update readme

README.md

@@ -24,6 +24,8 @@ xinfer is a modular Python framework that provides a unified interface for perfo
 - Ultralytics YOLO: State-of-the-art real-time object detection models.
 - Custom Models: Support for your own machine learning models and architectures.
 
+## Prerequisites
+Install [PyTorch](https://pytorch.org/get-started/locally/).
 
 ## Installation
 Install xinfer using pip:

nbs/example.ipynb

@@ -5,6 +5,14 @@
    "execution_count": 1,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/dnth/mambaforge-pypy3/envs/xinfer-test/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    },
     {
      "data": {
       "text/html": [
@@ -49,24 +57,11 @@
    "execution_count": 2,
    "metadata": {},
    "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "a59fae37d6c24825ba4afafa57451156",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
+      "Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  6.83it/s]\n",
       "Expanding inputs for image tokens in BLIP-2 should be done in processing. Please follow instruction here (https://gist.github.com/zucchini-nlp/e9f20b054fa322f84ac9311d9ab67042) to update your BLIP-2 model. Using processors without these attributes in the config is deprecated and will throw an error in v4.47.\n",
       "Expanding inputs for image tokens in BLIP-2 should be done in processing. Please follow instruction here (https://gist.github.com/zucchini-nlp/e9f20b054fa322f84ac9311d9ab67042) to update your BLIP-2 model. Using processors without these attributes in the config is deprecated and will throw an error in v4.47.\n"
      ]
@@ -140,27 +135,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "89a91a5de03f4a23952cb941a30c3025",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "Loading checkpoint shards:   0%|          | 0/8 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
+      "Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.21it/s]\n",
       "Both `max_new_tokens` (=200) and `max_length`(=51) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)\n"
      ]
     },
@@ -213,7 +195,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.15"
+   "version": "3.11.10"
   }
  },
  "nbformat": 4,

pyproject.toml

@@ -41,6 +41,7 @@ ultralytics = [
 ]
 
 
+
 [tool]
 [tool.setuptools.packages.find]
 include = ["xinfer*"]

xinfer/model_factory.py

@@ -1,20 +1,10 @@
-from enum import Enum
-
 from rich.console import Console
 from rich.table import Table
 
-from .model_registry import ModelRegistry
+from .model_registry import InputOutput, ModelRegistry
 from .transformers.blip2 import BLIP2, VLRMBlip2
 
 
-class InputOutput(Enum):
-    IMAGE_TO_TEXT = "image --> text"
-    IMAGE_TEXT_TO_TEXT = "image-text --> text"
-    TEXT_TO_TEXT = "text --> text"
-    IMAGE_TO_BBOX = "image --> bbox"
-    IMAGE_TO_CLASS = "image --> class"
-
-
 def register_models():
     ModelRegistry.register(
         "transformers",
@@ -34,6 +24,7 @@ def create_model(model_id: str, backend: str, **kwargs):
     return ModelRegistry.get_model(model_id, backend, **kwargs)
 
 
+# TODO: list by backend or wildcard
 def list_models():
     console = Console()
     table = Table(title="Available Models")

xinfer/model_registry.py

@@ -1,13 +1,22 @@
 from dataclasses import dataclass, field
+from enum import Enum
 from typing import Dict, List, Type
 
 from .base_model import BaseModel
 
 
+class InputOutput(Enum):
+    IMAGE_TO_TEXT = "image --> text"
+    IMAGE_TEXT_TO_TEXT = "image-text --> text"
+    TEXT_TO_TEXT = "text --> text"
+    IMAGE_TO_BBOX = "image --> bbox"
+    IMAGE_TO_CLASS = "image --> class"
+
+
 @dataclass
 class ModelInfo:
     model_class: Type[BaseModel]
-    input_output: str = ""
+    input_output: InputOutput
 
 
 @dataclass
@@ -25,7 +34,7 @@ def register(
         backend: str,
         model_id: str,
         model_class: Type[BaseModel],
-        input_output: str = "",
+        input_output: InputOutput,
     ):
         if backend not in cls._registry:
             cls._registry[backend] = BackendRegistry(backend_name=backend)