diff --git a/.gitlab-ci.yml.bak b/.gitlab-ci.backup.yml similarity index 100% rename from .gitlab-ci.yml.bak rename to .gitlab-ci.backup.yml diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index d2c7989..1c16d00 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -1,12 +1,17 @@ include: - project: "Gitlab/gitlab" - ref: main - file: "/ci-templates/research/versioning-build+azure_model-test-release.gitlab-ci.yml" + ref: 0.3.0 + file: "/ci-templates/research/dvc-versioning-build-release.gitlab-ci.yml" variables: - NEXUS_PROJECT_DIR: ff + NEXUS_PROJECT_DIR: red IMAGENAME: "${CI_PROJECT_NAME}" +################################# +# temp. disable integration tests, b/c they don't cover the CV analysis case yet +trigger integration tests: + rules: + - when: never ###### # DOCS @@ -15,6 +20,7 @@ pages: before_script: - !reference [.activate-venv, script] script: + - cp ./README.md ./docs/source/README.md && cp -r ./data ./docs/source/data/ - sphinx-apidoc ./src -o ./docs/source/modules --no-toc --module-first --follow-links --separate --force - sphinx-build -b html docs/source/ public/ -E -a artifacts: @@ -30,51 +36,18 @@ pages: # MAKE SURE TO SWITCH OUT ALL YOUR MODEL NAMES + VERSIONS # name the job after the model it's using in the build, keep the prefix referencing `docker-build::` -docker-build::keyword-extraction-multi: - extends: .docker-build - variables: - MODEL_NAME: ${MODEL_MULTI_NAME} - MODEL_VERSION: ${MODEL_MULTI_VERSION} - -docker-build::cv-analysis-service: - extends: .docker-build +release build: + stage: release needs: - - !reference [.needs-versioning, needs] - - job: docker-build::cv-analysis-service - variables: - MODEL_NAME: ${MODEL_EN_NAME} - MODEL_VERSION: ${MODEL_EN_VERSION} - -docker-build::keyword-extraction-de: - extends: .docker-build - needs: - - !reference [.needs-versioning, needs] - - job: docker-build::cv-analysis-service - variables: - MODEL_NAME: ${MODEL_DE_NAME} - MODEL_VERSION: ${MODEL_DE_VERSION} - - -################### -# INTEGRATION TESTS -trigger-integration-tests: - extends: .integration-tests - needs: - ###### UPDATE/EDIT ###### - # YOU NEED ONLY TO DEFINE ONE - # reason is that we want to have one built image to use with the integration tests - # this should be the same image you uploaded test data for - - job: docker-build::cv-analysis-service - artifacts: true - rules: - - when: never # temp. disable integration tests - -######### -# RELEASE -release: - extends: .release - needs: - - !reference [.release, needs] # LEAVE THIS LINE AS IS - ###### UPDATE/EDIT ###### - # DEFINE ONE BUILD JOB THAT NEEDS TO EXIST BEFORE RELEASE - - job: docker-build::cv-analysis-service + - job: set custom version + artifacts: true + optional: true + - job: calculate patch version + artifacts: true + optional: true + - job: calculate minor version + artifacts: true + optional: true + - job: build docker nexus + artifacts: true +################################# diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index 1155696..3c633ba 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -10,7 +10,7 @@ repos: - id: trailing-whitespace - id: end-of-file-fixer - id: check-yaml - args: [--unsafe] + args: [--unsafe] # needed for .gitlab-ci.yml - id: check-toml - id: detect-private-key - id: check-added-large-files diff --git a/Makefile b/Makefile index 7498375..c989ab2 100644 --- a/Makefile +++ b/Makefile @@ -9,8 +9,8 @@ export DOCKER=docker export DOCKERFILE=Dockerfile -export IMAGE_NAME=keyword_extraction_service-image -export CONTAINER_NAME=keyword_extraction_service-container +export IMAGE_NAME=cv_analysis_service-image +export CONTAINER_NAME=cv_analysis_service-container export HOST_PORT=9999 export CONTAINER_PORT=9999 export PYTHON_VERSION=python3.10 @@ -88,4 +88,4 @@ sphinx_html: poetry run sphinx-build -b html docs/source/ docs/build/html -E -a sphinx_apidoc: - poetry run sphinx-apidoc ./src -o ./docs/source/modules --no-toc --module-first --follow-links --separate --force + cp ./README.md ./docs/source/README.md && cp -r ./data ./docs/source/data/ && poetry run sphinx-apidoc ./src -o ./docs/source/modules --no-toc --module-first --follow-links --separate --force diff --git a/docs/build/html/.doctrees/README.doctree b/docs/build/html/.doctrees/README.doctree index 91c38e3..834b670 100644 Binary files a/docs/build/html/.doctrees/README.doctree and b/docs/build/html/.doctrees/README.doctree differ diff --git a/docs/build/html/.doctrees/environment.pickle b/docs/build/html/.doctrees/environment.pickle index 58ecfcf..ba3b85c 100644 Binary files a/docs/build/html/.doctrees/environment.pickle and b/docs/build/html/.doctrees/environment.pickle differ diff --git a/docs/build/html/.doctrees/index.doctree b/docs/build/html/.doctrees/index.doctree index f5b0768..acc570b 100644 Binary files a/docs/build/html/.doctrees/index.doctree and b/docs/build/html/.doctrees/index.doctree differ diff --git a/docs/build/html/README.html b/docs/build/html/README.html index 2e810dc..7557b96 100644 --- a/docs/build/html/README.html +++ b/docs/build/html/README.html @@ -8,7 +8,7 @@ - Keyword-Service — CV Analysis Service 2.5.1 documentation + cv-analysis - Visual (CV-Based) Document Parsing — CV Analysis Service 2.5.1 documentation @@ -44,7 +44,7 @@ - + @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -343,7 +343,7 @@ document.write(` -
  • Keyword-Service
    • +
  • cv-analysis...
    • @@ -360,193 +360,163 @@ document.write(`
    -
    -

    Keyword-Service#

    -

    Service to get keywords of a paragraph or whole document.

    - -
    +
    +

    cv-analysis - Visual (CV-Based) Document Parsing#

    +

    parse_pdf() +This repository implements computer vision based approaches for detecting and parsing visual features such as tables or +previous redactions in documents.

    +

    API#

    -
    -

    REST#

    -

    The service provides endpoints to extract keywords from a text and to embed a text. For details, download -OpenAPI schema and view it in a browser.

    -
    -
    -

    RabbitMQ#

    -

    The service listens to a queue and processes the messages. This method is ment to be used for extracting keywords from -whole documents. All RabbitMQ parameters including the queue names are set in environment variables, refer to the -service respective HELM chart for more information.

    -

    The input message should be a JSON object with the following structure:

    +

    Input message:

    {
-  "targetFilePath": string,
-  "responseFilePath": string
+  "targetFilePath": {
+    "pdf": "absolute file path",
+    "vlp_output": "absolute file path"
+  },
+  "responseFilePath": "absolute file path",
+  "operation": "table_image_inference"
 }
    -

    The service downloads the file specified in targetFilePath. Supported data structures for the target file are:

    -
      -

    • simplified text data (signifier key: paragraphs)

      • -

    • structure object data (signifier key: structureObjects)

      • -
    -

    As a response, the service uploads a JSON-structured file (as defined in responseFilePath) with the result under the -data key. The structure of the response file is as follows:

    -
    {
-    "targetFilePath"
-:
-    string,
-        "responseFilePath"
-:
-    string,
-        // and eventually further fields if present in the input message      
-        "data"
-:
-    [
+
    Response is uploaded to the storage as specified in the responseFilePath field. The structure is as follows:
    
+
    {
+  ...,
+  "data": [
+    {
+      'pageNum': 0,
+      'bbox': {
+        'x1': 55.3407,
+        'y1': 247.0246,
+        'x2': 558.5602,
+        'y2': 598.0585
+      },
+      'uuid': '2b10c1a2-393c-4fca-b9e3-0ad5b774ac84',
+      'label': 'table',
+      'tableLines': [
         {
-            "keywords": Array[string],
-            "paragraphId": int,
-            "embedding": Array[float]  // 384 dimensions
-        }
-    ]
+          'x1': 0,
+          'y1': 16,
+          'x2': 1399,
+          'y2': 16
+        },
+        ...
+      ],
+      'imageInfo': {
+        'height': 693,
+        'width': 1414
+      }
+    },
+    ...
+  ]
 }
 
    
 
    
-
    Note that
    
-
      
-
    • the embedding key is optional and can be omitted. The service will not calculate the embedding if the environment
-variable MODEL__COMPUTE_EMBEDDINGS is set to false.
      • 
-
    • the service also computes the keywords for the whole document. In this case, the paragraphId is set to -1.
      • 
-
    -
    -
    -

    Service Configuration#

    -

    The service is configured via environment variables. The following variables are available:

    -

    | Variable | Description | Default | -| —————————————— | ———————————————————————————– | ——- | -| LOGGING__LEVEL | Logging level | INFO | -| MODEL__MAX_KEYWORDS_PER_PARAGRAPH | Maximum number of keywords per paragraph | 5 | -| MODLE__MAX_KEYWORDS_PER_DOCUMENT | Maximum number of keywords per document, when set to 0, no keywords are extracted | 0 | -| MODEL__COMPUTE_EMBEDDINGS | Whether to compute keyword embeddings or not | true | -| MODEL__PREPROCESSING__MIN_PARAGRAPH_LENGTH | Minimum number of characters in a paragraph to be considered for keyword extraction | 1 | -| MODEL__POSTPROCESSING__FILTER_SUBWORDS | Whether to filter out subwords from the keywords or not | true |

    -

    NOTE that these variables are subject to change. For the most recent configuration, refer to the service respective -HELM chart.

    -
    -
    -

    Language#

    -

    Currently, there is an english, a german and a multi-language model for keyword extraction. The models are uploaded to -mlflow and can -be set in the Dockerfile when building the container:

    -

    example for german model:

    -
    ENV AZURE_RESOURCE_GROUP="mriedl"
-ENV AZURE_AML_WORKSPACE="azureml-ws"
-ENV AZURE_AML_MODEL_NAME="keyword-extraction-de"
-ENV AZURE_AML_MODEL_VERSION="1"
-
    -
    -

    and example for english model:

    -
    ENV AZURE_RESOURCE_GROUP="mriedl"
-ENV AZURE_AML_WORKSPACE="azureml-ws"
-ENV AZURE_AML_MODEL_NAME="keyword-extraction-de"
-ENV AZURE_AML_MODEL_VERSION="1"
+
    
+
    Installation#
    
+
    git clone ssh://git@git.iqser.com:2222/rr/cv-analysis.git
+cd cv-analysis
+
+python -m venv env
+source env/bin/activate
+
+pip install -e .
+pip install -r requirements.txt
+
+dvc pull
 
    
 
    
 
    
 
    
 
    Usage#
    
-
    Two Options:
    
-
      
-
    1. REST: Send text per request to endpoint, endpoint returns keywords
      2. 
-
    2. Queue: Service gets text from queue, model calculates keywords, save keywords in queue
      3. 
-
    
-
    To test the REST endpoint you have to set up an environment and do poetry install (
-see https://gitlab.knecon.com/knecon/research/template-python-project for details for setting up poetry)
-Then run
    
-
    python ./src/serve.py 
-
    
-
    
-
    You don’t need to start a queue for that, just ignore the AMQP Error.
-Port and host are set in settings.toml .
-You can use the FastAPI under 127.0.0.1:8001/docs to send request to endpoint.
    
-
    You can also test the service with docker:
    
-
    
-
    Run Docker Commands#
    
-
    docker build -t ${IMAGE_NAME} -f Dockerfile --build-arg GITLAB_USER=${GITLAB_USER} \
-                                            --build-arg GITLAB_ACCESS_TOKEN=${GITLAB_ACCESS_TOKEN} \
-                                            --build-arg AZURE_TENANT_ID=${AZURE_TENANT_ID} \
-                                            --build-arg AZURE_SUBSCRIPTION_ID=${AZURE_SUBSCRIPTION_ID} \
-                                            --build-arg AZURE_CLIENT_ID=${AZURE_CLIENT_ID} \
-                                            --build-arg AZURE_CLIENT_SECRET=${AZURE_CLIENT_SECRET} \
-                                            --build-arg AZURE_AML_MODEL_VERSION=${AZURE_AML_MODEL_VERSION} \
-                                            --build-arg AZURE_AML_MODEL_NAME=${AZURE_AML_MODEL_NAME} \
-                                            --build-arg AZURE_RESOURCE_GROUP=${AZURE_RESOURCE_GROUP} \
-                                            --build-arg AZURE_AML_WORKSPACE=${AZURE_AML_WORKSPACE}
-
    
-
    
-
    docker run --net=host -it --rm --name ${CONTAINER_NAME} ${IMAGE_NAME}
+
    
+
    As an API#
    
+
    The module provided functions for the individual tasks that all return some kind of collection of points, depending on
+the specific task.
    
+
    
+
    Redaction Detection (API)#
    
+
    The below snippet shows hot to find the outlines of previous redactions.
    
+
    from cv_analysis.redaction_detection import find_redactions
+import pdf2image
+import numpy as np
+
+pdf_path = ...
+page_index = ...
+
+page = pdf2image.convert_from_path(pdf_path, first_page=page_index, last_page=page_index)[0]
+page = np.array(page)
+
+redaction_contours = find_redactions(page)
 
    
 
    
 
    
-
    
-
    Run locally#
    
-
    First you need to download the model from mlflow. This can be done with the “src/ml_flow/download_model.py” script.
-This scripts downloads a model and copies config and model data to the specific locations, such that the model can
-be loaded.
    
-
    For running/testing the keyword extraction locally you can use the src/tests/test_process.py script.
    
-
    Model ist stored and loaded via DVC, you need the connection string under
-https://portal.azure.com/#@knecon.com/resource/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourceGroups/taas-rg/providers/Microsoft.Storage/storageAccounts/taassaracer/keys
    
 
    
 
    
-
    
-
    
-
    Upload models to ML Flow#
    
-
    To upload the models to mlflow, you can use following script: src/mlflow/upload_model.py
-For authentication following environment variables need to be set:
    
-
    #AZURE_TENANT_ID=""
-#AZURE_SUBSCRIPTION_ID=""
-#AZURE_CLIENT_ID=""
-#AZURE_CLIENT_SECRET=""
+
    
+
    As a CLI Tool#
    
+
    Core API functionalities can be used through a CLI.
    
+
    
+
    Table Parsing#
    
+
    The tables parsing utility detects and segments tables into individual cells.
    
+
    python scripts/annotate.py data/test_pdf.pdf 7 --type table
 
    
 
    
-
    Additional settings (resource group, experiment name, etc.) can be specified in the config (
-./src/mlflow/config/azure_config.toml).
-The upload_model.py has the following parameters:
    
-
    options:
-  -h, --help            show this help message and exit
-  -a AZURE_CONFIG, --azure_config AZURE_CONFIG
-                        Location of the configuration file for Azure (default: src/mlflow/config/azure_config.toml)
-  -b BASE_CONFIG, --base_config BASE_CONFIG
-                        Location of the basic training configuration (default: src/mlflow/config/settings_de.toml)
-  
+
    The below image shows a parsed table, where each table cell has been detected individually.
    
+
    Table Parsing Demonstration
    
+
    
+
    
+
    Redaction Detection (CLI)#
    
+
    The redaction detection utility detects previous redactions in PDFs (filled black rectangles).
    
+
    python scripts/annotate.py data/test_pdf.pdf 2 --type redaction
 
    
 
    
-
    the base config contains all information for the models used. Examples for German and
-English are placed in /src/mlflow/config/
    
-
    Note: Multi-language model tracking does not work for now. After the upload script reports an error, you have to
-manually track the
-model here
-where you can find the run. Adhere to the naming conventions for the model name and versions,
-see here
    
+
    The below image shows the detected redactions with green outlines.
    
