and camera, and all houses may have a smart light or weather feed\n",
     "air quality, or a smart speaker or television. The needs of a home owner\n",
     "may vary if they are a single person or family, or where in the world they\n",
-    "live. For example, a high rise aparment probably does not have a backyard.\n",
+    "live. For example, a high rise apartment probably does not have a backyard.\n",
     "\n",
     "Do not be cliche. Do not assume everyone has a smart watch. Don't assume\n",
     "every room has a smart light just because it is listed, but of course many will.\n",

generation/seeds/devices.yaml

@@ -2,11 +2,11 @@
 home:
   name: Coastal Bungalow
   thoughts:
-  - The bungalow in a seaside village may be a vacation home or a tranquil retreat.
-  - The cozy living room with a fireplace hints at a cozy atmosphere, and there may
-    be smart climate control for comfort.
-  - The outdoor shower is convenient for beach days, so there may be smart water heating
-    systems.
+    - The bungalow in a seaside village may be a vacation home or a tranquil retreat.
+    - The cozy living room with a fireplace hints at a cozy atmosphere, and there may
+      be smart climate control for comfort.
+    - The outdoor shower is convenient for beach days, so there may be smart water heating
+      systems.
   desc: Bungalow in a seaside village in Norway
   areas:
     - Living Room
@@ -20,7 +20,7 @@ devices:
     - name: Living Room
       device_type: light-dimmable
       device_info:
-        model: Enbrighten Z-Wave 
+        model: Enbrighten Z-Wave
         manufacturer: GE
         sw_version: 2.3.29-7
     - name: Thermostat
@@ -42,7 +42,7 @@ devices:
     - name: Port light
       device_type: light
       device_info:
-        model: Enbrighten Z-Wave 
+        model: Enbrighten Z-Wave
         manufacturer: GE
         sw_version: 2.3.29-7
   Outdoor Shower:
@@ -57,17 +57,17 @@ devices:
 home:
   name: Home2
   thoughts:
-  - The apartment with city views may cater to a single person or a couple.
-  - The shared rooftop terrace could have restrictions on device installations.
-  - In-unit laundry suggests convenience, so there may be a smart washer and dryer.
+    - The apartment with city views may cater to a single person or a couple.
+    - The shared rooftop terrace could have restrictions on device installations.
+    - In-unit laundry suggests convenience, so there may be a smart washer and dryer.
   desc: Apartment in urban area in New York City, USA
   areas:
     - Living Room
     - Kitchen
     - Bedroom
     - Balcony
 devices-thoughts:
-  - I will genearate a set of devices commonly found in a New York City Apartment
+  - I will generate a set of devices commonly found in a New York City Apartment
   - Given the compact nature of the living space, the devices may focus on essential functionalities and convenience for the resident
 devices:
   Living Room:
@@ -113,12 +113,12 @@ devices:
 home:
   name: Loft Industrial
   thoughts:
-  - The industrial loft with an open space and high ceilings may have a modern and
-    minimalist design.
-  - The panoramic windows offer plenty of natural light, ideal for a smart lighting
-    system.
-  - The rooftop terrace could be a great spot for outdoor entertainment, so there
-    may be smart speakers or a sound system.
+    - The industrial loft with an open space and high ceilings may have a modern and
+      minimalist design.
+    - The panoramic windows offer plenty of natural light, ideal for a smart lighting
+      system.
+    - The rooftop terrace could be a great spot for outdoor entertainment, so there
+      may be smart speakers or a sound system.
   desc: Industrial loft in Barcelona, Spain
   areas:
     - Living Space

home_assistant_datasets/tool/archive_evals/anomaly/output/gemma-3-shot.yaml

@@ -313,7 +313,7 @@ response: anomaly
 uuid: 1e4a7577-5d18-4dc5-b292-92ac333efa93
 task_id: d23ea7aac86432d2b3137e86965178ff96336f895742bdfa6dc627f67073dde5-anomaly
 task:
-  text: The smart plugs connected to the refridgerator is currently off.
+  text: The smart plugs connected to the refrigerator is currently off.
   label: anomaly
 response: anomaly
 ---

home_assistant_datasets/tool/archive_evals/anomaly/output/gpt-3.5-10-shot.yaml

@@ -2,7 +2,7 @@
 uuid: 1845a3d9-359e-4043-bbde-593687853c79
 task_id: d23ea7aac86432d2b3137e86965178ff96336f895742bdfa6dc627f67073dde5-anomaly
 task:
-  text: The smart plugs connected to the refridgerator is currently off.
+  text: The smart plugs connected to the refrigerator is currently off.
   label: anomaly
 response: anomaly
 ---

