Inference endpoint

README.md

@@ -100,7 +100,7 @@ language:
 license: cc-by-nc-sa-4.0
 ---
 
-This is a [COMET](https://github.com/Unbabel/COMET) quality estimation model: It receives a source sentence and the respective translation and returns a score that reflects the quality of the translation.
+This is a [COMET](https://github.com/Unbabel/COMET) quality estimation model by Unbabel: It receives a source sentence and the respective translation and returns a score that reflects the quality of the translation.
 
 # Paper
 
@@ -110,49 +110,61 @@ This is a [COMET](https://github.com/Unbabel/COMET) quality estimation model: It
 
 cc-by-nc-sa-4.0
 
-# Usage (unbabel-comet)
-
-Using this model requires unbabel-comet to be installed:
-
-```bash
-pip install --upgrade pip  # ensures that pip is current 
-pip install unbabel-comet
-```
-
-Then you can use it through comet CLI:
-
-```bash
-comet-score -s {source-input}.txt -t {translation-output}.txt --model Unbabel/wmt22-cometkiwi-da
-```
-
-Or using Python:
+# Usage for Inference Endpoint
 
 ```python
-from comet import download_model, load_from_checkpoint
-
-model_path = download_model("Unbabel/wmt22-cometkiwi-da")
-model = load_from_checkpoint(model_path)
-data = [
-    {
-        "src": "The output signal provides constant sync so the display never glitches.",
-        "mt": "Das Ausgangssignal bietet eine konstante Synchronisation, so dass die Anzeige nie stört."
-    },
-    {
-        "src": "Kroužek ilustrace je určen všem milovníkům umění ve věku od 10 do 15 let.",
-        "mt": "Кільце ілюстрації призначене для всіх любителів мистецтва у віці від 10 до 15 років."
-    },
-    {
-        "src": "Mandela then became South Africa's first black president after his African National Congress party won the 1994 election.",
-        "mt": "その後、1994年の選挙でアフリカ国民会議派が勝利し、南アフリカ初の黒人大統領となった。"
+import json
+import requests
+
+API_URL = ""
+API_TOKEN="MY_API_KEY"
+headers = {
+    "Authorization": f"Bearer {API_TOKEN}",
+    "Content-Type": "application/json",
+}
+
+def query(url, headers, payload):
+    data = json.dumps(payload)
+    response = requests.request("POST", url, headers=headers, data=data)
+    return json.loads(response.content.decode("utf-8"))
+
+payload = {
+    "inputs": {
+        "batch_size": 8,
+        "workers": None,
+        "data": [
+            {
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },{
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },{
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },{
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },{
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },{
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },{
+                "src": "Youll be picking fruit and generally helping us do all the usual farm work",
+                "mt": "당신은 과일을 따기도 하고 대체로 우리가 하는 일상적인 농장 일을 돕게 될 겁니다",
+            },
+        ]
     }
-]
-model_output = model.predict(data, batch_size=8, gpus=1)
-print (model_output)
+}
+
+scores = query(API_URL, headers, payload)
 ```
 
 # Intended uses
 
-Our model is intented to be used for **reference-free MT evaluation**. 
+Unbabel's model is intented to be used for **reference-free MT evaluation**. 
 
 Given a source text and its translation, outputs a single score between 0 and 1 where 1 represents a perfect translation. 
 

handler.py

@@ -0,0 +1,27 @@
+from typing import Dict, List, Any
+from comet import load_from_checkpoint
+
+
+class EndpointHandler():
+    def __init__(self, path=""):
+        self.model = load_from_checkpoint("checkpoints/model.ckpt")
+
+    def __call__(self, data: Dict[str, Any]) -> List[Any]:
+        """
+        data args:
+            inputs (:obj: `dict[str, str]`)
+        Return:
+            A :obj: `dict`: will be serialized and returned
+        """
+
+        inputs = data.pop("inputs")
+
+        batch_size = inputs.pop("batch_size")
+        workers = inputs.pop("workers")
+
+        data = inputs.pop("data")
+        
+        model_output = self.model.predict(data, batch_size=batch_size, num_workers=workers, gpus=0)
+        scores = model_output["scores"]
+
+        return scores

requirements.txt

@@ -0,0 +1 @@
+git+https://github.com/Unbabel/COMET.git