+
    Redaction Detection Demonstration
    
+
    
+
    
+
    Layout Parsing#
    
+
    The layout parsing utility detects elements such as paragraphs, tables and figures.
    
+
    python scripts/annotate.py data/test_pdf.pdf 7 --type layout
+
    
+
    
+
    The below image shows the detected layout elements on a page.
    
+
    Layout Parsing Demonstration
    
+
    
+
    
+
    Figure Detection#
    
+
    The figure detection utility detects figures specifically, which can be missed by the generic layout parsing utility.
    
+
    python scripts/annotate.py data/test_pdf.pdf 3 --type figure
+
    
+
    
+
    The below image shows the detected figure on a page.
    
+
    Figure Detection Demonstration
    
+
    
+
    
+
    
+
    Running as a service#
    
+
    
+
    Building#
    
+
    Build base image
    
+
    bash setup/docker.sh
+
    
+
    
+
    Build head image
    
+
    docker build -f Dockerfile -t cv-analysis . --build-arg BASE_ROOT=""
+
    
+
    
+
    
+
    
+
    Usage (service)#
    
+
    Shell 1
    
+
    docker run --rm --net=host --rm cv-analysis
+
    
+
    
+
    Shell 2
    
+
    python scripts/client_mock.py --pdf_path /path/to/a/pdf
+
    
+
    
+
    
+
    
 
    
 
 
@@ -565,7 +535,7 @@ see 
       
    
         
    previous
    
-        
    Welcome to Keyword Extraction Service documentation!
    
+        
    Welcome to CV Analysis Service documentation!
    diff --git a/docs/build/html/_images/figure_detection.png b/docs/build/html/_images/figure_detection.png new file mode 100644 index 0000000..7716ade Binary files /dev/null and b/docs/build/html/_images/figure_detection.png differ diff --git a/docs/build/html/_images/layout_parsing.png b/docs/build/html/_images/layout_parsing.png new file mode 100644 index 0000000..6b2a12a Binary files /dev/null and b/docs/build/html/_images/layout_parsing.png differ diff --git a/docs/build/html/_images/redaction_detection.png b/docs/build/html/_images/redaction_detection.png new file mode 100644 index 0000000..6df3f30 Binary files /dev/null and b/docs/build/html/_images/redaction_detection.png differ diff --git a/docs/build/html/_images/table_parsing.png b/docs/build/html/_images/table_parsing.png new file mode 100644 index 0000000..c3d4f9a Binary files /dev/null and b/docs/build/html/_images/table_parsing.png differ diff --git a/docs/build/html/_sources/README.md.txt b/docs/build/html/_sources/README.md.txt index b1901d1..2c44979 100644 --- a/docs/build/html/_sources/README.md.txt +++ b/docs/build/html/_sources/README.md.txt @@ -1,203 +1,178 @@ -# Keyword-Service +# cv-analysis - Visual (CV-Based) Document Parsing -Service to get keywords of a paragraph or whole document. - - - -- [Keyword-Service](#keyword-service) - - [API](#api) - - [REST](#rest) - - [RabbitMQ](#rabbitmq) - - [Service Configuration](#service-configuration) - - [Language](#language) - - [Usage](#usage) - - [Run Docker Commands](#run-docker-commands) - - [Run locally](#run-locally) -- [Upload models to ML Flow](#upload-models-to-ml-flow) - - +parse_pdf() +This repository implements computer vision based approaches for detecting and parsing visual features such as tables or +previous redactions in documents. ## API -### REST - -The service provides endpoints to extract keywords from a text and to embed a text. For details, download -[OpenAPI schema](docs/openapi_redoc.html) and view it in a browser. - -### RabbitMQ - -The service listens to a queue and processes the messages. This method is ment to be used for extracting keywords from -whole documents. All RabbitMQ parameters including the queue names are set in environment variables, refer to the -service respective HELM chart for more information. - -The input message should be a JSON object with the following structure: +Input message: ```json { - "targetFilePath": string, - "responseFilePath": string + "targetFilePath": { + "pdf": "absolute file path", + "vlp_output": "absolute file path" + }, + "responseFilePath": "absolute file path", + "operation": "table_image_inference" } ``` -The service downloads the file specified in `targetFilePath`. Supported data structures for the target file are: +Response is uploaded to the storage as specified in the `responseFilePath` field. The structure is as follows: -- simplified text data (signifier key: `paragraphs`) -- structure object data (signifier key: `structureObjects`) - -As a response, the service uploads a JSON-structured file (as defined in `responseFilePath`) with the result under the -`data` key. The structure of the response file is as follows: - -```javascript +```json { - "targetFilePath" -: - string, - "responseFilePath" -: - string, - // and eventually further fields if present in the input message - "data" -: - [ + ..., + "data": [ + { + 'pageNum': 0, + 'bbox': { + 'x1': 55.3407, + 'y1': 247.0246, + 'x2': 558.5602, + 'y2': 598.0585 + }, + 'uuid': '2b10c1a2-393c-4fca-b9e3-0ad5b774ac84', + 'label': 'table', + 'tableLines': [ { - "keywords": Array[string], - "paragraphId": int, - "embedding": Array[float] // 384 dimensions - } - ] + 'x1': 0, + 'y1': 16, + 'x2': 1399, + 'y2': 16 + }, + ... + ], + 'imageInfo': { + 'height': 693, + 'width': 1414 + } + }, + ... + ] } -``` - -**Note** that - -- the `embedding` key is optional and can be omitted. The service will not calculate the embedding if the environment - variable `MODEL__COMPUTE_EMBEDDINGS` is set to `false`. -- the service also computes the keywords for the whole document. In this case, the `paragraphId` is set to `-1`. - -## Service Configuration - -The service is configured via environment variables. The following variables are available: - -| Variable | Description | Default | -| ------------------------------------------ | ----------------------------------------------------------------------------------- | ------- | -| LOGGING__LEVEL | Logging level | INFO | -| MODEL__MAX_KEYWORDS_PER_PARAGRAPH | Maximum number of keywords per paragraph | 5 | -| MODLE__MAX_KEYWORDS_PER_DOCUMENT | Maximum number of keywords per document, when set to 0, no keywords are extracted | 0 | -| MODEL__COMPUTE_EMBEDDINGS | Whether to compute keyword embeddings or not | true | -| MODEL__PREPROCESSING__MIN_PARAGRAPH_LENGTH | Minimum number of characters in a paragraph to be considered for keyword extraction | 1 | -| MODEL__POSTPROCESSING__FILTER_SUBWORDS | Whether to filter out subwords from the keywords or not | true | - -**NOTE** that these variables are subject to change. For the most recent configuration, refer to the service respective -HELM chart. - -## Language - -Currently, there is an english, a german and a multi-language model for keyword extraction. The models are uploaded to -mlflow and can -be set in the Dockerfile when building the container: - -example for german model: -``` -ENV AZURE_RESOURCE_GROUP="mriedl" -ENV AZURE_AML_WORKSPACE="azureml-ws" -ENV AZURE_AML_MODEL_NAME="keyword-extraction-de" -ENV AZURE_AML_MODEL_VERSION="1" ``` -and example for english model: +## Installation -``` -ENV AZURE_RESOURCE_GROUP="mriedl" -ENV AZURE_AML_WORKSPACE="azureml-ws" -ENV AZURE_AML_MODEL_NAME="keyword-extraction-de" -ENV AZURE_AML_MODEL_VERSION="1" +```bash +git clone ssh://git@git.iqser.com:2222/rr/cv-analysis.git +cd cv-analysis + +python -m venv env +source env/bin/activate + +pip install -e . +pip install -r requirements.txt + +dvc pull ``` ## Usage -**Two Options:** +### As an API -1. REST: Send text per request to endpoint, endpoint returns keywords -2. Queue: Service gets text from queue, model calculates keywords, save keywords in queue +The module provided functions for the individual tasks that all return some kind of collection of points, depending on +the specific task. -To test the REST endpoint you have to set up an environment and do poetry install ( -see https://gitlab.knecon.com/knecon/research/template-python-project for details for setting up poetry) -Then run +#### Redaction Detection (API) -``` -python ./src/serve.py +The below snippet shows hot to find the outlines of previous redactions. + +```python +from cv_analysis.redaction_detection import find_redactions +import pdf2image +import numpy as np + +pdf_path = ... +page_index = ... + +page = pdf2image.convert_from_path(pdf_path, first_page=page_index, last_page=page_index)[0] +page = np.array(page) + +redaction_contours = find_redactions(page) ``` -You don't need to start a queue for that, just ignore the AMQP Error. -Port and host are set in settings.toml . -You can use the FastAPI under 127.0.0.1:8001/docs to send request to endpoint. +## As a CLI Tool -You can also test the service with docker: +Core API functionalities can be used through a CLI. -#### Run Docker Commands +### Table Parsing + +The tables parsing utility detects and segments tables into individual cells. ```bash -docker build -t ${IMAGE_NAME} -f Dockerfile --build-arg GITLAB_USER=${GITLAB_USER} \ - --build-arg GITLAB_ACCESS_TOKEN=${GITLAB_ACCESS_TOKEN} \ - --build-arg AZURE_TENANT_ID=${AZURE_TENANT_ID} \ - --build-arg AZURE_SUBSCRIPTION_ID=${AZURE_SUBSCRIPTION_ID} \ - --build-arg AZURE_CLIENT_ID=${AZURE_CLIENT_ID} \ - --build-arg AZURE_CLIENT_SECRET=${AZURE_CLIENT_SECRET} \ - --build-arg AZURE_AML_MODEL_VERSION=${AZURE_AML_MODEL_VERSION} \ - --build-arg AZURE_AML_MODEL_NAME=${AZURE_AML_MODEL_NAME} \ - --build-arg AZURE_RESOURCE_GROUP=${AZURE_RESOURCE_GROUP} \ - --build-arg AZURE_AML_WORKSPACE=${AZURE_AML_WORKSPACE} +python scripts/annotate.py data/test_pdf.pdf 7 --type table ``` +The below image shows a parsed table, where each table cell has been detected individually. + +![Table Parsing Demonstration](data/table_parsing.png) + +### Redaction Detection (CLI) + +The redaction detection utility detects previous redactions in PDFs (filled black rectangles). + ```bash -docker run --net=host -it --rm --name ${CONTAINER_NAME} ${IMAGE_NAME} +python scripts/annotate.py data/test_pdf.pdf 2 --type redaction ``` -#### Run locally +The below image shows the detected redactions with green outlines. -First you need to download the model from mlflow. This can be done with the *"src/ml_flow/download_model.py"* script. -This scripts downloads a model and copies config and model data to the specific locations, such that the model can -be loaded. +![Redaction Detection Demonstration](data/redaction_detection.png) -For running/testing the keyword extraction locally you can use the *src/tests/test_process.py* script. +### Layout Parsing -Model ist stored and loaded via DVC, you need the connection string under -https://portal.azure.com/#@knecon.com/resource/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourceGroups/taas-rg/providers/Microsoft.Storage/storageAccounts/taassaracer/keys +The layout parsing utility detects elements such as paragraphs, tables and figures. -# Upload models to ML Flow - -To upload the models to mlflow, you can use following script: src/mlflow/upload_model.py -For authentication following environment variables need to be set: - -``` -#AZURE_TENANT_ID="" -#AZURE_SUBSCRIPTION_ID="" -#AZURE_CLIENT_ID="" -#AZURE_CLIENT_SECRET="" +```bash +python scripts/annotate.py data/test_pdf.pdf 7 --type layout ``` -Additional settings (resource group, experiment name, etc.) can be specified in the config ( -*./src/mlflow/config/azure_config.toml*). -The *upload_model.py* has the following parameters: +The below image shows the detected layout elements on a page. -``` -options: - -h, --help show this help message and exit - -a AZURE_CONFIG, --azure_config AZURE_CONFIG - Location of the configuration file for Azure (default: src/mlflow/config/azure_config.toml) - -b BASE_CONFIG, --base_config BASE_CONFIG - Location of the basic training configuration (default: src/mlflow/config/settings_de.toml) - +![Layout Parsing Demonstration](data/layout_parsing.png) +### Figure Detection + +The figure detection utility detects figures specifically, which can be missed by the generic layout parsing utility. + +```bash +python scripts/annotate.py data/test_pdf.pdf 3 --type figure ``` -the base config contains all information for the models used. Examples for German and -English are placed in */src/mlflow/config/* +The below image shows the detected figure on a page. -Note: Multi-language model tracking does not work for now. After the upload script reports an error, you have to -manually track the -model [here](https://ml.azure.com/experiments?wsid=/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourcegroups/fforesight-rg/providers/Microsoft.MachineLearningServices/workspaces/ff-aml-main&tid=b44be368-e4f2-4ade-a089-cd2825458048) -where you can find the run. Adhere to the naming conventions for the model name and versions, -see [here](https://ml.azure.com/model/list?wsid=/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourcegroups/fforesight-rg/providers/Microsoft.MachineLearningServices/workspaces/ff-aml-main&tid=b44be368-e4f2-4ade-a089-cd2825458048) +![Figure Detection Demonstration](data/figure_detection.png) +## Running as a service + +### Building + +Build base image + +```bash +bash setup/docker.sh +``` + +Build head image + +```bash +docker build -f Dockerfile -t cv-analysis . --build-arg BASE_ROOT="" +``` + +### Usage (service) + +Shell 1 + +```bash +docker run --rm --net=host --rm cv-analysis +``` + +Shell 2 + +```bash +python scripts/client_mock.py --pdf_path /path/to/a/pdf +``` diff --git a/docs/build/html/_sources/index.rst.txt b/docs/build/html/_sources/index.rst.txt index e356eff..05aa9b4 100644 --- a/docs/build/html/_sources/index.rst.txt +++ b/docs/build/html/_sources/index.rst.txt @@ -3,9 +3,9 @@ You can adapt this file completely to your liking, but it should at least contain the root `toctree` directive. -==================================================== -Welcome to Keyword Extraction Service documentation! -==================================================== +============================================= +Welcome to CV Analysis Service documentation! +============================================= .. note:: diff --git a/docs/build/html/genindex.html b/docs/build/html/genindex.html index 2baeb60..7896582 100644 --- a/docs/build/html/genindex.html +++ b/docs/build/html/genindex.html @@ -154,7 +154,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -254,7 +254,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/index.html b/docs/build/html/index.html index 4a672e3..ec81aa5 100644 --- a/docs/build/html/index.html +++ b/docs/build/html/index.html @@ -8,7 +8,7 @@ - Welcome to Keyword Extraction Service documentation! — CV Analysis Service 2.5.1 documentation + Welcome to CV Analysis Service documentation! — CV Analysis Service 2.5.1 documentation @@ -43,7 +43,7 @@ - + @@ -156,7 +156,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -260,7 +260,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -326,8 +326,8 @@ document.write(`
    -
    -

    Welcome to Keyword Extraction Service documentation!#

    +
    +

    Welcome to CV Analysis Service documentation!#

    Note

    If you’d like to change the looks of things 👉 https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html