home_assistant_datasets/tool/archive_evals/anomaly/output/llama3-10-shot.yaml

@@ -273,7 +273,7 @@ response: normal
 uuid: eb56c6db-7127-491c-aa9c-eafc23e830e4
 task_id: d23ea7aac86432d2b3137e86965178ff96336f895742bdfa6dc627f67073dde5-anomaly
 task:
-  text: The smart plugs connected to the refridgerator is currently off.
+  text: The smart plugs connected to the refrigerator is currently off.
   label: anomaly
 response: anomaly
 ---

home_assistant_datasets/tool/archive_evals/anomaly/output/llama3-zero-shot.yaml

@@ -48,7 +48,7 @@ response: normal
 uuid: 383a566a-f7d6-4694-976c-9a0717432ed2
 task_id: d23ea7aac86432d2b3137e86965178ff96336f895742bdfa6dc627f67073dde5-anomaly
 task:
-  text: The smart plugs connected to the refridgerator is currently off.
+  text: The smart plugs connected to the refrigerator is currently off.
   label: anomaly
 response: anomaly
 ---

home_assistant_datasets/tool/archive_evals/anomaly/output/mistral-7b-instruct-10-shot.yaml

@@ -540,7 +540,7 @@ response: normal
 uuid: 5e37847c-31ab-4de2-85c3-1ae45c2b2dbe
 task_id: d23ea7aac86432d2b3137e86965178ff96336f895742bdfa6dc627f67073dde5-anomaly
 task:
-  text: The smart plugs connected to the refridgerator is currently off.
+  text: The smart plugs connected to the refrigerator is currently off.
   label: anomaly
 response: normal
 ---

home_assistant_datasets/tool/archive_evals/anomaly/output/mistral-7b-instruct-zero-shot.yaml

@@ -46,7 +46,7 @@ response: anomaly
 uuid: 70a75c30-e7c6-4a9b-b5a5-a0c78afa9b22
 task_id: d23ea7aac86432d2b3137e86965178ff96336f895742bdfa6dc627f67073dde5-anomaly
 task:
-  text: The smart plugs connected to the refridgerator is currently off.
+  text: The smart plugs connected to the refrigerator is currently off.
   label: anomaly
 response: anomaly
 ---

home_assistant_datasets/tool/archive_evals/anomaly/test_anomaly_eval.py

@@ -58,7 +58,7 @@
     ],
 )
 def model_id_fixture(request: pytest.FixtureRequest) -> str:
-    """Fiture that defines which model is being evaluated."""
+    """Fixture that defines which model is being evaluated."""
     return request.param
 
 

home_assistant_datasets/tool/archive_evals/archive/area_summary/eval.py

@@ -17,7 +17,7 @@
 
 
 AREA_SUMMARY_PROMPT = """
-Please summarize the status of an area of the home. Your summaries are succint,
+Please summarize the status of an area of the home. Your summaries are succinct,
 and do not mention boring details or things that seem very mundane or minor. A
 one sentence summary is best.
 
@@ -52,7 +52,6 @@
 """
 
 
-
 def make_prompt(area_name: str) -> str:
     """Create a prompt for the agent to summarize the area."""
     return AREA_PROMPT.format(
@@ -81,25 +80,32 @@ async def async_process(self, hass: HomeAssistant, text: str) -> str:
         return response["speech"]["plain"]["speech"]
 
 
-
 async def async_run_eval(hass: HomeAssistant, config: dict[str, Any]) -> None:
     """Run the evaluation."""
 
     eval_dir = pathlib.Path(config["eval_dir"])
     os.makedirs(eval_dir, exist_ok=True)
 
     # Find the conversation agent id to evaluate based on the domain in the config file
-    config_entry_ids = [ entry.entry_id for entry in hass.config_entries.async_entries() if entry.domain in config["conversation_agent_domain"] ]
+    config_entry_ids = [
+        entry.entry_id
+        for entry in hass.config_entries.async_entries()
+        if entry.domain in config["conversation_agent_domain"]
+    ]
     if len(config_entry_ids) > 1:
-        raise ValueError("Found more than one config entry for domain {config['convesation_agent_domain']}")
+        raise ValueError(
+            "Found more than one config entry for domain {config['convesation_agent_domain']}"
+        )
     if not config_entry_ids:
-        raise ValueError("Found no config entry for domain {config['convesation_agent_domain']}")
+        raise ValueError(
+            "Found no config entry for domain {config['convesation_agent_domain']}"
+        )
     conversation_agent_id = config_entry_ids[0]
 
     agent = ConversationAgent(conversation_agent_id)
 
     area_registry = ar.async_get(hass)
-    area_names = [ area.name for area in area_registry.async_list_areas() ]
+    area_names = [area.name for area in area_registry.async_list_areas()]
 
     _LOGGER.info("Loaded %s areas to evaluate", len(area_names))