Update README.md

README.md

@@ -9,22 +9,31 @@ base_model: timbrooks/instruct-pix2pix
 library_name: generic
 ---
 
-# Fork of [timbrooks/instruct-pix2pix](https://huggingface.co/timbrooks/instruct-pix2pix) for an `image-to-image` Inference endpoint.
+## timbrooks/instruct-pix2pix to deploy with Inference Endpoints
 
-This repository implements a `custom` task for `image-to-image` with instructions for 🤗 Inference Endpoints. The code for the customized 
-pipeline is in the handler.py.
+Expected payload:
 
-To use deploy this model an Inference Endpoint you have to select `Custom` as task to use the `handler.py` file. 
-
-### expected Request payload
-
-```json
-{
-  "image": encoded_image,
-  "parameters": {
-    "candidate_labels": "green, yellow, blue, white, silver"
-  }
-}
+```python
+def predict(path_to_image, prompt):
+    with open(path_to_image, "rb") as i:
+        b64 = base64.b64encode(i.read()).decode()
+    payload = {
+        "inputs": {
+            "image": b64, 
+            "prompt": prompt
+        }
+    }
+    response = r.post(
+        ENDPOINT_URL, json=payload, headers={"Content-Type": "application/json"}
+    )
+    return response.json()
 ```
 
-`encoded_image` is a base64 encoded image. 
+Call it with:
+```python
+resp = predict(
+    path_to_image="car.png", 
+    prompt="make the car green"
+)
+img = Image.open(BytesIO(base64.b64decode(resp)))
+```