docs: PREINSTALL & POSTINSTALL

Author: pr-Mais

extensions/firestore-huggingface-inference-api/POSTINSTALL.md

@@ -1,17 +1,44 @@
-# See it in action
+## See it in action
 
 You can test out this extension right away!
 
-Visit the following URL:
-${function:greetTheWorld.url}
-
-# Using the extension
-
-When triggered by an HTTP request, this extension responds with the following specified greeting: "${param:GREETING} World from ${param:EXT_INSTANCE_ID}".
-
-To learn more about HTTP functions, visit the [functions documentation](https://firebase.google.com/docs/functions/http-events).
-
-<!-- We recommend keeping the following section to explain how to monitor extensions with Firebase -->
+Depending on the task you want to run, add a new Firestore document to `${param:collectionPath}`.
+
+For example, if you want to run a text classification task using the model [`distilbert-base-uncased-finetuned-sst-2-english`](https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english), add a new document to `${param:collectionPath}` with the following fields:
+```
+{
+  "inputs": "You're amazing!"
+}
+```
+
+The response will get written back to the same document.
+
+```
+{
+  "inputs": "You're amazing!",
+  "response": [
+    {
+      "label": "POSITIVE",
+      "score": 0.9998762602806091
+    },
+    {
+      "label": "NEGATIVE",
+      "score": 0.00012373875781264198
+    }
+  ]
+}
+```
+
+## Error handling
+
+If the extension encounters an error, it will write the error message to the document in `${param:collectionPath}`.
+
+```
+{
+  "text": "You're amazing!",
+  "error": "Field `inputs` must be provided and must be a string or a list of strings"
+}
+```
 
 # Monitoring
 

extensions/firestore-huggingface-inference-api/PREINSTALL.md

@@ -2,19 +2,11 @@ Use this extension to run inferences in a Firestore collection using the [Huggin
 
 When triggered by a write operation in a Firestore document, this extension performs a Natural Language Processing task on a specified field and updates the document with the results.
 
-The extension supports both using Hugging Face Hosted Inference API and using a custom model hosted on your own server.
+The extension supports both using Hugging Face [Hosted Inference API](https://huggingface.co/docs/api-inference) and [Inference Endpoints](https://huggingface.co/docs/inference-endpoints/index).
 
 ## Supported tasks
 
-This extension **currently** supports the following Natural Language Processing tasks:
-- [Fill Mask](https://huggingface.co/docs/api-inference#fill-mask)
-- [Summarization](https://huggingface.co/docs/api-inference#summarization)
-- [Question Answering](https://huggingface.co/docs/api-inference/detailed_parameters#question-answering-task)
-- [Table Question Answering task](https://huggingface.co/docs/api-inference/detailed_parameters#table-question-answering-task)
-- [Sentence Similarity task](https://huggingface.co/docs/api-inference/detailed_parameters#sentence-similarity-task)
-- [Text Classification task](https://huggingface.co/docs/api-inference/detailed_parameters#text-classification-task)
-  
-For more information about the Hugging Face inference API, see the [documentation](https://huggingface.co/docs/api-inference).
+This extension **currently** supports all Natural Language Processing tasks that are supported by HuggingFace, [click here to see the full list of NLP tasks](https://huggingface.co/docs/api-inference).
 
 # Billing
 

extensions/firestore-huggingface-inference-api/README.md

@@ -10,19 +10,11 @@
 
 When triggered by a write operation in a Firestore document, this extension performs a Natural Language Processing task on a specified field and updates the document with the results.
 
-The extension supports both using Hugging Face Hosted Inference API and using a custom model hosted on your own server.
+The extension supports both using Hugging Face [Hosted Inference API](https://huggingface.co/docs/api-inference) and [Inference Endpoints](https://huggingface.co/docs/inference-endpoints/index).
 
 ## Supported tasks
 
-This extension **currently** supports the following Natural Language Processing tasks:
-- [Fill Mask](https://huggingface.co/docs/api-inference#fill-mask)
-- [Summarization](https://huggingface.co/docs/api-inference#summarization)
-- [Question Answering](https://huggingface.co/docs/api-inference/detailed_parameters#question-answering-task)
-- [Table Question Answering task](https://huggingface.co/docs/api-inference/detailed_parameters#table-question-answering-task)
-- [Sentence Similarity task](https://huggingface.co/docs/api-inference/detailed_parameters#sentence-similarity-task)
-- [Text Classification task](https://huggingface.co/docs/api-inference/detailed_parameters#text-classification-task)
-  
-For more information about the Hugging Face inference API, see the [documentation](https://huggingface.co/docs/api-inference).
+This extension **currently** supports all Natural Language Processing tasks that are supported by HuggingFace, [click here to see the full list of NLP tasks](https://huggingface.co/docs/api-inference).
 
 # Billing