    @@ -337,22 +337,27 @@ document.write(`

    README

    @@ -405,7 +410,7 @@ document.write(` title="next page">

    next

    -

    Keyword-Service

    +

    cv-analysis - Visual (CV-Based) Document Parsing

    diff --git a/docs/build/html/modules/cv_analysis.config.html b/docs/build/html/modules/cv_analysis.config.html index a737afd..01241dc 100644 --- a/docs/build/html/modules/cv_analysis.config.html +++ b/docs/build/html/modules/cv_analysis.config.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.figure_detection.figure_detection.html b/docs/build/html/modules/cv_analysis.figure_detection.figure_detection.html index 23b7d48..2fb5a6e 100644 --- a/docs/build/html/modules/cv_analysis.figure_detection.figure_detection.html +++ b/docs/build/html/modules/cv_analysis.figure_detection.figure_detection.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.figure_detection.figures.html b/docs/build/html/modules/cv_analysis.figure_detection.figures.html index 2dea945..4637184 100644 --- a/docs/build/html/modules/cv_analysis.figure_detection.figures.html +++ b/docs/build/html/modules/cv_analysis.figure_detection.figures.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.figure_detection.html b/docs/build/html/modules/cv_analysis.figure_detection.html index e96d62b..ba2f320 100644 --- a/docs/build/html/modules/cv_analysis.figure_detection.html +++ b/docs/build/html/modules/cv_analysis.figure_detection.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.figure_detection.text.html b/docs/build/html/modules/cv_analysis.figure_detection.text.html index 6abca43..0384498 100644 --- a/docs/build/html/modules/cv_analysis.figure_detection.text.html +++ b/docs/build/html/modules/cv_analysis.figure_detection.text.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.html b/docs/build/html/modules/cv_analysis.html index c4fffaf..8b9ea80 100644 --- a/docs/build/html/modules/cv_analysis.html +++ b/docs/build/html/modules/cv_analysis.html @@ -44,7 +44,7 @@ - + @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -576,7 +576,7 @@ document.write(`

