Update first try

.stignore

@@ -0,0 +1,101 @@
+.git
+# Byte-compiled / optimized / DLL files
+__pycache__
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build
+develop-eggs
+dist
+downloads
+eggs
+.eggs
+lib
+lib64
+parts
+sdist
+var
+wheels
+pip-wheel-metadata
+share/python-wheels
+*.egg-info
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+
+# Flask stuff:
+instance
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build
+
+# PyBuilder
+target
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default
+ipython_config.py
+
+# pyenv
+.python-version
+
+# celery beat schedule file
+celerybeat-schedule
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env
+venv
+ENV
+env.bak
+venv.bak
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mypy
+.mypy_cache
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre

app.py

@@ -1,15 +1,44 @@
+"""Bootstrap."""
+# pylint: disable=invalid-name
+import numpy as np
 import gradio as gr
 from model_pool import load_model_s
 from logzero import logger
 
+from gradio_cmat import gradio_cmat
+
 model_s = load_model_s()
-logger.info("model_s.encode('a'): %s", model_s.encode('a').shape)
 
-_ = """
+
+def fn(text1: str, text2: str) -> np.ndarray:
+    """Define."""
+    list1 = [elm.strip() for elm in text1.splitlines() if elm.strip()]
+    list2 = [elm.strip() for elm in text2.splitlines() if elm.strip()]
+    try:
+        res = gradio_cmat(list1, list2)
+    except Exception as e:
+        logger.error("gradio_cmat error: %s", e)
+        raise
+
+    return res
+
+
+# _ = """
 try:
-  interface = gr.Interface.load(model)
+    interface = gr.Interface.load(
+        fn,
+        [
+            gr.inputs.Textbox(
+                lines=3, default="The quick brown fox jumped over the lazy dogs."
+            ),
+            gr.inputs.Textbox(lines=3, default="The fast brown fox jumps over lazy dogs."),
+        ],
+        "numpy",
+        description="Gen corralation matrix",
+    )
 except Exception as e:
-  logger.error("gr.Interface.load(%s): %s", model, e)
-  raise
+    logger.exception("")
+    logger.error("gr.Interface.load(%s): %s", "fn", e)
+    raise
+
 interface.launch()
-# """

gradio_cmat/__init__.py

@@ -0,0 +1,5 @@
+"""Init."""
+__version__ = "0.1.0"
+from .gradio_cmat import gradio_cmat
+
+__all__ = ("gradio_cmat",)

gradio_cmat/gradio_cmat.py

@@ -0,0 +1,47 @@
+"""Calculate the corr matrix."""
+# pylint: disable=invalid-name
+from typing import List, Optional
+
+import numpy as np
+from logzero import logger
+from model_pool import load_model_s
+
+model_s = load_model_s()
+
+
+def gradio_cmat(
+    list1: List[str],
+    list2_: Optional[List[str]] = None,
+) -> np.ndarray:
+    """Gen corr matrix given two lists of str.
+
+    Args:
+        list1: list of strings
+        list2_: list of strings, if None, set to list1
+    Returns:
+        numpy.ndarray, (len(list1)xlen(list2))
+    """
+    if not list2_:
+        list2 = list1[:]
+    else:
+        list2 = list2_[:]
+
+    try:
+        vec1 = model_s.encode(list1)
+    except Exception as e:
+        logger.error("mode_s.encode(list1) error: %s", e)
+        raise
+
+    try:
+        vec2 = model_s.encode(list2)
+    except Exception as e:
+        logger.error("mode_s.encode(list2) error: %s", e)
+        raise
+
+    try:
+        res = vec1.dot(vec2.T)
+    except Exception as e:
+        logger.error("vec1.dot(vec2.T) error: %s", e)
+        raise
+
+    return res

