Updated README

README.md

@@ -38,9 +38,9 @@ def predict(path_to_image: str = None):
     with open(path_to_image, "rb") as i:
         image = i.read()
     payload = {
-        "inputs": {"image": image},
+        "inputs": [image],
         "parameters": {
-                   "sample": True,
+                   "do_sample": True,
                    "top_p":0.9,
                    "min_length":5,
                    "max_length":20
@@ -55,6 +55,42 @@ prediction = predict(
 )
 
 ```
+Example parameters depending on the decoding strategy:
+
+1. Beam search
+
+``` 
+        "parameters": {
+                   "num_beams":5,
+                   "max_length":20
+        }
+```
+
+2. Nucleus sampling
+
+``` 
+        "parameters": {
+                   "num_beams":1,
+                   "max_length":20,
+                   "do_sample": True,
+                   "top_k":50,
+                   "top_p":0.95
+        }
+```
+
+3. Contrastive search
+
+``` 
+        "parameters": {
+                   "penalty_alpha":0.6,
+                   "top_k":4
+                   "max_length":512
+        }
+```
+
+See [generate()](https://huggingface.co/docs/transformers/v4.25.1/en/main_classes/text_generation#transformers.GenerationMixin.generate) doc for additional detail
+
+
 expected output
 ```python
 ['buckingham palace with flower beds and red flowers']