    previous

    -

    Keyword-Service

    +

    cv-analysis - Visual (CV-Based) Document Parsing

    - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.locations.html b/docs/build/html/modules/cv_analysis.locations.html index 139ec5e..f33a4d9 100644 --- a/docs/build/html/modules/cv_analysis.locations.html +++ b/docs/build/html/modules/cv_analysis.locations.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.redaction_detection.html b/docs/build/html/modules/cv_analysis.redaction_detection.html index 7566d07..c28394b 100644 --- a/docs/build/html/modules/cv_analysis.redaction_detection.html +++ b/docs/build/html/modules/cv_analysis.redaction_detection.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.server.html b/docs/build/html/modules/cv_analysis.server.html index 0036e6d..169a5d0 100644 --- a/docs/build/html/modules/cv_analysis.server.html +++ b/docs/build/html/modules/cv_analysis.server.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.server.pipeline.html b/docs/build/html/modules/cv_analysis.server.pipeline.html index c7eb2f6..49084bc 100644 --- a/docs/build/html/modules/cv_analysis.server.pipeline.html +++ b/docs/build/html/modules/cv_analysis.server.pipeline.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.table_inference.html b/docs/build/html/modules/cv_analysis.table_inference.html index 54fd7cb..d4a8d0d 100644 --- a/docs/build/html/modules/cv_analysis.table_inference.html +++ b/docs/build/html/modules/cv_analysis.table_inference.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.table_parsing.html b/docs/build/html/modules/cv_analysis.table_parsing.html index 6b2329b..110f20d 100644 --- a/docs/build/html/modules/cv_analysis.table_parsing.html +++ b/docs/build/html/modules/cv_analysis.table_parsing.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.annotate.html b/docs/build/html/modules/cv_analysis.utils.annotate.html index fcdb659..df4fdfc 100644 --- a/docs/build/html/modules/cv_analysis.utils.annotate.html +++ b/docs/build/html/modules/cv_analysis.utils.annotate.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.banner.html b/docs/build/html/modules/cv_analysis.utils.banner.html index 797a2cc..2cc865b 100644 --- a/docs/build/html/modules/cv_analysis.utils.banner.html +++ b/docs/build/html/modules/cv_analysis.utils.banner.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.connect_rects.html b/docs/build/html/modules/cv_analysis.utils.connect_rects.html index 4376a00..c01b453 100644 --- a/docs/build/html/modules/cv_analysis.utils.connect_rects.html +++ b/docs/build/html/modules/cv_analysis.utils.connect_rects.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.display.html b/docs/build/html/modules/cv_analysis.utils.display.html index aa8a3b6..7f747a6 100644 --- a/docs/build/html/modules/cv_analysis.utils.display.html +++ b/docs/build/html/modules/cv_analysis.utils.display.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.draw.html b/docs/build/html/modules/cv_analysis.utils.draw.html index 4d5ee08..6c58a47 100644 --- a/docs/build/html/modules/cv_analysis.utils.draw.html +++ b/docs/build/html/modules/cv_analysis.utils.draw.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.filters.html b/docs/build/html/modules/cv_analysis.utils.filters.html index 19f4474..c3dd534 100644 --- a/docs/build/html/modules/cv_analysis.utils.filters.html +++ b/docs/build/html/modules/cv_analysis.utils.filters.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.html b/docs/build/html/modules/cv_analysis.utils.html index df05c1c..d7c3044 100644 --- a/docs/build/html/modules/cv_analysis.utils.html +++ b/docs/build/html/modules/cv_analysis.utils.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.image_extraction.html b/docs/build/html/modules/cv_analysis.utils.image_extraction.html index 67b5ee2..7bee4d7 100644 --- a/docs/build/html/modules/cv_analysis.utils.image_extraction.html +++ b/docs/build/html/modules/cv_analysis.utils.image_extraction.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.open_pdf.html b/docs/build/html/modules/cv_analysis.utils.open_pdf.html index 28e34ed..3cf8f07 100644 --- a/docs/build/html/modules/cv_analysis.utils.open_pdf.html +++ b/docs/build/html/modules/cv_analysis.utils.open_pdf.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.postprocessing.html b/docs/build/html/modules/cv_analysis.utils.postprocessing.html index 52da476..723deff 100644 --- a/docs/build/html/modules/cv_analysis.utils.postprocessing.html +++ b/docs/build/html/modules/cv_analysis.utils.postprocessing.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.preprocessing.html b/docs/build/html/modules/cv_analysis.utils.preprocessing.html index 3692f62..3fc24d2 100644 --- a/docs/build/html/modules/cv_analysis.utils.preprocessing.html +++ b/docs/build/html/modules/cv_analysis.utils.preprocessing.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.structures.html b/docs/build/html/modules/cv_analysis.utils.structures.html index 82af23a..b4ab438 100644 --- a/docs/build/html/modules/cv_analysis.utils.structures.html +++ b/docs/build/html/modules/cv_analysis.utils.structures.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.test_metrics.html b/docs/build/html/modules/cv_analysis.utils.test_metrics.html index 28d8dfd..a2f1697 100644 --- a/docs/build/html/modules/cv_analysis.utils.test_metrics.html +++ b/docs/build/html/modules/cv_analysis.utils.test_metrics.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.utils.html b/docs/build/html/modules/cv_analysis.utils.utils.html index 3065095..29fda0b 100644 --- a/docs/build/html/modules/cv_analysis.utils.utils.html +++ b/docs/build/html/modules/cv_analysis.utils.utils.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/cv_analysis.utils.visual_logging.html b/docs/build/html/modules/cv_analysis.utils.visual_logging.html index c6024f5..a781392 100644 --- a/docs/build/html/modules/cv_analysis.utils.visual_logging.html +++ b/docs/build/html/modules/cv_analysis.utils.visual_logging.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -259,7 +259,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/modules/serve.html b/docs/build/html/modules/serve.html index 7809498..c0b8b81 100644 --- a/docs/build/html/modules/serve.html +++ b/docs/build/html/modules/serve.html @@ -156,7 +156,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -258,7 +258,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/objects.inv b/docs/build/html/objects.inv index 6ca6054..d8e0682 100644 Binary files a/docs/build/html/objects.inv and b/docs/build/html/objects.inv differ diff --git a/docs/build/html/py-modindex.html b/docs/build/html/py-modindex.html index ae52033..56aecc0 100644 --- a/docs/build/html/py-modindex.html +++ b/docs/build/html/py-modindex.html @@ -157,7 +157,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -257,7 +257,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/search.html b/docs/build/html/search.html index eb4549d..a1b05b4 100644 --- a/docs/build/html/search.html +++ b/docs/build/html/search.html @@ -156,7 +156,7 @@ READTHEDOCS_DATA = JSON.parse(document.getElementById('READTHEDOCS_DATA').innerH
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • @@ -256,7 +256,7 @@ document.write(`
  • - Keyword-Service + cv-analysis - Visual (CV-Based) Document Parsing
    • diff --git a/docs/build/html/searchindex.js b/docs/build/html/searchindex.js index f8a89ce..58d0b8c 100644 --- a/docs/build/html/searchindex.js +++ b/docs/build/html/searchindex.js @@ -1 +1 @@ -Search.setIndex({"alltitles": {"API": [[0, "api"]], "Indices and tables": [[1, "indices-and-tables"]], "Keyword-Service": [[0, "keyword-service"]], "Language": [[0, "language"]], "Modules": [[1, null]], "README": [[1, null]], "REST": [[0, "rest"]], "RabbitMQ": [[0, "rabbitmq"]], "Run Docker Commands": [[0, "run-docker-commands"]], "Run locally": [[0, "run-locally"]], "Service Configuration": [[0, "service-configuration"]], "Submodules": [[2, "submodules"], [4, "submodules"], [11, "submodules"], [15, "submodules"]], "Subpackages": [[2, "subpackages"]], "Table of Contents": [[1, "table-of-contents"]], "Upload models to ML Flow": [[0, "upload-models-to-ml-flow"]], "Usage": [[0, "usage"]], "Welcome to Keyword Extraction Service documentation!": [[1, "welcome-to-keyword-extraction-service-documentation"]], "cv_analysis package": [[2, "module-cv_analysis"]], "cv_analysis.config module": [[3, "module-cv_analysis.config"]], "cv_analysis.figure_detection package": [[4, "module-cv_analysis.figure_detection"]], "cv_analysis.figure_detection.figure_detection module": [[5, "module-cv_analysis.figure_detection.figure_detection"]], "cv_analysis.figure_detection.figures module": [[6, "module-cv_analysis.figure_detection.figures"]], "cv_analysis.figure_detection.text module": [[7, "module-cv_analysis.figure_detection.text"]], "cv_analysis.layout_parsing module": [[8, "cv-analysis-layout-parsing-module"]], "cv_analysis.locations module": [[9, "module-cv_analysis.locations"]], "cv_analysis.redaction_detection module": [[10, "cv-analysis-redaction-detection-module"]], "cv_analysis.server package": [[11, "module-cv_analysis.server"]], "cv_analysis.server.pipeline module": [[12, "cv-analysis-server-pipeline-module"]], "cv_analysis.table_inference module": [[13, "module-cv_analysis.table_inference"]], "cv_analysis.table_parsing module": [[14, "cv-analysis-table-parsing-module"]], "cv_analysis.utils package": [[15, "module-cv_analysis.utils"]], "cv_analysis.utils.annotate module": [[16, "module-cv_analysis.utils.annotate"]], "cv_analysis.utils.banner module": [[17, "module-cv_analysis.utils.banner"]], "cv_analysis.utils.connect_rects module": [[18, "module-cv_analysis.utils.connect_rects"]], "cv_analysis.utils.display module": [[19, "module-cv_analysis.utils.display"]], "cv_analysis.utils.draw module": [[20, "module-cv_analysis.utils.draw"]], "cv_analysis.utils.filters module": [[21, "module-cv_analysis.utils.filters"]], "cv_analysis.utils.image_extraction module": [[22, "module-cv_analysis.utils.image_extraction"]], "cv_analysis.utils.open_pdf module": [[23, "module-cv_analysis.utils.open_pdf"]], "cv_analysis.utils.postprocessing module": [[24, "module-cv_analysis.utils.postprocessing"]], "cv_analysis.utils.preprocessing module": [[25, "module-cv_analysis.utils.preprocessing"]], "cv_analysis.utils.structures module": [[26, "module-cv_analysis.utils.structures"]], "cv_analysis.utils.test_metrics module": [[27, "module-cv_analysis.utils.test_metrics"]], "cv_analysis.utils.utils module": [[28, "module-cv_analysis.utils.utils"]], "cv_analysis.utils.visual_logging module": [[29, "cv-analysis-utils-visual-logging-module"]], "serve module": [[30, "module-serve"]]}, "docnames": ["README", "index", "modules/cv_analysis", "modules/cv_analysis.config", "modules/cv_analysis.figure_detection", "modules/cv_analysis.figure_detection.figure_detection", "modules/cv_analysis.figure_detection.figures", "modules/cv_analysis.figure_detection.text", "modules/cv_analysis.layout_parsing", "modules/cv_analysis.locations", "modules/cv_analysis.redaction_detection", "modules/cv_analysis.server", "modules/cv_analysis.server.pipeline", "modules/cv_analysis.table_inference", "modules/cv_analysis.table_parsing", "modules/cv_analysis.utils", "modules/cv_analysis.utils.annotate", "modules/cv_analysis.utils.banner", "modules/cv_analysis.utils.connect_rects", "modules/cv_analysis.utils.display", "modules/cv_analysis.utils.draw", "modules/cv_analysis.utils.filters", "modules/cv_analysis.utils.image_extraction", "modules/cv_analysis.utils.open_pdf", "modules/cv_analysis.utils.postprocessing", "modules/cv_analysis.utils.preprocessing", "modules/cv_analysis.utils.structures", "modules/cv_analysis.utils.test_metrics", "modules/cv_analysis.utils.utils", "modules/cv_analysis.utils.visual_logging", "modules/serve"], "envversion": {"sphinx": 61, "sphinx.domains.c": 3, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 9, "sphinx.domains.index": 1, "sphinx.domains.javascript": 3, "sphinx.domains.math": 2, "sphinx.domains.python": 4, "sphinx.domains.rst": 2, "sphinx.domains.std": 2}, "filenames": ["README.md", "index.rst", "modules/cv_analysis.rst", "modules/cv_analysis.config.rst", "modules/cv_analysis.figure_detection.rst", "modules/cv_analysis.figure_detection.figure_detection.rst", "modules/cv_analysis.figure_detection.figures.rst", "modules/cv_analysis.figure_detection.text.rst", "modules/cv_analysis.layout_parsing.rst", "modules/cv_analysis.locations.rst", "modules/cv_analysis.redaction_detection.rst", "modules/cv_analysis.server.rst", "modules/cv_analysis.server.pipeline.rst", "modules/cv_analysis.table_inference.rst", "modules/cv_analysis.table_parsing.rst", "modules/cv_analysis.utils.rst", "modules/cv_analysis.utils.annotate.rst", "modules/cv_analysis.utils.banner.rst", "modules/cv_analysis.utils.connect_rects.rst", "modules/cv_analysis.utils.display.rst", "modules/cv_analysis.utils.draw.rst", "modules/cv_analysis.utils.filters.rst", "modules/cv_analysis.utils.image_extraction.rst", "modules/cv_analysis.utils.open_pdf.rst", "modules/cv_analysis.utils.postprocessing.rst", "modules/cv_analysis.utils.preprocessing.rst", "modules/cv_analysis.utils.structures.rst", "modules/cv_analysis.utils.test_metrics.rst", "modules/cv_analysis.utils.utils.rst", "modules/cv_analysis.utils.visual_logging.rst", "modules/serve.rst"], "indexentries": {"adjacent() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.adjacent", false]], "annotate_page() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.annotate_page", false]], "annotate_pdf() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.annotate_pdf", false]], "apply_threshold_to_image() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.apply_threshold_to_image", false]], "area() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.area", false]], "compute_document_score() (in module cv_analysis.utils.test_metrics)": [[27, "cv_analysis.utils.test_metrics.compute_document_score", false]], "compute_page_iou() (in module cv_analysis.utils.test_metrics)": [[27, "cv_analysis.utils.test_metrics.compute_page_iou", false]], "connect_related_rects() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.connect_related_rects", false]], "connect_related_rects2() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.connect_related_rects2", false]], "contour (class in cv_analysis.utils.structures)": [[26, "cv_analysis.utils.structures.Contour", false]], "copy_and_normalize_channels() (in module cv_analysis.utils.utils)": [[28, "cv_analysis.utils.utils.copy_and_normalize_channels", false]], "cv_analysis": [[2, "module-cv_analysis", false]], "cv_analysis.config": [[3, "module-cv_analysis.config", false]], "cv_analysis.figure_detection": [[4, "module-cv_analysis.figure_detection", false]], "cv_analysis.figure_detection.figure_detection": [[5, "module-cv_analysis.figure_detection.figure_detection", false]], "cv_analysis.figure_detection.figures": [[6, "module-cv_analysis.figure_detection.figures", false]], "cv_analysis.figure_detection.text": [[7, "module-cv_analysis.figure_detection.text", false]], "cv_analysis.locations": [[9, "module-cv_analysis.locations", false]], "cv_analysis.server": [[11, "module-cv_analysis.server", false]], "cv_analysis.table_inference": [[13, "module-cv_analysis.table_inference", false]], "cv_analysis.utils": [[15, "module-cv_analysis.utils", false]], "cv_analysis.utils.annotate": [[16, "module-cv_analysis.utils.annotate", false]], "cv_analysis.utils.banner": [[17, "module-cv_analysis.utils.banner", false]], "cv_analysis.utils.connect_rects": [[18, "module-cv_analysis.utils.connect_rects", false]], "cv_analysis.utils.display": [[19, "module-cv_analysis.utils.display", false]], "cv_analysis.utils.draw": [[20, "module-cv_analysis.utils.draw", false]], "cv_analysis.utils.filters": [[21, "module-cv_analysis.utils.filters", false]], "cv_analysis.utils.image_extraction": [[22, "module-cv_analysis.utils.image_extraction", false]], "cv_analysis.utils.open_pdf": [[23, "module-cv_analysis.utils.open_pdf", false]], "cv_analysis.utils.postprocessing": [[24, "module-cv_analysis.utils.postprocessing", false]], "cv_analysis.utils.preprocessing": [[25, "module-cv_analysis.utils.preprocessing", false]], "cv_analysis.utils.structures": [[26, "module-cv_analysis.utils.structures", false]], "cv_analysis.utils.test_metrics": [[27, "module-cv_analysis.utils.test_metrics", false]], "cv_analysis.utils.utils": [[28, "module-cv_analysis.utils.utils", false]], "detect_figures() (in module cv_analysis.figure_detection.figure_detection)": [[5, "cv_analysis.figure_detection.figure_detection.detect_figures", false]], "detect_large_coherent_structures() (in module cv_analysis.figure_detection.figures)": [[6, "cv_analysis.figure_detection.figures.detect_large_coherent_structures", false]], "dpi (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.dpi", false]], "draw_bboxes() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.draw_bboxes", false]], "draw_contours() (in module cv_analysis.utils.draw)": [[20, "cv_analysis.utils.draw.draw_contours", false]], "draw_rectangles() (in module cv_analysis.utils.draw)": [[20, "cv_analysis.utils.draw.draw_rectangles", false]], "extract_images_from_pdf() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.extract_images_from_pdf", false]], "filter_array() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.filter_array", false]], "filter_fp_col_lines() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.filter_fp_col_lines", false]], "find_max_overlap() (in module cv_analysis.utils.test_metrics)": [[27, "cv_analysis.utils.test_metrics.find_max_overlap", false]], "find_related_rects() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.find_related_rects", false]], "from_dict_xywh() (cv_analysis.utils.structures.rectangle class method)": [[26, "cv_analysis.utils.structures.Rectangle.from_dict_xywh", false]], "from_xywh() (cv_analysis.utils.structures.rectangle class method)": [[26, "cv_analysis.utils.structures.Rectangle.from_xywh", false]], "from_xyxy() (cv_analysis.utils.structures.rectangle class method)": [[26, "cv_analysis.utils.structures.Rectangle.from_xyxy", false]], "fuse_rects() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.fuse_rects", false]], "get_config() (in module cv_analysis.config)": [[3, "cv_analysis.config.get_config", false]], "get_lines_either() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.get_lines_either", false]], "has_acceptable_format() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.has_acceptable_format", false]], "has_correct_position1() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.has_correct_position1", false]], "has_no_parent() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.has_no_parent", false]], "height (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.height", false]], "image_height (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.image_height", false]], "image_width (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.image_width", false]], "img_bytes_to_array() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.img_bytes_to_array", false]], "includes() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.includes", false]], "infer_lines() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.infer_lines", false]], "intersection() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.intersection", false]], "iou() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.iou", false]], "is_boxy() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_boxy", false]], "is_filled() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_filled", false]], "is_included() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.is_included", false]], "is_large_enough() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_large_enough", false]], "is_likely_figure() (in module cv_analysis.figure_detection.figure_detection)": [[5, "cv_analysis.figure_detection.figure_detection.is_likely_figure", false]], "is_likely_primary_text_segment() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.is_likely_primary_text_segment", false]], "is_near_enough() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_near_enough", false]], "is_not_too_large() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_not_too_large", false]], "is_on_same_line() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_on_same_line", false]], "is_overlapping() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_overlapping", false]], "is_related() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_related", false]], "json() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json", false]], "json_full() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json_full", false]], "json_xywh() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json_xywh", false]], "json_xyxy() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json_xyxy", false]], "make_art() (in module cv_analysis.utils.banner)": [[17, "cv_analysis.utils.banner.make_art", false]], "make_gaussian_kernel() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.make_gaussian_kernel", false]], "make_gaussian_nonpositive_kernel() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.make_gaussian_nonpositive_kernel", false]], "make_quadratic_kernel() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.make_quadratic_kernel", false]], "min_avg_for_interval() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.min_avg_for_interval", false]], "mirror_on_x_axis() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.mirror_on_x_axis", false]], "module": [[2, "module-cv_analysis", false], [3, "module-cv_analysis.config", false], [4, "module-cv_analysis.figure_detection", false], [5, "module-cv_analysis.figure_detection.figure_detection", false], [6, "module-cv_analysis.figure_detection.figures", false], [7, "module-cv_analysis.figure_detection.text", false], [9, "module-cv_analysis.locations", false], [11, "module-cv_analysis.server", false], [13, "module-cv_analysis.table_inference", false], [15, "module-cv_analysis.utils", false], [16, "module-cv_analysis.utils.annotate", false], [17, "module-cv_analysis.utils.banner", false], [18, "module-cv_analysis.utils.connect_rects", false], [19, "module-cv_analysis.utils.display", false], [20, "module-cv_analysis.utils.draw", false], [21, "module-cv_analysis.utils.filters", false], [22, "module-cv_analysis.utils.image_extraction", false], [23, "module-cv_analysis.utils.open_pdf", false], [24, "module-cv_analysis.utils.postprocessing", false], [25, "module-cv_analysis.utils.preprocessing", false], [26, "module-cv_analysis.utils.structures", false], [27, "module-cv_analysis.utils.test_metrics", false], [28, "module-cv_analysis.utils.utils", false], [30, "module-serve", false]], "npconvert() (in module cv_analysis.utils.utils)": [[28, "cv_analysis.utils.utils.npconvert", false]], "open_pdf() (in module cv_analysis.utils.open_pdf)": [[23, "cv_analysis.utils.open_pdf.open_pdf", false]], "page2image() (in module cv_analysis.utils.preprocessing)": [[25, "cv_analysis.utils.preprocessing.page2image", false]], "page_num (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.page_num", false]], "pageinfo (class in cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.PageInfo", false]], "preprocess_page_array() (in module cv_analysis.utils.preprocessing)": [[25, "cv_analysis.utils.preprocessing.preprocess_page_array", false]], "provide_byte_stream() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.provide_byte_stream", false]], "rectangle (class in cv_analysis.utils.structures)": [[26, "cv_analysis.utils.structures.Rectangle", false]], "rects_not_the_same() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.rects_not_the_same", false]], "remove_included() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.remove_included", false]], "remove_isolated() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.remove_isolated", false]], "remove_overlapping() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.remove_overlapping", false]], "remove_primary_text_regions() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.remove_primary_text_regions", false]], "rescale_to_pdf() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.rescale_to_pdf", false]], "rotation (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.rotation", false]], "rotation_matrix (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.rotation_matrix", false]], "save_image() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.save_image", false]], "save_lines() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.save_lines", false]], "save_plot() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.save_plot", false]], "search_intervals() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.search_intervals", false]], "serve": [[30, "module-serve", false]], "show() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.show", false]], "show_image() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.show_image", false]], "show_image_cv2() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.show_image_cv2", false]], "show_image_mpl() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.show_image_mpl", false]], "show_multiple() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.show_multiple", false]], "transform_image_coordinates_to_pdf_coordinates() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.transform_image_coordinates_to_pdf_coordinates", false]], "transform_table_lines_by_page_info() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.transform_table_lines_by_page_info", false]], "transformation_matrix (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.transformation_matrix", false]], "width (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.width", false]], "xywh() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.xywh", false]], "xyxy() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.xyxy", false]]}, "objects": {"": [[2, 0, 0, "-", "cv_analysis"], [30, 0, 0, "-", "serve"]], "cv_analysis": [[3, 0, 0, "-", "config"], [4, 0, 0, "-", "figure_detection"], [9, 0, 0, "-", "locations"], [11, 0, 0, "-", "server"], [13, 0, 0, "-", "table_inference"], [15, 0, 0, "-", "utils"]], "cv_analysis.config": [[3, 1, 1, "", "get_config"]], "cv_analysis.figure_detection": [[5, 0, 0, "-", "figure_detection"], [6, 0, 0, "-", "figures"], [7, 0, 0, "-", "text"]], "cv_analysis.figure_detection.figure_detection": [[5, 1, 1, "", "detect_figures"], [5, 1, 1, "", "is_likely_figure"]], "cv_analysis.figure_detection.figures": [[6, 1, 1, "", "detect_large_coherent_structures"]], "cv_analysis.figure_detection.text": [[7, 1, 1, "", "apply_threshold_to_image"], [7, 1, 1, "", "draw_bboxes"], [7, 1, 1, "", "is_likely_primary_text_segment"], [7, 1, 1, "", "remove_primary_text_regions"]], "cv_analysis.table_inference": [[13, 1, 1, "", "filter_array"], [13, 1, 1, "", "filter_fp_col_lines"], [13, 1, 1, "", "get_lines_either"], [13, 1, 1, "", "img_bytes_to_array"], [13, 1, 1, "", "infer_lines"], [13, 1, 1, "", "make_gaussian_kernel"], [13, 1, 1, "", "make_gaussian_nonpositive_kernel"], [13, 1, 1, "", "make_quadratic_kernel"], [13, 1, 1, "", "min_avg_for_interval"], [13, 1, 1, "", "save_lines"], [13, 1, 1, "", "save_plot"], [13, 1, 1, "", "search_intervals"], [13, 1, 1, "", "show"], [13, 1, 1, "", "show_multiple"]], "cv_analysis.utils": [[16, 0, 0, "-", "annotate"], [17, 0, 0, "-", "banner"], [18, 0, 0, "-", "connect_rects"], [19, 0, 0, "-", "display"], [20, 0, 0, "-", "draw"], [21, 0, 0, "-", "filters"], [22, 0, 0, "-", "image_extraction"], [23, 0, 0, "-", "open_pdf"], [24, 0, 0, "-", "postprocessing"], [25, 0, 0, "-", "preprocessing"], [26, 0, 0, "-", "structures"], [27, 0, 0, "-", "test_metrics"], [28, 0, 0, "-", "utils"]], "cv_analysis.utils.annotate": [[16, 1, 1, "", "annotate_page"], [16, 1, 1, "", "annotate_pdf"], [16, 1, 1, "", "mirror_on_x_axis"], [16, 1, 1, "", "provide_byte_stream"]], "cv_analysis.utils.banner": [[17, 1, 1, "", "make_art"]], "cv_analysis.utils.connect_rects": [[18, 1, 1, "", "connect_related_rects"], [18, 1, 1, "", "connect_related_rects2"], [18, 1, 1, "", "find_related_rects"], [18, 1, 1, "", "fuse_rects"], [18, 1, 1, "", "has_correct_position1"], [18, 1, 1, "", "is_near_enough"], [18, 1, 1, "", "is_on_same_line"], [18, 1, 1, "", "is_overlapping"], [18, 1, 1, "", "is_related"], [18, 1, 1, "", "rects_not_the_same"]], "cv_analysis.utils.display": [[19, 1, 1, "", "save_image"], [19, 1, 1, "", "show_image"], [19, 1, 1, "", "show_image_cv2"], [19, 1, 1, "", "show_image_mpl"]], "cv_analysis.utils.draw": [[20, 1, 1, "", "draw_contours"], [20, 1, 1, "", "draw_rectangles"]], "cv_analysis.utils.filters": [[21, 1, 1, "", "has_acceptable_format"], [21, 1, 1, "", "is_boxy"], [21, 1, 1, "", "is_filled"], [21, 1, 1, "", "is_large_enough"], [21, 1, 1, "", "is_not_too_large"]], "cv_analysis.utils.image_extraction": [[22, 2, 1, "", "PageInfo"], [22, 1, 1, "", "extract_images_from_pdf"], [22, 1, 1, "", "rescale_to_pdf"], [22, 1, 1, "", "transform_image_coordinates_to_pdf_coordinates"], [22, 1, 1, "", "transform_table_lines_by_page_info"]], "cv_analysis.utils.image_extraction.PageInfo": [[22, 3, 1, "", "dpi"], [22, 3, 1, "", "height"], [22, 3, 1, "", "image_height"], [22, 3, 1, "", "image_width"], [22, 3, 1, "", "page_num"], [22, 3, 1, "", "rotation"], [22, 3, 1, "", "rotation_matrix"], [22, 3, 1, "", "transformation_matrix"], [22, 3, 1, "", "width"]], "cv_analysis.utils.open_pdf": [[23, 1, 1, "", "open_pdf"]], "cv_analysis.utils.postprocessing": [[24, 1, 1, "", "has_no_parent"], [24, 1, 1, "", "remove_included"], [24, 1, 1, "", "remove_isolated"], [24, 1, 1, "", "remove_overlapping"]], "cv_analysis.utils.preprocessing": [[25, 1, 1, "", "page2image"], [25, 1, 1, "", "preprocess_page_array"]], "cv_analysis.utils.structures": [[26, 2, 1, "", "Contour"], [26, 2, 1, "", "Rectangle"]], "cv_analysis.utils.structures.Rectangle": [[26, 4, 1, "", "adjacent"], [26, 4, 1, "", "area"], [26, 4, 1, "", "from_dict_xywh"], [26, 4, 1, "", "from_xywh"], [26, 4, 1, "", "from_xyxy"], [26, 4, 1, "", "includes"], [26, 4, 1, "", "intersection"], [26, 4, 1, "", "iou"], [26, 4, 1, "", "is_included"], [26, 4, 1, "", "json"], [26, 4, 1, "", "json_full"], [26, 4, 1, "", "json_xywh"], [26, 4, 1, "", "json_xyxy"], [26, 4, 1, "", "xywh"], [26, 4, 1, "", "xyxy"]], "cv_analysis.utils.test_metrics": [[27, 1, 1, "", "compute_document_score"], [27, 1, 1, "", "compute_page_iou"], [27, 1, 1, "", "find_max_overlap"]], "cv_analysis.utils.utils": [[28, 1, 1, "", "copy_and_normalize_channels"], [28, 1, 1, "", "npconvert"]]}, "objnames": {"0": ["py", "module", "Python module"], "1": ["py", "function", "Python function"], "2": ["py", "class", "Python class"], "3": ["py", "attribute", "Python attribute"], "4": ["py", "method", "Python method"]}, "objtypes": {"0": "py:module", "1": "py:function", "2": "py:class", "3": "py:attribute", "4": "py:method"}, "terms": {"0": [0, 23], "0cc173c1f97d": 0, "1": 0, "127": 0, "14": 18, "200": 22, "3": 26, "384": 0, "4": 26, "4b11": 0, "4b9531fc": 0, "5": 0, "58349726": 7, "60095520": 21, "60259169": 6, "7": 26, "700": 19, "8001": 0, "8492": 0, "As": 0, "For": 0, "If": 1, "In": 0, "The": 0, "Then": 0, "To": 0, "_": 0, "__": 0, "addit": 0, "adher": 0, "adjac": [15, 26], "after": 0, "all": 0, "also": 0, "amqp": 0, "an": [0, 6, 7], "annot": [2, 15, 20], "annotate_pag": [2, 15, 16], "annotate_pdf": [2, 15, 16], "annotation_dict": 27, "api": 1, "appli": 6, "apply_threshold_to_imag": [2, 4, 7], "ar": 0, "area": [15, 26], "arg": 0, "arr": 13, "arrai": [0, 13], "authent": 0, "avail": 0, "azur": 0, "azure_aml_model_nam": 0, "azure_aml_model_vers": 0, "azure_aml_workspac": 0, "azure_client_id": 0, "azure_client_secret": 0, "azure_config": 0, "azure_resource_group": 0, "azure_subscription_id": 0, "azure_tenant_id": 0, "azureml": 0, "b": [0, 26], "backend": 19, "background": 7, "banner": [2, 15], "base": [0, 22, 26], "base_config": 0, "basic": [0, 1], "bbox": [7, 16, 22], "binari": 6, "black": 7, "bodi": 7, "box": 27, "box_list": 27, "browser": 0, "build": 0, "byte": [13, 16, 22], "c5e4": 0, "calcul": 0, "callabl": 13, "can": 0, "case": 0, "chang": [0, 1], "charact": 0, "chart": 0, "check": 21, "circl": 21, "class": [22, 26], "classmethod": 26, "cnt": [5, 7, 21], "coher": 6, "color": [6, 20], "com": [0, 6, 7, 21], "command": 1, "comput": 0, "compute_document_scor": [2, 15, 27], "compute_page_i": [2, 15, 27], "config": [0, 1, 2], "configur": 1, "connect": 0, "connect_rect": [2, 15], "connect_related_rect": [2, 15, 18], "connect_related_rects2": [2, 15, 18], "consid": 0, "constant": 9, "cont": 21, "contain": 0, "container_nam": 0, "contour": [2, 6, 15, 20, 21, 26], "convent": 0, "convert": 7, "copi": 0, "copy_and_normalize_channel": [2, 15, 28], "cross": 6, "current": 0, "cv_analysi": 1, "d": 1, "data": 0, "de": 0, "default": 0, "defin": [0, 9], "descript": [0, 7], "detail": 0, "detect": 6, "detect_figur": [2, 4, 5], "detect_large_coherent_structur": [2, 4, 6], "dict": [13, 22], "dimens": 0, "discret": 26, "displai": [2, 15], "distinguish": 21, "do": 0, "doc": [0, 1], "docker": 1, "dockerfil": 0, "document": 0, "doe": [0, 26], "don": 0, "done": 0, "download": 0, "dpi": [15, 22], "draw": [2, 15], "draw_bbox": [2, 4, 7], "draw_contour": [2, 15, 20], "draw_rectangl": [2, 15, 20], "dvc": 0, "e": 6, "emb": 0, "embed": 0, "en": 1, "endpoint": 0, "english": 0, "env": 0, "environ": 0, "error": 0, "etc": 0, "eventu": 0, "exampl": [0, 7], "exit": 0, "expect": 6, "experi": 0, "extract": 0, "extract_images_from_pdf": [2, 15, 22], "f": 0, "fals": [0, 20], "fastapi": 0, "field": 0, "figur": [2, 4, 7], "figure_detect": [1, 2], "file": 0, "fill": 21, "filt_sum": 13, "filter": [0, 2, 13, 15], "filter_arrai": [2, 13], "filter_fp_col_lin": [2, 13], "find": 0, "find_max_overlap": [2, 15, 27], "find_related_rect": [2, 15, 18], "first": 0, "first_pag": 23, "fitz": [], "float": [0, 13, 22], "flow": 1, "follow": 0, "format": 26, "from": [0, 7], "from_dict_xywh": [15, 26], "from_xywh": [15, 26], "from_xyxi": [15, 26], "function": 13, "further": 0, "fuse_rect": [2, 15, 18], "g": 6, "german": 0, "get": 0, "get_config": [2, 3], "get_lines_eith": [2, 13], "gitlab": 0, "gitlab_access_token": 0, "gitlab_us": 0, "ground_truth_box": 27, "group": [0, 6], "h": [0, 26], "ha": 0, "has_acceptable_format": [2, 15, 21], "has_correct_position1": [2, 15, 18], "has_no_par": [2, 15, 24], "have": 0, "height": [15, 22], "helm": 0, "help": 0, "here": 0, "hierarchi": [21, 24], "horizont": 13, "host": 0, "how": [6, 7, 21], "html": 1, "http": [0, 1, 6, 7, 21], "i": [0, 21], "ignor": 0, "imag": [5, 6, 7, 19, 20, 28], "image_extract": [2, 15], "image_height": [15, 22], "image_nam": 0, "image_width": [15, 22], "img": 13, "img_byt": 13, "img_bytes_to_arrai": [2, 13], "includ": [0, 15, 26], "indent": 26, "index": 1, "infer_lin": [2, 13], "info": 0, "inform": 0, "input": 0, "input_unsort": 24, "instal": 0, "int": [0, 13, 22], "intersect": [15, 26], "interv": 13, "iou": [15, 26], "is_boxi": [2, 15, 21], "is_fil": [2, 15, 21], "is_includ": [15, 26], "is_large_enough": [2, 15, 21], "is_likely_figur": [2, 4, 5], "is_likely_primary_text_seg": [2, 4, 7], "is_near_enough": [2, 15, 18], "is_not_too_larg": [2, 15, 21], "is_on_same_lin": [2, 15, 18], "is_overlap": [2, 15, 18], "is_rel": [2, 15, 18], "ist": 0, "iter": [18, 24, 26, 27], "json": [0, 15, 26], "json_ful": [15, 26], "json_xywh": [15, 26], "json_xyxi": [15, 26], "just": 0, "kei": 0, "kernel_s": 13, "knecon": 0, "lambda": 13, "languag": 1, "larg": 6, "last_pag": 23, "layout_pars": [1, 2], "length": 0, "level": 0, "like": 1, "line": 13, "line_list": 13, "list": [13, 22, 24], "listen": 0, "load": 0, "local": 1, "locat": [0, 1, 2], "log": 0, "look": 1, "m": 19, "main": 7, "make_art": [2, 15, 17], "make_gaussian_kernel": [2, 13], "make_gaussian_nonpositive_kernel": [2, 13], "make_quadratic_kernel": [2, 13], "manual": 0, "master": 1, "matrix": 22, "max": 0, "max_area": [5, 21], "max_gap": 18, "max_interv": 13, "max_width_to_height_ratio": [5, 21], "maxdim": 19, "maximum": 0, "mean": 7, "ment": 0, "messag": 0, "method": 0, "microsoft": 0, "min": 0, "min_area": [5, 21], "min_avg_for_interv": [2, 13], "min_interv": 13, "minimum": 0, "mirror_on_x_axi": [2, 15, 16], "ml": 1, "mlflow": 0, "model": 1, "model__compute_embed": 0, "modl": 0, "modul": [2, 4, 11, 15], "more": 0, "most": 0, "mriedl": 0, "multi": 0, "name": [0, 13], "ndarrai": [5, 6, 13, 22], "nearbi": 6, "need": 0, "net": 0, "none": [13, 16, 20, 22, 23, 26], "note": 0, "now": 0, "npconvert": [2, 15, 28], "number": 0, "numpi": 13, "ob": 28, "object": [0, 22, 26], "offset": 22, "omit": 0, "open_pdf": [2, 15], "openapi": 0, "opencv": [6, 7, 21], "option": 0, "org": 1, "other": 26, "out": 0, "output_path": 16, "packag": 1, "pad": 13, "pad_value_funct": 13, "page": [1, 16, 25], "page2imag": [2, 15, 25], "page_height": 16, "page_info": 22, "page_num": [15, 22], "pageinfo": [2, 15, 22], "paragraph": [0, 7], "paragraphid": 0, "paramet": [0, 7], "path": [9, 16, 19], "pdf": [16, 23], "pdf_byte": 22, "per": 0, "pipelin": [2, 11], "place": 0, "poetri": 0, "port": 0, "portal": 0, "postprocess": [0, 2, 15], "predict": 16, "preprocess": [0, 2, 15], "preprocess_page_arrai": [2, 15, 25], "present": 0, "primari": 7, "process": 0, "project": 0, "provid": 0, "provide_byte_stream": [2, 15, 16], "py": 0, "python": [0, 6], "question": [6, 7, 21], "queue": 0, "r": 18, "rabbitmq": 1, "ratio": 13, "recent": 0, "rect": [18, 26], "rect1": 18, "rect2": [18, 26], "rect_pair": 18, "rectangl": [2, 15, 20, 24, 26, 27], "rects_not_the_sam": [2, 15, 18], "redaction_detect": [1, 2], "refer": [0, 6, 7, 21], "region": 7, "rel": 9, "remov": 7, "remove_includ": [2, 15, 24], "remove_isol": [2, 15, 24], "remove_overlap": [2, 15, 24], "remove_primary_text_region": [2, 4, 7], "report": 0, "request": 0, "rescale_to_pdf": [2, 15, 22], "research": 0, "resourc": 0, "resourcegroup": 0, "respect": 0, "respons": 0, "responsefilepath": 0, "rest": 1, "restructuredtext": 1, "result": 0, "results_box": 27, "results_dict": 27, "return": [0, 6, 7, 13, 16, 22, 24], "rg": 0, "rm": 0, "root": 9, "rotat": [15, 22], "rotation_matrix": [15, 22], "run": 1, "save": 0, "save_imag": [2, 15, 