Upload with huggingface_hub

app.py

@@ -0,0 +1,66 @@
+import minichain
+from dataclasses import fields, dataclass, is_dataclass
+from typing import List
+from enum import Enum
+
+class ColorType(Enum):
+    RED = 1
+    GREEN = 2
+    BLUE = 3
+
+@dataclass
+class Color:
+    color: ColorType
+    object: str
+    explanation: str
+
+
+
+# class StatType(Enum):
+#     POINTS = 1
+#     REBOUNDS = 2
+#     ASSISTS = 3
+
+# @dataclass
+# class Stat:
+#     value: int
+#     stat: StatType
+    
+# @dataclass
+# class Player:
+#     player: str
+#     stats: List[Stat]
+
+class T(minichain.TypedTemplatePrompt):
+    template_file = "stats.pmpt.tpl"
+    Out = Color
+
+# print(T().show({"passage": "hello"}, '[{"player": "Harden", "stats": {"value": 10, "stat": 2}}]'))
+
+with minichain.start_chain("stats") as backend:
+    p = T(backend.OpenAI(max_tokens=512))
+    print(p({"passage": open("sixers.txt").read()}))
+
+# def enum(x):
+#     d = {e.name: e.value for e in x}
+#     # d["__enum__"] = True
+#     return d
+    
+            
+# def walk(x):
+#     print(x)
+#     if issubclass(x, Enum):
+#         return enum(x)
+#     if is_dataclass(x):
+#         return {y.name: walk(y.type) for y in fields(x)}
+#     return x.__name__
+#     # return [x for x in fields(B)]
+#             # print(x.name)
+#             # print(x.type)
+#             # if issubclass(x.type, Enum):
+#             # for e in x.type:
+#             # print(e.value)
+#             # print(e.name)
+#             # print(x)]
+
+# print(walk(B))

requirements.txt

@@ -0,0 +1,3 @@
+gradio
+git+https://github.com/srush/minichain
+manifest-ml