okteto.yml

@@ -0,0 +1,42 @@
+name: gradio-cmat
+
+# The build section defines how to build the images of
+# your development environment
+# More info: https://www.okteto.com/docs/reference/manifest/#build
+# build:
+#   my-service:
+#     context: .
+
+# The deploy section defines how to deploy your development environment
+# More info: https://www.okteto.com/docs/reference/manifest/#deploy
+# deploy:
+  # commands:
+    # - name: Deploy
+      # command: echo 'Replace this line with the proper 'helm'
+
+#  or 'kubectl' commands to deploy your development environment'
+
+# The dependencies section defines other git repositories to be
+# deployed as part of your development environment
+# More info: https://www.okteto.com/docs/reference/manifest/#dependencies
+# dependencies:
+#   - https://github.com/okteto/sample
+# The dev section defines how to activate a development container
+# More info: https://www.okteto.com/docs/reference/manifest/#dev
+dev:
+  gradio-cmat:
+    # image: okteto/dev:latest
+    # image: python:3.8.13-bullseye
+    # image: simbachain/poetry-3.8
+    image: python:3.8
+    command: bash
+    workdir: /usr/src/app
+    sync:
+      - .:/usr/src/app
+    environment:
+      - name=$USER
+    forward:
+      - 8080:80
+    reverse:
+      - 9000:9000
+    autocreate: true

poetry.toml

@@ -0,0 +1,3 @@
+[virtualenvs]
+create = true
+in-project = false

pyproject.toml

@@ -0,0 +1,19 @@
+[tool.poetry]
+name = "gradio-cmat"
+version = "0.1.0"
+description = "Gen correlation matrix via gradio (mainly for hf spaces)"
+authors = ["mikeee"]
+license = "MIT"
+
+[tool.poetry.dependencies]
+python = "^3.8"
+logzero = "^1.7.0"
+model-pool = "^0.1.3"
+gradio = "^2.9.4"
+
+[tool.poetry.dev-dependencies]
+pytest = "^7.1.2"
+
+[build-system]
+requires = ["poetry-core>=1.0.0"]
+build-backend = "poetry.core.masonry.api"

requirements.txt

@@ -1,2 +1,40 @@
-logzero
-model-pool
+about-time==3.1.1; python_version >= "3.7" and python_version < "4" and python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+alive-progress==2.4.1; python_version >= "3.7" and python_version < "4" and python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+asttokens==2.0.5; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+certifi==2021.10.8; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+charset-normalizer==2.0.12; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3"
+click==8.1.3; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+colorama==0.4.4; python_full_version >= "3.7.9" and sys_platform == "win32" and python_full_version < "4.0.0" and platform_system == "Windows" and python_version >= "3.7"
+executing==0.8.3; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+filelock==3.6.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+grapheme==0.6.0; python_version >= "3.7" and python_version < "4" and python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+huggingface-hub==0.4.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+icecream==2.1.2; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+idna==3.3; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.5"
+install==1.3.5; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.5"
+joblib==1.1.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+logzero==1.7.0
+model-pool==0.1.3; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+nltk==3.7; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+numpy==1.22.3; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.8"
+packaging==21.3; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.6"
+pillow==9.1.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+pygments==2.12.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.6"
+pyparsing==3.0.8; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.6"
+pyyaml==6.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.6"
+regex==2022.4.24; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+requests==2.27.1; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+sacremoses==0.0.53; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+scikit-learn==1.0.2; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+scipy==1.6.1; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+sentence-transformers==2.2.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+sentencepiece==0.1.96; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+six==1.16.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+threadpoolctl==3.1.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+tokenizers==0.12.1; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+torch==1.11.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+torchvision==0.12.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+tqdm==4.64.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+transformers==4.18.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0"
+typing-extensions==4.2.0; python_full_version >= "3.7.9" and python_full_version < "4.0.0" and python_version >= "3.7"
+urllib3==1.26.9; python_full_version >= "3.7.9" and python_version < "4" and python_full_version < "4.0.0"

tests/test_gradio_cmat.py

@@ -0,0 +1,16 @@
+"""Test gradio_cmat."""
+from gradio_cmat import gradio_cmat
+
+
+def test_gradio_cmat():
+    """Test gradio_cmat."""
+    list1 = ["test", "测试", "love"]
+    list2 = ["this is a test"]
+    cmat = gradio_cmat(list1, list2)
+    assert cmat.shape == (3, 1)
+
+    assert cmat[0, 0] > 0.6
+    assert cmat[1, 0] > 0.6
+    assert cmat[2, 0] < 0.2
+
+    # cmat.round(2)

tests/test_version.py

@@ -0,0 +1,6 @@
+"""Test __version__/sanity."""
+from gradio_cmat import __version__
+
+def test_version():
+    """Test version."""
+    assert __version__[:3] == "0.1"