19], "save_lin": [2, 13], "save_plot": [2, 13], "schema": 0, "script": 0, "sd": 13, "search": 1, "search_interv": [2, 13], "see": 0, "send": 0, "serv": [0, 1], "server": [1, 2], "set": 0, "settings_d": 0, "should": 0, "show": [0, 2, 13], "show_imag": [2, 15, 19], "show_image_cv2": [2, 15, 19], "show_image_mpl": [2, 15, 19], "show_multipl": [2, 13], "signifi": 0, "simplifi": 0, "space": 6, "specif": 0, "specifi": 0, "sphinx": 1, "src": 0, "stackoverflow": [6, 7, 21], "start": 0, "storag": 0, "storageaccount": 0, "store": 0, "str": [13, 16], "string": 0, "structur": [0, 2, 6, 15], "structureobject": 0, "subject": 0, "submodul": 1, "subpackag": 1, "subscript": 0, "subword": 0, "sum_filt": 13, "support": 0, "t": 0, "taa": 0, "taassarac": 0, "table_arrai": 13, "table_infer": [1, 2], "table_pars": [1, 2], "target": 0, "targetfilepath": 0, "templat": 0, "test": 0, "test_metr": [2, 15], "text": [0, 2, 4], "thi": 0, "thing": 1, "threshold": 6, "titl": 13, "tol": 26, "toler": 26, "toml": 0, "track": 0, "train": 0, "transform_image_coordinates_to_pdf_coordin": [2, 15, 22], "transform_table_lines_by_page_info": [2, 15, 22], "transformation_matrix": [15, 22], "true": [0, 13, 24, 26], "tupl": [13, 18, 22], "two": 0, "type": [13, 16, 22, 24], "under": 0, "unfil": 21, "up": 0, "upload": 1, "us": 0, "usag": 1, "util": [1, 2], "variabl": 0, "version": 0, "via": 0, "view": 0, "visual_log": [2, 15], "vlp_output": 22, "w": [0, 26], "when": 0, "where": 0, "whether": [0, 21], "white": 7, "whole": 0, "width": [15, 22], "work": 0, "www": 1, "x1": 26, "x2": 26, "xywh": [15, 26], "xywh_dict": 26, "xywh_tupl": 26, "xyxi": [15, 26], "xyxy_tupl": 26, "y1": 26, "y2": 26, "you": [0, 1], "zebra": 6}, "titles": ["Keyword-Service", "Welcome to Keyword Extraction Service documentation!", "cv_analysis package", "cv_analysis.config module", "cv_analysis.figure_detection package", "cv_analysis.figure_detection.figure_detection module", "cv_analysis.figure_detection.figures module", "cv_analysis.figure_detection.text module", "cv_analysis.layout_parsing module", "cv_analysis.locations module", "cv_analysis.redaction_detection module", "cv_analysis.server package", "cv_analysis.server.pipeline module", "cv_analysis.table_inference module", "cv_analysis.table_parsing module", "cv_analysis.utils package", "cv_analysis.utils.annotate module", "cv_analysis.utils.banner module", "cv_analysis.utils.connect_rects module", "cv_analysis.utils.display module", "cv_analysis.utils.draw module", "cv_analysis.utils.filters module", "cv_analysis.utils.image_extraction module", "cv_analysis.utils.open_pdf module", "cv_analysis.utils.postprocessing module", "cv_analysis.utils.preprocessing module", "cv_analysis.utils.structures module", "cv_analysis.utils.test_metrics module", "cv_analysis.utils.utils module", "cv_analysis.utils.visual_logging module", "serve module"], "titleterms": {"annot": 16, "api": 0, "banner": 17, "command": 0, "config": 3, "configur": 0, "connect_rect": 18, "content": 1, "cv_analysi": [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29], "displai": 19, "docker": 0, "document": 1, "draw": 20, "extract": 1, "figur": 6, "figure_detect": [4, 5, 6, 7], "filter": 21, "flow": 0, "image_extract": 22, "indic": 1, "keyword": [0, 1], "languag": 0, "layout_pars": 8, "local": 0, "locat": 9, "ml": 0, "model": 0, "modul": [1, 3, 5, 6, 7, 8, 9, 10, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30], "open_pdf": 23, "packag": [2, 4, 11, 15], "pipelin": 12, "postprocess": 24, "preprocess": 25, "rabbitmq": 0, "readm": 1, "redaction_detect": 10, "rest": 0, "run": 0, "serv": 30, "server": [11, 12], "servic": [0, 1], "structur": 26, "submodul": [2, 4, 11, 15], "subpackag": 2, "tabl": 1, "table_infer": 13, "table_pars": 14, "test_metr": 27, "text": 7, "upload": 0, "usag": 0, "util": [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29], "visual_log": 29, "welcom": 1}}) \ No newline at end of file +Search.setIndex({"alltitles": {"API": [[0, "api"]], "As a CLI Tool": [[0, "as-a-cli-tool"]], "As an API": [[0, "as-an-api"]], "Building": [[0, "building"]], "Figure Detection": [[0, "figure-detection"]], "Indices and tables": [[1, "indices-and-tables"]], "Installation": [[0, "installation"]], "Layout Parsing": [[0, "layout-parsing"]], "Modules": [[1, null]], "README": [[1, null]], "Redaction Detection (API)": [[0, "redaction-detection-api"]], "Redaction Detection (CLI)": [[0, "redaction-detection-cli"]], "Running as a service": [[0, "running-as-a-service"]], "Submodules": [[2, "submodules"], [4, "submodules"], [11, "submodules"], [15, "submodules"]], "Subpackages": [[2, "subpackages"]], "Table Parsing": [[0, "table-parsing"]], "Table of Contents": [[1, "table-of-contents"]], "Usage": [[0, "usage"]], "Usage (service)": [[0, "usage-service"]], "Welcome to CV Analysis Service documentation!": [[1, "welcome-to-cv-analysis-service-documentation"]], "cv-analysis - Visual (CV-Based) Document Parsing": [[0, "cv-analysis-visual-cv-based-document-parsing"]], "cv_analysis package": [[2, "module-cv_analysis"]], "cv_analysis.config module": [[3, "module-cv_analysis.config"]], "cv_analysis.figure_detection package": [[4, "module-cv_analysis.figure_detection"]], "cv_analysis.figure_detection.figure_detection module": [[5, "module-cv_analysis.figure_detection.figure_detection"]], "cv_analysis.figure_detection.figures module": [[6, "module-cv_analysis.figure_detection.figures"]], "cv_analysis.figure_detection.text module": [[7, "module-cv_analysis.figure_detection.text"]], "cv_analysis.layout_parsing module": [[8, "cv-analysis-layout-parsing-module"]], "cv_analysis.locations module": [[9, "module-cv_analysis.locations"]], "cv_analysis.redaction_detection module": [[10, "cv-analysis-redaction-detection-module"]], "cv_analysis.server package": [[11, "module-cv_analysis.server"]], "cv_analysis.server.pipeline module": [[12, "cv-analysis-server-pipeline-module"]], "cv_analysis.table_inference module": [[13, "module-cv_analysis.table_inference"]], "cv_analysis.table_parsing module": [[14, "cv-analysis-table-parsing-module"]], "cv_analysis.utils package": [[15, "module-cv_analysis.utils"]], "cv_analysis.utils.annotate module": [[16, "module-cv_analysis.utils.annotate"]], "cv_analysis.utils.banner module": [[17, "module-cv_analysis.utils.banner"]], "cv_analysis.utils.connect_rects module": [[18, "module-cv_analysis.utils.connect_rects"]], "cv_analysis.utils.display module": [[19, "module-cv_analysis.utils.display"]], "cv_analysis.utils.draw module": [[20, "module-cv_analysis.utils.draw"]], "cv_analysis.utils.filters module": [[21, "module-cv_analysis.utils.filters"]], "cv_analysis.utils.image_extraction module": [[22, "module-cv_analysis.utils.image_extraction"]], "cv_analysis.utils.open_pdf module": [[23, "module-cv_analysis.utils.open_pdf"]], "cv_analysis.utils.postprocessing module": [[24, "module-cv_analysis.utils.postprocessing"]], "cv_analysis.utils.preprocessing module": [[25, "module-cv_analysis.utils.preprocessing"]], "cv_analysis.utils.structures module": [[26, "module-cv_analysis.utils.structures"]], "cv_analysis.utils.test_metrics module": [[27, "module-cv_analysis.utils.test_metrics"]], "cv_analysis.utils.utils module": [[28, "module-cv_analysis.utils.utils"]], "cv_analysis.utils.visual_logging module": [[29, "cv-analysis-utils-visual-logging-module"]], "serve module": [[30, "module-serve"]]}, "docnames": ["README", "index", "modules/cv_analysis", "modules/cv_analysis.config", "modules/cv_analysis.figure_detection", "modules/cv_analysis.figure_detection.figure_detection", "modules/cv_analysis.figure_detection.figures", "modules/cv_analysis.figure_detection.text", "modules/cv_analysis.layout_parsing", "modules/cv_analysis.locations", "modules/cv_analysis.redaction_detection", "modules/cv_analysis.server", "modules/cv_analysis.server.pipeline", "modules/cv_analysis.table_inference", "modules/cv_analysis.table_parsing", "modules/cv_analysis.utils", "modules/cv_analysis.utils.annotate", "modules/cv_analysis.utils.banner", "modules/cv_analysis.utils.connect_rects", "modules/cv_analysis.utils.display", "modules/cv_analysis.utils.draw", "modules/cv_analysis.utils.filters", "modules/cv_analysis.utils.image_extraction", "modules/cv_analysis.utils.open_pdf", "modules/cv_analysis.utils.postprocessing", "modules/cv_analysis.utils.preprocessing", "modules/cv_analysis.utils.structures", "modules/cv_analysis.utils.test_metrics", "modules/cv_analysis.utils.utils", "modules/cv_analysis.utils.visual_logging", "modules/serve"], "envversion": {"sphinx": 61, "sphinx.domains.c": 3, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 9, "sphinx.domains.index": 1, "sphinx.domains.javascript": 3, "sphinx.domains.math": 2, "sphinx.domains.python": 4, "sphinx.domains.rst": 2, "sphinx.domains.std": 2}, "filenames": ["README.md", "index.rst", "modules/cv_analysis.rst", "modules/cv_analysis.config.rst", "modules/cv_analysis.figure_detection.rst", "modules/cv_analysis.figure_detection.figure_detection.rst", "modules/cv_analysis.figure_detection.figures.rst", "modules/cv_analysis.figure_detection.text.rst", "modules/cv_analysis.layout_parsing.rst", "modules/cv_analysis.locations.rst", "modules/cv_analysis.redaction_detection.rst", "modules/cv_analysis.server.rst", "modules/cv_analysis.server.pipeline.rst", "modules/cv_analysis.table_inference.rst", "modules/cv_analysis.table_parsing.rst", "modules/cv_analysis.utils.rst", "modules/cv_analysis.utils.annotate.rst", "modules/cv_analysis.utils.banner.rst", "modules/cv_analysis.utils.connect_rects.rst", "modules/cv_analysis.utils.display.rst", "modules/cv_analysis.utils.draw.rst", "modules/cv_analysis.utils.filters.rst", "modules/cv_analysis.utils.image_extraction.rst", "modules/cv_analysis.utils.open_pdf.rst", "modules/cv_analysis.utils.postprocessing.rst", "modules/cv_analysis.utils.preprocessing.rst", "modules/cv_analysis.utils.structures.rst", "modules/cv_analysis.utils.test_metrics.rst", "modules/cv_analysis.utils.utils.rst", "modules/cv_analysis.utils.visual_logging.rst", "modules/serve.rst"], "indexentries": {"adjacent() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.adjacent", false]], "annotate_page() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.annotate_page", false]], "annotate_pdf() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.annotate_pdf", false]], "apply_threshold_to_image() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.apply_threshold_to_image", false]], "area() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.area", false]], "compute_document_score() (in module cv_analysis.utils.test_metrics)": [[27, "cv_analysis.utils.test_metrics.compute_document_score", false]], "compute_page_iou() (in module cv_analysis.utils.test_metrics)": [[27, "cv_analysis.utils.test_metrics.compute_page_iou", false]], "connect_related_rects() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.connect_related_rects", false]], "connect_related_rects2() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.connect_related_rects2", false]], "contour (class in cv_analysis.utils.structures)": [[26, "cv_analysis.utils.structures.Contour", false]], "copy_and_normalize_channels() (in module cv_analysis.utils.utils)": [[28, "cv_analysis.utils.utils.copy_and_normalize_channels", false]], "cv_analysis": [[2, "module-cv_analysis", false]], "cv_analysis.config": [[3, "module-cv_analysis.config", false]], "cv_analysis.figure_detection": [[4, "module-cv_analysis.figure_detection", false]], "cv_analysis.figure_detection.figure_detection": [[5, "module-cv_analysis.figure_detection.figure_detection", false]], "cv_analysis.figure_detection.figures": [[6, "module-cv_analysis.figure_detection.figures", false]], "cv_analysis.figure_detection.text": [[7, "module-cv_analysis.figure_detection.text", false]], "cv_analysis.locations": [[9, "module-cv_analysis.locations", false]], "cv_analysis.server": [[11, "module-cv_analysis.server", false]], "cv_analysis.table_inference": [[13, "module-cv_analysis.table_inference", false]], "cv_analysis.utils": [[15, "module-cv_analysis.utils", false]], "cv_analysis.utils.annotate": [[16, "module-cv_analysis.utils.annotate", false]], "cv_analysis.utils.banner": [[17, "module-cv_analysis.utils.banner", false]], "cv_analysis.utils.connect_rects": [[18, "module-cv_analysis.utils.connect_rects", false]], "cv_analysis.utils.display": [[19, "module-cv_analysis.utils.display", false]], "cv_analysis.utils.draw": [[20, "module-cv_analysis.utils.draw", false]], "cv_analysis.utils.filters": [[21, "module-cv_analysis.utils.filters", false]], "cv_analysis.utils.image_extraction": [[22, "module-cv_analysis.utils.image_extraction", false]], "cv_analysis.utils.open_pdf": [[23, "module-cv_analysis.utils.open_pdf", false]], "cv_analysis.utils.postprocessing": [[24, "module-cv_analysis.utils.postprocessing", false]], "cv_analysis.utils.preprocessing": [[25, "module-cv_analysis.utils.preprocessing", false]], "cv_analysis.utils.structures": [[26, "module-cv_analysis.utils.structures", false]], "cv_analysis.utils.test_metrics": [[27, "module-cv_analysis.utils.test_metrics", false]], "cv_analysis.utils.utils": [[28, "module-cv_analysis.utils.utils", false]], "detect_figures() (in module cv_analysis.figure_detection.figure_detection)": [[5, "cv_analysis.figure_detection.figure_detection.detect_figures", false]], "detect_large_coherent_structures() (in module cv_analysis.figure_detection.figures)": [[6, "cv_analysis.figure_detection.figures.detect_large_coherent_structures", false]], "dpi (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.dpi", false]], "draw_bboxes() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.draw_bboxes", false]], "draw_contours() (in module cv_analysis.utils.draw)": [[20, "cv_analysis.utils.draw.draw_contours", false]], "draw_rectangles() (in module cv_analysis.utils.draw)": [[20, "cv_analysis.utils.draw.draw_rectangles", false]], "extract_images_from_pdf() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.extract_images_from_pdf", false]], "filter_array() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.filter_array", false]], "filter_fp_col_lines() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.filter_fp_col_lines", false]], "find_max_overlap() (in module cv_analysis.utils.test_metrics)": [[27, "cv_analysis.utils.test_metrics.find_max_overlap", false]], "find_related_rects() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.find_related_rects", false]], "from_dict_xywh() (cv_analysis.utils.structures.rectangle class method)": [[26, "cv_analysis.utils.structures.Rectangle.from_dict_xywh", false]], "from_xywh() (cv_analysis.utils.structures.rectangle class method)": [[26, "cv_analysis.utils.structures.Rectangle.from_xywh", false]], "from_xyxy() (cv_analysis.utils.structures.rectangle class method)": [[26, "cv_analysis.utils.structures.Rectangle.from_xyxy", false]], "fuse_rects() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.fuse_rects", false]], "get_config() (in module cv_analysis.config)": [[3, "cv_analysis.config.get_config", false]], "get_lines_either() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.get_lines_either", false]], "has_acceptable_format() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.has_acceptable_format", false]], "has_correct_position1() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.has_correct_position1", false]], "has_no_parent() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.has_no_parent", false]], "height (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.height", false]], "image_height (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.image_height", false]], "image_width (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.image_width", false]], "img_bytes_to_array() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.img_bytes_to_array", false]], "includes() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.includes", false]], "infer_lines() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.infer_lines", false]], "intersection() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.intersection", false]], "iou() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.iou", false]], "is_boxy() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_boxy", false]], "is_filled() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_filled", false]], "is_included() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.is_included", false]], "is_large_enough() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_large_enough", false]], "is_likely_figure() (in module cv_analysis.figure_detection.figure_detection)": [[5, "cv_analysis.figure_detection.figure_detection.is_likely_figure", false]], "is_likely_primary_text_segment() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.is_likely_primary_text_segment", false]], "is_near_enough() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_near_enough", false]], "is_not_too_large() (in module cv_analysis.utils.filters)": [[21, "cv_analysis.utils.filters.is_not_too_large", false]], "is_on_same_line() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_on_same_line", false]], "is_overlapping() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_overlapping", false]], "is_related() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.is_related", false]], "json() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json", false]], "json_full() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json_full", false]], "json_xywh() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json_xywh", false]], "json_xyxy() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.json_xyxy", false]], "make_art() (in module cv_analysis.utils.banner)": [[17, "cv_analysis.utils.banner.make_art", false]], "make_gaussian_kernel() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.make_gaussian_kernel", false]], "make_gaussian_nonpositive_kernel() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.make_gaussian_nonpositive_kernel", false]], "make_quadratic_kernel() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.make_quadratic_kernel", false]], "min_avg_for_interval() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.min_avg_for_interval", false]], "mirror_on_x_axis() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.mirror_on_x_axis", false]], "module": [[2, "module-cv_analysis", false], [3, "module-cv_analysis.config", false], [4, "module-cv_analysis.figure_detection", false], [5, "module-cv_analysis.figure_detection.figure_detection", false], [6, "module-cv_analysis.figure_detection.figures", false], [7, "module-cv_analysis.figure_detection.text", false], [9, "module-cv_analysis.locations", false], [11, "module-cv_analysis.server", false], [13, "module-cv_analysis.table_inference", false], [15, "module-cv_analysis.utils", false], [16, "module-cv_analysis.utils.annotate", false], [17, "module-cv_analysis.utils.banner", false], [18, "module-cv_analysis.utils.connect_rects", false], [19, "module-cv_analysis.utils.display", false], [20, "module-cv_analysis.utils.draw", false], [21, "module-cv_analysis.utils.filters", false], [22, "module-cv_analysis.utils.image_extraction", false], [23, "module-cv_analysis.utils.open_pdf", false], [24, "module-cv_analysis.utils.postprocessing", false], [25, "module-cv_analysis.utils.preprocessing", false], [26, "module-cv_analysis.utils.structures", false], [27, "module-cv_analysis.utils.test_metrics", false], [28, "module-cv_analysis.utils.utils", false], [30, "module-serve", false]], "npconvert() (in