Rename Predictor to BasePredictor

Signed-off-by: Ben Firshman <[email protected]>

Author: bfirsh

README.md

@@ -27,10 +27,10 @@ predict: "predict.py:Predictor"
 And define how predictions are run on your model with `predict.py`:
 
 ```python
-from cog import Predictor, Input, Path
+from cog import BasePredictor, Input, Path
 import torch
 
-class ColorizationPredictor(Predictor):
+class Predictor(BasePredictor):
     def setup(self):
         """Load the model into memory to make running multiple predictions efficient"""
         self.model = torch.load("./weights.pth")

docs/getting-started-own-model.md

@@ -66,11 +66,10 @@ With `cog.yaml`, you can also install system packages and other things. [Take a
 The next step is to update `predict.py` to define the interface for running predictions on your model. The `predict.py` generated by `cog init` looks something like this:
 
 ```python
-import cog
-from cog import Path, Input
+from cog import BasePredictor, Path, Input
 import torch
 
-class Predictor(cog.Predictor):
+class Predictor(BasePredictor):
     def setup(self):
         """Load the model into memory to make running multiple predictions efficient"""
         self.net = torch.load("weights.pth")

docs/getting-started.md

@@ -60,15 +60,14 @@ Then, we need to write some code to describe how predictions are run on the mode
 
 ```python
 from typing import Any
-import cog
-from cog import Input, Path
+from cog import BasePredictor, Input, Path
 from tensorflow.keras.applications.resnet50 import ResNet50
 from tensorflow.keras.preprocessing import image as keras_image
 from tensorflow.keras.applications.resnet50 import preprocess_input, decode_predictions
 import numpy as np
 
 
-class ResNetPredictor(cog.Predictor):
+class ResNetPredictor(BasePredictor):
     def setup(self):
         """Load the model into memory to make running multiple predictions efficient"""
         self.model = ResNet50(weights='resnet50_weights_tf_dim_ordering_tf_kernels.h5')

docs/python.md

@@ -1,13 +1,12 @@
 # Prediction interface reference
 
-You define how Cog runs predictions on your model by defining a class that inherits from `cog.Predictor`. It looks something like this:
+You define how Cog runs predictions on your model by defining a class that inherits from `BasePredictor`. It looks something like this:
 
 ```python
-import cog
-from cog import Path, Input
+from cog import BasePredictor, Path, Input
 import torch
 
-class ImageScalingPredictor(cog.Predictor):
+class Predictor(BasePredictor):
     def setup(self):
         """Load the model into memory to make running multiple predictions efficient"""
         self.model = torch.load("weights.pth")

docs/yaml.md

@@ -12,7 +12,7 @@ build:
   system_packages:
     - "ffmpeg"
     - "libavcodec-dev"
-predict: "predict.py:JazzSoloComposerPredictor"
+predict: "predict.py:Predictor"
 ```
 
 Tip: Run [`cog init`](getting-started-own-model#initialization) to generate an annotated `cog.yaml` file that can be used as a starting point for setting up your model.
@@ -102,12 +102,12 @@ If you don't provide this, a name will be generated from the directory name.
 
 ## `predict`
 
-The pointer to the `cog.Predictor` object in your code, which defines how predictions are run on your model.
+The pointer to the `Predictor` object in your code, which defines how predictions are run on your model.
 
 For example:
 
 ```yaml
-predict: "predict.py:HotdogPredictor"
+predict: "predict.py:Predictor"
 ```
 
 See [the Python API documentation for more information](python.md).

pkg/config/config.go

@@ -122,7 +122,7 @@ func (c *Config) ValidateAndCompleteConfig() error {
 	}
 	if c.Predict != "" {
 		if len(strings.Split(c.Predict, ".py:")) != 2 {
-			return fmt.Errorf("'predict' in cog.yaml must be in the form 'predict.py:PredictorClass")
+			return fmt.Errorf("'predict' in cog.yaml must be in the form 'predict.py:Predictor")
 		}
 	}
 

python/cog/__init__.py

@@ -1,8 +1,11 @@
-from .predictor import Predictor
+from .predictor import BasePredictor
 from .types import File, Input, Path
 
+# Backwards compatibility. Will be deprecated before 1.0.0.
+Predictor = BasePredictor
 
 __all__ = [
+    "BasePredictor",
     "File",
     "Input",
     "Path",

python/cog/predictor.py

@@ -17,7 +17,7 @@
 from .types import Input
 
 
-class Predictor(ABC):
+class BasePredictor(ABC):
     def setup(self):
         pass
 
@@ -63,7 +63,7 @@ def load_predictor():
     return predictor_class()
 
 
-def get_input_type(predictor: Predictor):
+def get_input_type(predictor: BasePredictor):
     signature = inspect.signature(predictor.predict)
     create_model_kwargs = {}
 
@@ -106,7 +106,7 @@ def get_input_type(predictor: Predictor):
     return create_model("Input", **create_model_kwargs)
 
 
-def get_output_type(predictor: Predictor):
+def get_output_type(predictor: BasePredictor):
     signature = inspect.signature(predictor.predict)
     if signature.return_annotation is inspect.Signature.empty:
         OutputType = Literal[None]

python/cog/server/http.py

@@ -10,13 +10,13 @@
 
 from ..files import upload_file
 from ..json import encode_json
-from ..predictor import Predictor, get_input_type, get_output_type, load_predictor
+from ..predictor import BasePredictor, get_input_type, get_output_type, load_predictor
 from ..response import Status, get_response_type
 
 logger = logging.getLogger("cog")
 
 
-def create_app(predictor: Predictor) -> FastAPI:
+def create_app(predictor: BasePredictor) -> FastAPI:
     app = FastAPI(
         title="Cog",  # TODO: mention model name?
         # version=None # TODO

python/cog/server/redis_queue.py

@@ -14,9 +14,8 @@
 import requests
 
 from .redis_log_capture import capture_log
-from ..predictor import Predictor, get_input_type, load_predictor
+from ..predictor import BasePredictor, get_input_type, load_predictor
 from ..json import encode_json
-from ..predictor import Predictor, load_predictor
 
 
 class timeout:
@@ -53,7 +52,7 @@ class RedisQueueWorker:
 
     def __init__(
         self,
-        predictor: Predictor,
+        predictor: BasePredictor,
         redis_host: str,
         redis_port: int,
         input_queue: str,