module cv_analysis.utils.utils)": [[28, "cv_analysis.utils.utils.npconvert", false]], "open_pdf() (in module cv_analysis.utils.open_pdf)": [[23, "cv_analysis.utils.open_pdf.open_pdf", false]], "page2image() (in module cv_analysis.utils.preprocessing)": [[25, "cv_analysis.utils.preprocessing.page2image", false]], "page_num (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.page_num", false]], "pageinfo (class in cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.PageInfo", false]], "preprocess_page_array() (in module cv_analysis.utils.preprocessing)": [[25, "cv_analysis.utils.preprocessing.preprocess_page_array", false]], "provide_byte_stream() (in module cv_analysis.utils.annotate)": [[16, "cv_analysis.utils.annotate.provide_byte_stream", false]], "rectangle (class in cv_analysis.utils.structures)": [[26, "cv_analysis.utils.structures.Rectangle", false]], "rects_not_the_same() (in module cv_analysis.utils.connect_rects)": [[18, "cv_analysis.utils.connect_rects.rects_not_the_same", false]], "remove_included() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.remove_included", false]], "remove_isolated() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.remove_isolated", false]], "remove_overlapping() (in module cv_analysis.utils.postprocessing)": [[24, "cv_analysis.utils.postprocessing.remove_overlapping", false]], "remove_primary_text_regions() (in module cv_analysis.figure_detection.text)": [[7, "cv_analysis.figure_detection.text.remove_primary_text_regions", false]], "rescale_to_pdf() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.rescale_to_pdf", false]], "rotation (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.rotation", false]], "rotation_matrix (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.rotation_matrix", false]], "save_image() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.save_image", false]], "save_lines() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.save_lines", false]], "save_plot() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.save_plot", false]], "search_intervals() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.search_intervals", false]], "serve": [[30, "module-serve", false]], "show() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.show", false]], "show_image() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.show_image", false]], "show_image_cv2() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.show_image_cv2", false]], "show_image_mpl() (in module cv_analysis.utils.display)": [[19, "cv_analysis.utils.display.show_image_mpl", false]], "show_multiple() (in module cv_analysis.table_inference)": [[13, "cv_analysis.table_inference.show_multiple", false]], "transform_image_coordinates_to_pdf_coordinates() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.transform_image_coordinates_to_pdf_coordinates", false]], "transform_table_lines_by_page_info() (in module cv_analysis.utils.image_extraction)": [[22, "cv_analysis.utils.image_extraction.transform_table_lines_by_page_info", false]], "transformation_matrix (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.transformation_matrix", false]], "width (cv_analysis.utils.image_extraction.pageinfo attribute)": [[22, "cv_analysis.utils.image_extraction.PageInfo.width", false]], "xywh() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.xywh", false]], "xyxy() (cv_analysis.utils.structures.rectangle method)": [[26, "cv_analysis.utils.structures.Rectangle.xyxy", false]]}, "objects": {"": [[2, 0, 0, "-", "cv_analysis"], [30, 0, 0, "-", "serve"]], "cv_analysis": [[3, 0, 0, "-", "config"], [4, 0, 0, "-", "figure_detection"], [9, 0, 0, "-", "locations"], [11, 0, 0, "-", "server"], [13, 0, 0, "-", "table_inference"], [15, 0, 0, "-", "utils"]], "cv_analysis.config": [[3, 1, 1, "", "get_config"]], "cv_analysis.figure_detection": [[5, 0, 0, "-", "figure_detection"], [6, 0, 0, "-", "figures"], [7, 0, 0, "-", "text"]], "cv_analysis.figure_detection.figure_detection": [[5, 1, 1, "", "detect_figures"], [5, 1, 1, "", "is_likely_figure"]], "cv_analysis.figure_detection.figures": [[6, 1, 1, "", "detect_large_coherent_structures"]], "cv_analysis.figure_detection.text": [[7, 1, 1, "", "apply_threshold_to_image"], [7, 1, 1, "", "draw_bboxes"], [7, 1, 1, "", "is_likely_primary_text_segment"], [7, 1, 1, "", "remove_primary_text_regions"]], "cv_analysis.table_inference": [[13, 1, 1, "", "filter_array"], [13, 1, 1, "", "filter_fp_col_lines"], [13, 1, 1, "", "get_lines_either"], [13, 1, 1, "", "img_bytes_to_array"], [13, 1, 1, "", "infer_lines"], [13, 1, 1, "", "make_gaussian_kernel"], [13, 1, 1, "", "make_gaussian_nonpositive_kernel"], [13, 1, 1, "", "make_quadratic_kernel"], [13, 1, 1, "", "min_avg_for_interval"], [13, 1, 1, "", "save_lines"], [13, 1, 1, "", "save_plot"], [13, 1, 1, "", "search_intervals"], [13, 1, 1, "", "show"], [13, 1, 1, "", "show_multiple"]], "cv_analysis.utils": [[16, 0, 0, "-", "annotate"], [17, 0, 0, "-", "banner"], [18, 0, 0, "-", "connect_rects"], [19, 0, 0, "-", "display"], [20, 0, 0, "-", "draw"], [21, 0, 0, "-", "filters"], [22, 0, 0, "-", "image_extraction"], [23, 0, 0, "-", "open_pdf"], [24, 0, 0, "-", "postprocessing"], [25, 0, 0, "-", "preprocessing"], [26, 0, 0, "-", "structures"], [27, 0, 0, "-", "test_metrics"], [28, 0, 0, "-", "utils"]], "cv_analysis.utils.annotate": [[16, 1, 1, "", "annotate_page"], [16, 1, 1, "", "annotate_pdf"], [16, 1, 1, "", "mirror_on_x_axis"], [16, 1, 1, "", "provide_byte_stream"]], "cv_analysis.utils.banner": [[17, 1, 1, "", "make_art"]], "cv_analysis.utils.connect_rects": [[18, 1, 1, "", "connect_related_rects"], [18, 1, 1, "", "connect_related_rects2"], [18, 1, 1, "", "find_related_rects"], [18, 1, 1, "", "fuse_rects"], [18, 1, 1, "", "has_correct_position1"], [18, 1, 1, "", "is_near_enough"], [18, 1, 1, "", "is_on_same_line"], [18, 1, 1, "", "is_overlapping"], [18, 1, 1, "", "is_related"], [18, 1, 1, "", "rects_not_the_same"]], "cv_analysis.utils.display": [[19, 1, 1, "", "save_image"], [19, 1, 1, "", "show_image"], [19, 1, 1, "", "show_image_cv2"], [19, 1, 1, "", "show_image_mpl"]], "cv_analysis.utils.draw": [[20, 1, 1, "", "draw_contours"], [20, 1, 1, "", "draw_rectangles"]], "cv_analysis.utils.filters": [[21, 1, 1, "", "has_acceptable_format"], [21, 1, 1, "", "is_boxy"], [21, 1, 1, "", "is_filled"], [21, 1, 1, "", "is_large_enough"], [21, 1, 1, "", "is_not_too_large"]], "cv_analysis.utils.image_extraction": [[22, 2, 1, "", "PageInfo"], [22, 1, 1, "", "extract_images_from_pdf"], [22, 1, 1, "", "rescale_to_pdf"], [22, 1, 1, "", "transform_image_coordinates_to_pdf_coordinates"], [22, 1, 1, "", "transform_table_lines_by_page_info"]], "cv_analysis.utils.image_extraction.PageInfo": [[22, 3, 1, "", "dpi"], [22, 3, 1, "", "height"], [22, 3, 1, "", "image_height"], [22, 3, 1, "", "image_width"], [22, 3, 1, "", "page_num"], [22, 3, 1, "", "rotation"], [22, 3, 1, "", "rotation_matrix"], [22, 3, 1, "", "transformation_matrix"], [22, 3, 1, "", "width"]], "cv_analysis.utils.open_pdf": [[23, 1, 1, "", "open_pdf"]], "cv_analysis.utils.postprocessing": [[24, 1, 1, "", "has_no_parent"], [24, 1, 1, "", "remove_included"], [24, 1, 1, "", "remove_isolated"], [24, 1, 1, "", "remove_overlapping"]], "cv_analysis.utils.preprocessing": [[25, 1, 1, "", "page2image"], [25, 1, 1, "", "preprocess_page_array"]], "cv_analysis.utils.structures": [[26, 2, 1, "", "Contour"], [26, 2, 1, "", "Rectangle"]], "cv_analysis.utils.structures.Rectangle": [[26, 4, 1, "", "adjacent"], [26, 4, 1, "", "area"], [26, 4, 1, "", "from_dict_xywh"], [26, 4, 1, "", "from_xywh"], [26, 4, 1, "", "from_xyxy"], [26, 4, 1, "", "includes"], [26, 4, 1, "", "intersection"], [26, 4, 1, "", "iou"], [26, 4, 1, "", "is_included"], [26, 4, 1, "", "json"], [26, 4, 1, "", "json_full"], [26, 4, 1, "", "json_xywh"], [26, 4, 1, "", "json_xyxy"], [26, 4, 1, "", "xywh"], [26, 4, 1, "", "xyxy"]], "cv_analysis.utils.test_metrics": [[27, 1, 1, "", "compute_document_score"], [27, 1, 1, "", "compute_page_iou"], [27, 1, 1, "", "find_max_overlap"]], "cv_analysis.utils.utils": [[28, 1, 1, "", "copy_and_normalize_channels"], [28, 1, 1, "", "npconvert"]]}, "objnames": {"0": ["py", "module", "Python module"], "1": ["py", "function", "Python function"], "2": ["py", "class", "Python class"], "3": ["py", "attribute", "Python attribute"], "4": ["py", "method", "Python method"]}, "objtypes": {"0": "py:module", "1": "py:function", "2": "py:class", "3": "py:attribute", "4": "py:method"}, "terms": {"0": [0, 23], "0246": 0, "0585": 0, "0ad5b774ac84": 0, "1": 0, "1399": 0, "14": 18, "1414": 0, "16": 0, "2": 0, "200": 22, "2222": 0, "247": 0, "2b10c1a2": 0, "3": [0, 26], "3407": 0, "393c": 0, "4": 26, "4fca": 0, "55": 0, "558": 0, "5602": 0, "58349726": 7, "598": 0, "60095520": 21, "60259169": 6, "693": 0, "7": [0, 26], "700": 19, "As": 1, "If": 1, "The": 0, "_": 0, "absolut": 0, "activ": 0, "adjac": [15, 26], "all": 0, "an": [1, 6, 7], "annot": [0, 2, 15, 20], "annotate_pag": [2, 15, 16], "annotate_pdf": [2, 15, 16], "annotation_dict": 27, "api": 1, "appli": 6, "apply_threshold_to_imag": [2, 4, 7], "approach": 0, "area": [15, 26], "arg": 0, "arr": 13, "arrai": [0, 13], "b": 26, "b9e3": 0, "backend": 19, "background": 7, "banner": [2, 15], "base": [1, 22, 26], "base_root": 0, "bash": 0, "basic": 1, "bbox": [0, 7, 16, 22], "been": 0, "below": 0, "bin": 0, "binari": 6, "black": [0, 7], "bodi": 7, "box": 27, "box_list": 27, "build": 1, "byte": [13, 16, 22], "callabl": 13, "can": 0, "cd": 0, "cell": 0, "chang": 1, "check": 21, "circl": 21, "class": [22, 26], "classmethod": 26, "cli": 1, "client_mock": 0, "clone": 0, "cnt": [5, 7, 21], "coher": 6, "collect": 0, "color": [6, 20], "com": [0, 6, 7, 21], "comput": 0, "compute_document_scor": [2, 15, 27], "compute_page_i": [2, 15, 27], "config": [1, 2], "connect_rect": [2, 15], "connect_related_rect": [2, 15, 18], "connect_related_rects2": [2, 15, 18], "constant": 9, "cont": 21, "contour": [2, 6, 15, 20, 21, 26], "convert": 7, "convert_from_path": 0, "copy_and_normalize_channel": [2, 15, 28], "core": 0, "cross": 6, "cv_analysi": [0, 1], "d": 1, "data": 0, "defin": 9, "depend": 0, "descript": 7, "detect": [1, 6], "detect_figur": [2, 4, 5], "detect_large_coherent_structur": [2, 4, 6], "dict": [13, 22], "discret": 26, "displai": [2, 15], "distinguish": 21, "doc": 1, "docker": 0, "dockerfil": 0, "doe": 26, "dpi": [15, 22], "draw": [2, 15], "draw_bbox": [2, 4, 7], "draw_contour": [2, 15, 20], "draw_rectangl": [2, 15, 20], "dvc": 0, "e": [0, 6], "each": 0, "element": 0, "en": 1, "env": 0, "exampl": 7, "expect": 6, "extract_images_from_pdf": [2, 15, 22], "f": 0, "fals": 20, "featur": 0, "field": 0, "figur": [1, 2, 4, 7], "figure_detect": [1, 2], "file": 0, "fill": [0, 21], "filt_sum": 13, "filter": [2, 13, 15], "filter_arrai": [2, 13], "filter_fp_col_lin": [2, 13], "find": 0, "find_max_overlap": [2, 15, 27], "find_redact": 0, "find_related_rect": [2, 15, 18], "first_pag": [0, 23], "float": [13, 22], "follow": 0, "format": 26, "from": [0, 7], "from_dict_xywh": [15, 26], "from_xywh": [15, 26], "from_xyxi": [15, 26], "function": [0, 13], "fuse_rect": [2, 15, 18], "g": 6, "gener": 0, "get_config": [2, 3], "get_lines_eith": [2, 13], "git": 0, "green": 0, "ground_truth_box": 27, "group": 6, "h": 26, "ha": 0, "has_acceptable_format": [2, 15, 21], "has_correct_position1": [2, 15, 18], "has_no_par": [2, 15, 24], "head": 0, "height": [0, 15, 22], "hierarchi": [21, 24], "horizont": 13, "host": 0, "hot": 0, "how": [6, 7, 21], "html": 1, "http": [1, 6, 7, 21], "i": [0, 21], "imag": [0, 5, 6, 7, 19, 20, 28], "image_extract": [2, 15], "image_height": [15, 22], "image_width": [15, 22], "imageinfo": 0, "img": 13, "img_byt": 13, "img_bytes_to_arrai": [2, 13], "implement": 0, "import": 0, "includ": [15, 26], "indent": 26, "index": 1, "individu": 0, "infer_lin": [2, 13], "input": 0, "input_unsort": 24, "instal": 1, "int": [13, 22], "intersect": [15, 26], "interv": 13, "iou": [15, 26], "iqser": 0, "is_boxi": [2, 15, 21], "is_fil": [2, 15, 21], "is_includ": [15, 26], "is_large_enough": [2, 15, 21], "is_likely_figur": [2, 4, 5], "is_likely_primary_text_seg": [2, 4, 7], "is_near_enough": [2, 15, 18], "is_not_too_larg": [2, 15, 21], "is_on_same_lin": [2, 15, 18], "is_overlap": [2, 15, 18], "is_rel": [2, 15, 18], "iter": [18, 24, 26, 27], "json": [15, 26], "json_ful": [15, 26], "json_xywh": [15, 26], "json_xyxi": [15, 26], "kernel_s": 13, "kind": 0, "label": 0, "lambda": 13, "larg": 6, "last_pag": [0, 23], "layout": 1, "layout_pars": [1, 2], "like": 1, "line": 13, "line_list": 13, "list": [13, 22, 24], "locat": [1, 2], "look": 1, "m": [0, 19], "main": 7, "make_art": [2, 15, 17], "make_gaussian_kernel": [2, 13], "make_gaussian_nonpositive_kernel": [2, 13], "make_quadratic_kernel": [2, 13], "master": 1, "matrix": 22, "max_area": [5, 21], "max_gap": 18, "max_interv": 13, "max_width_to_height_ratio": [5, 21], "maxdim": 19, "mean": 7, "messag": 0, "min_area": [5, 21], "min_avg_for_interv": [2, 13], "min_interv": 13, "mirror_on_x_axi": [2, 15, 16], "miss": 0, "modul": [0, 2, 4, 11, 15], "name": 13, "ndarrai": [5, 6, 13, 22], "nearbi": 6, "net": 0, "none": [13, 16, 20, 22, 23, 26], "np": 0, "npconvert": [2, 15, 28], "numpi": [0, 13], "ob": 28, "object": [22, 26], "offset": 22, "open_pdf": [2, 15], "opencv": [6, 7, 21], "oper": 0, "org": 1, "other": 26, "outlin": 0, "output_path": 16, "packag": 1, "pad": 13, "pad_value_funct": 13, "page": [0, 1, 16, 25], "page2imag": [2, 15, 25], "page_height": 16, "page_index": 0, "page_info": 22, "page_num": [15, 22], "pageinfo": [2, 15, 22], "pagenum": 0, "paragraph": [0, 7], "paramet": 7, "pars": 1, "path": [0, 9, 16, 19], "pdf": [0, 16, 23], "pdf2imag": 0, "pdf_byte": 22, "pdf_path": 0, "pip": 0, "pipelin": [2, 11], "point": 0, "postprocess": [2, 15], "predict": 16, "preprocess": [2, 15], "preprocess_page_arrai": [2, 15, 25], "previou": 0, "primari": 7, "provid": 0, "provide_byte_stream": [2, 15, 16], "pull": 0, "py": 0, "python": [0, 6], "question": [6, 7, 21], "r": [0, 18], "ratio": 13, "rect": [18, 26], "rect1": 18, "rect2": [18, 26], "rect_pair": 18, "rectangl": [0, 2, 15, 20, 24, 26, 27], "rects_not_the_sam": [2, 15, 18], "redact": 1, "redaction_contour": 0, "redaction_detect": [0, 1, 2], "refer": [6, 7, 21], "region": 7, "rel": 9, "remov": 7, "remove_includ": [2, 15, 24], "remove_isol": [2, 15, 24], "remove_overlap": [2, 15, 24], "remove_primary_text_region": [2, 4, 7], "repositori": 0, "requir": 0, "rescale_to_pdf": [2, 15, 22], "respons": 0, "responsefilepath": 0, "restructuredtext": 1, "results_box": 27, "results_dict": 27, "return": [0, 6, 7, 13, 16, 22, 24], "rm": 0, "root": 9, "rotat": [15, 22], "rotation_matrix": [15, 22], "rr": 0, "run": 1, "save_imag": [2, 15, 19], "save_lin": [2, 13], "save_plot": [2, 13], "script": 0, "sd": 13, "search": 1, "search_interv": [2, 13], "segment": 0, "serv": 1, "server": [1, 2], "setup": 0, "sh": 0, "shell": 0, "show": [0, 2, 13], "show_imag": [2, 15, 19], "show_image_cv2": [2, 15, 19], "show_image_mpl": [2, 15, 19], "show_multipl": [2, 13], "snippet": 0, "some": 0, "sourc": 0, "space": 6, "specif": 0, "specifi": 0, "sphinx": 1, "ssh": 0, "stackoverflow": [6, 7, 21], "storag": 0, "str": [13, 16], "structur": [0, 2, 6, 15], "submodul": 1, "subpackag": 1, "sum_filt": 13, "t": 0, "table_arrai": 13, "table_image_infer": 0, "table_infer": [1, 2], "table_pars": [1, 2], "tablelin": 0, "targetfilepath": 0, "task": 0, "test_metr": [2, 15], "test_pdf": 0, "text": [2, 4], "thi": 0, "thing": 1, "threshold": 6, "through": 0, "titl": 13, "tol": 26, "toler": 26, "tool": 1, "transform_image_coordinates_to_pdf_coordin": [2, 15, 22], "transform_table_lines_by_page_info": [2, 15, 22], "transformation_matrix": [15, 22], "true": [13, 24, 26], "tupl": [13, 18, 22], "txt": 0, "type": [0, 13, 16, 22, 24], "unfil": 21, "upload": 0, "us": 0, "usag": 1, "util": [0, 1, 2], "uuid": 0, "venv": 0, "vision": 0, "visual": 1, "visual_log": [2, 15], "vlp_output": [0, 22], "w": 26, "where": 0, "whether": 21, "which": 0, "white": 7, "width": [0, 15, 22], "www": 1, "x1": [0, 26], "x2": [0, 26], "xywh": [15, 26], "xywh_dict": 26, "xywh_tupl": 26, "xyxi": [15, 26], "xyxy_tupl": 26, "y1": [0, 26], "y2": [0, 26], "you": 1, "zebra": 6}, "titles": ["cv-analysis - Visual (CV-Based) Document Parsing", "Welcome to CV Analysis Service documentation!", "cv_analysis package", "cv_analysis.config module", "cv_analysis.figure_detection package", "cv_analysis.figure_detection.figure_detection module", "cv_analysis.figure_detection.figures module", "cv_analysis.figure_detection.text module", "cv_analysis.layout_parsing module", "cv_analysis.locations module", "cv_analysis.redaction_detection module", "cv_analysis.server package", "cv_analysis.server.pipeline module", "cv_analysis.table_inference module", "cv_analysis.table_parsing module", "cv_analysis.utils package", "cv_analysis.utils.annotate module", "cv_analysis.utils.banner module", "cv_analysis.utils.connect_rects module", "cv_analysis.utils.display module", "cv_analysis.utils.draw module", "cv_analysis.utils.filters module", "cv_analysis.utils.image_extraction module", "cv_analysis.utils.open_pdf module", "cv_analysis.utils.postprocessing module", "cv_analysis.utils.preprocessing module", "cv_analysis.utils.structures module", "cv_analysis.utils.test_metrics module", "cv_analysis.utils.utils module", "cv_analysis.utils.visual_logging module", "serve module"], "titleterms": {"As": 0, "an": 0, "analysi": [0, 1], "annot": 16, "api": 0, "banner": 17, "base": 0, "build": 0, "cli": 0, "config": 3, "connect_rect": 18, "content": 1, "cv": [0, 1], "cv_analysi": [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29], "detect": 0, "displai": 19, "document": [0, 1], "draw": 20, "extract": [], "figur": [0, 6], "figure_detect": [4, 5, 6, 7], "filter": 21, "image_extract": 22, "indic": 1, "instal": 0, "keyword": [], "layout": 0, "layout_pars": 8, "locat": 9, "modul": [1, 3, 5, 6, 7, 8, 9, 10, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30], "open_pdf": 23, "packag": [2, 4, 11, 15], "pars": 0, "pipelin": 12, "postprocess": 24, "preprocess": 25, "readm": 1, "redact": 0, "redaction_detect": 10, "run": 0, "serv": 30, "server": [11, 12], "servic": [0, 1], "structur": 26, "submodul": [2, 4, 11, 15], "subpackag": 2, "tabl": [0, 1], "table_infer": 13, "table_pars": 14, "test_metr": 27, "text": 7, "tool": 0, "usag": 0, "util": [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29], "visual": 0, "visual_log": 29, "welcom": 1}}) \ No newline at end of file diff --git a/docs/build/json/README.fjson b/docs/build/json/README.fjson index b59fbf4..cf31906 100644 --- a/docs/build/json/README.fjson +++ b/docs/build/json/README.fjson @@ -1,8 +1,8 @@ { - "body": "
    \n

    Keyword-Service#

    \n

    Service to get keywords of a paragraph or whole document.

    \n\n
    \n

    API#

    \n
    \n

    REST#

    \n

    The service provides endpoints to extract keywords from a text and to embed a text. For details, download\nOpenAPI schema and view it in a browser.

    \n
    \n
    \n

    RabbitMQ#

    \n

    The service listens to a queue and processes the messages. This method is ment to be used for extracting keywords from\nwhole documents. All RabbitMQ parameters including the queue names are set in environment variables, refer to the\nservice respective HELM chart for more information.

    \n

    The input message should be a JSON object with the following structure:

    \n
    {\n  "targetFilePath": string,\n  "responseFilePath": string\n}\n
    \n
    \n

    The service downloads the file specified in targetFilePath. Supported data structures for the target file are:

    \n
      \n

    • simplified text data (signifier key: paragraphs)

      • \n

    • structure object data (signifier key: structureObjects)

      • \n
    \n

    As a response, the service uploads a JSON-structured file (as defined in responseFilePath) with the result under the\ndata key. The structure of the response file is as follows:

    \n
    {\n    "targetFilePath"\n:\n    string,\n        "responseFilePath"\n:\n    string,\n        // and eventually further fields if present in the input message      \n        "data"\n:\n    [\n        {\n            "keywords": Array[string],\n            "paragraphId": int,\n            "embedding": Array[float]  // 384 dimensions\n        }\n    ]\n}\n
    \n
    \n

    Note that

    \n
      \n

    • the embedding key is optional and can be omitted. The service will not calculate the embedding if the environment\nvariable MODEL__COMPUTE_EMBEDDINGS is set to false.

      • \n

    • the service also computes the keywords for the whole document. In this case, the paragraphId is set to -1.

      • \n
    \n
    \n
    \n
    \n

    Service Configuration#

    \n

    The service is configured via environment variables. The following variables are available:

    \n

    | Variable | Description | Default |\n| \u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014 | \u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2014\u2013 | \u2014\u2014- |\n| LOGGING__LEVEL | Logging level | INFO |\n| MODEL__MAX_KEYWORDS_PER_PARAGRAPH | Maximum number of keywords per paragraph | 5 |\n| MODLE__MAX_KEYWORDS_PER_DOCUMENT | Maximum number of keywords per document, when set to 0, no keywords are extracted | 0 |\n| MODEL__COMPUTE_EMBEDDINGS | Whether to compute keyword embeddings or not | true |\n| MODEL__PREPROCESSING__MIN_PARAGRAPH_LENGTH | Minimum number of characters in a paragraph to be considered for keyword extraction | 1 |\n| MODEL__POSTPROCESSING__FILTER_SUBWORDS | Whether to filter out subwords from the keywords or not | true |

    \n

    NOTE that these variables are subject to change. For the most recent configuration, refer to the service respective\nHELM chart.

    \n
    \n
    \n

    Language#

    \n

    Currently, there is an english, a german and a multi-language model for keyword extraction. The models are uploaded to\nmlflow and can\nbe set in the Dockerfile when building the container:

    \n

    example for german model:

    \n
    ENV AZURE_RESOURCE_GROUP="mriedl"\nENV AZURE_AML_WORKSPACE="azureml-ws"\nENV AZURE_AML_MODEL_NAME="keyword-extraction-de"\nENV AZURE_AML_MODEL_VERSION="1"\n
    \n
    \n

    and example for english model:

    \n
    ENV AZURE_RESOURCE_GROUP="mriedl"\nENV AZURE_AML_WORKSPACE="azureml-ws"\nENV AZURE_AML_MODEL_NAME="keyword-extraction-de"\nENV AZURE_AML_MODEL_VERSION="1"\n
    \n
    \n
    \n
    \n

    Usage#

    \n

    Two Options:

    \n
      \n

    1. REST: Send text per request to endpoint, endpoint returns keywords

      2. \n

    2. Queue: Service gets text from queue, model calculates keywords, save keywords in queue

      3. \n
    \n

    To test the REST endpoint you have to set up an environment and do poetry install (\nsee https://gitlab.knecon.com/knecon/research/template-python-project for details for setting up poetry)\nThen run

    \n
    python ./src/serve.py \n
    \n
    \n

    You don\u2019t need to start a queue for that, just ignore the AMQP Error.\nPort and host are set in settings.toml .\nYou can use the FastAPI under 127.0.0.1:8001/docs to send request to endpoint.

    \n

    You can also test the service with docker:

    \n
    \n

    Run Docker Commands#

    \n
    docker build -t ${IMAGE_NAME} -f Dockerfile --build-arg GITLAB_USER=${GITLAB_USER} \\\n                                            --build-arg GITLAB_ACCESS_TOKEN=${GITLAB_ACCESS_TOKEN} \\\n                                            --build-arg AZURE_TENANT_ID=${AZURE_TENANT_ID} \\\n                                            --build-arg AZURE_SUBSCRIPTION_ID=${AZURE_SUBSCRIPTION_ID} \\\n                                            --build-arg AZURE_CLIENT_ID=${AZURE_CLIENT_ID} \\\n                                            --build-arg AZURE_CLIENT_SECRET=${AZURE_CLIENT_SECRET} \\\n                                            --build-arg AZURE_AML_MODEL_VERSION=${AZURE_AML_MODEL_VERSION} \\\n                                            --build-arg AZURE_AML_MODEL_NAME=${AZURE_AML_MODEL_NAME} \\\n                                            --build-arg AZURE_RESOURCE_GROUP=${AZURE_RESOURCE_GROUP} \\\n                                            --build-arg AZURE_AML_WORKSPACE=${AZURE_AML_WORKSPACE}\n
    \n
    \n
    docker run --net=host -it --rm --name ${CONTAINER_NAME} ${IMAGE_NAME}\n
    \n
    \n
    \n
    \n

    Run locally#

    \n

    First you need to download the model from mlflow. This can be done with the \u201csrc/ml_flow/download_model.py\u201d script.\nThis scripts downloads a model and copies config and model data to the specific locations, such that the model can\nbe loaded.

    \n

    For running/testing the keyword extraction locally you can use the src/tests/test_process.py script.

    \n

    Model ist stored and loaded via DVC, you need the connection string under\nhttps://portal.azure.com/#@knecon.com/resource/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourceGroups/taas-rg/providers/Microsoft.Storage/storageAccounts/taassaracer/keys

    \n
    \n
    \n
    \n
    \n

    Upload models to ML Flow#

    \n

    To upload the models to mlflow, you can use following script: src/mlflow/upload_model.py\nFor authentication following environment variables need to be set:

    \n
    #AZURE_TENANT_ID=""\n#AZURE_SUBSCRIPTION_ID=""\n#AZURE_CLIENT_ID=""\n#AZURE_CLIENT_SECRET=""\n
    \n
    \n

    Additional settings (resource group, experiment name, etc.) can be specified in the config (\n./src/mlflow/config/azure_config.toml).\nThe upload_model.py has the following parameters:

    \n
    options:\n  -h, --help            show this help message and exit\n  -a AZURE_CONFIG, --azure_config AZURE_CONFIG\n                        Location of the configuration file for Azure (default: src/mlflow/config/azure_config.toml)\n  -b BASE_CONFIG, --base_config BASE_CONFIG\n                        Location of the basic training configuration (default: src/mlflow/config/settings_de.toml)\n  \n
    \n
    \n

    the base config contains all information for the models used. Examples for German and\nEnglish are placed in /src/mlflow/config/

    \n

    Note: Multi-language model tracking does not work for now. After the upload script reports an error, you have to\nmanually track the\nmodel here\nwhere you can find the run. Adhere to the naming conventions for the model name and versions,\nsee here

    \n
    \n", - "title": "Keyword-Service", + "body": "
    \n

    cv-analysis - Visual (CV-Based) Document Parsing#

    \n

    parse_pdf()\nThis repository implements computer vision based approaches for detecting and parsing visual features such as tables or\nprevious redactions in documents.

    \n
    \n

    API#

    \n

    Input message:

    \n
    {\n  "targetFilePath": {\n    "pdf": "absolute file path",\n    "vlp_output": "absolute file path"\n  },\n  "responseFilePath": "absolute file path",\n  "operation": "table_image_inference"\n}\n
    \n
    \n

    Response is uploaded to the storage as specified in the responseFilePath field. The structure is as follows:

    \n
    {\n  ...,\n  "data": [\n    {\n      'pageNum': 0,\n      'bbox': {\n        'x1': 55.3407,\n        'y1': 247.0246,\n        'x2': 558.5602,\n        'y2': 598.0585\n      },\n      'uuid': '2b10c1a2-393c-4fca-b9e3-0ad5b774ac84',\n      'label': 'table',\n      'tableLines': [\n        {\n          'x1': 0,\n          'y1': 16,\n          'x2': 1399,\n          'y2': 16\n        },\n        ...\n      ],\n      'imageInfo': {\n        'height': 693,\n        'width': 1414\n      }\n    },\n    ...\n  ]\n}\n
    \n
    \n
    \n
    \n

    Installation#

    \n
    git clone ssh://git@git.iqser.com:2222/rr/cv-analysis.git\ncd cv-analysis\n\npython -m venv env\nsource env/bin/activate\n\npip install -e .\npip install -r requirements.txt\n\ndvc pull\n
    \n
    \n
    \n
    \n

    Usage#

    \n
    \n

    As an API#

    \n

    The module provided functions for the individual tasks that all return some kind of collection of points, depending on\nthe specific task.

    \n
    \n

    Redaction Detection (API)#

    \n

    The below snippet shows hot to find the outlines of previous redactions.

    \n
    from cv_analysis.redaction_detection import find_redactions\nimport pdf2image\nimport numpy as np\n\npdf_path = ...\npage_index = ...\n\npage = pdf2image.convert_from_path(pdf_path, first_page=page_index, last_page=page_index)[0]\npage = np.array(page)\n\nredaction_contours = find_redactions(page)\n
    \n
    \n
    \n
    \n
    \n
    \n

    As a CLI Tool#

    \n

    Core API functionalities can be used through a CLI.

    \n
    \n

    Table Parsing#

    \n

    The tables parsing utility detects and segments tables into individual cells.

    \n
    python scripts/annotate.py data/test_pdf.pdf 7 --type table\n
    \n
    \n

    The below image shows a parsed table, where each table cell has been detected individually.

    \n

    \"Table

    \n
    \n
    \n

    Redaction Detection (CLI)#

    \n

    The redaction detection utility detects previous redactions in PDFs (filled black rectangles).

    \n
    python scripts/annotate.py data/test_pdf.pdf 2 --type redaction\n
    \n
    \n

    The below image shows the detected redactions with green outlines.

    \n

    \"Redaction

    \n
    \n
    \n

    Layout Parsing#

    \n

    The layout parsing utility detects elements such as paragraphs, tables and figures.

    \n
    python scripts/annotate.py data/test_pdf.pdf 7 --type layout\n
    \n
    \n

    The below image shows the detected layout elements on a page.

    \n

    \"Layout

    \n
    \n
    \n

    Figure Detection#

    \n

    The figure detection utility detects figures specifically, which can be missed by the generic layout parsing utility.

    \n
    python scripts/annotate.py data/test_pdf.pdf 3 --type figure\n
    \n
    \n

    The below image shows the detected figure on a page.

    \n

    \"Figure

    \n
    \n
    \n
    \n

    Running as a service#

    \n
    \n

    Building#

    \n

    Build base image

    \n
    bash setup/docker.sh\n
    \n
    \n

    Build head image

    \n
    docker build -f Dockerfile -t cv-analysis . --build-arg BASE_ROOT=""\n
    \n
    \n
    \n
    \n

    Usage (service)#

    \n

    Shell 1

    \n
    docker run --rm --net=host --rm cv-analysis\n
    \n
    \n

    Shell 2

    \n
    python scripts/client_mock.py --pdf_path /path/to/a/pdf\n
    \n
    \n
    \n
    \n
    \n", + "title": "cv-analysis - Visual (CV-Based) Document Parsing", "sourcename": "README.md.txt", "current_page_name": "README", - "toc": "\n", + "toc": "\n", "page_source_suffix": ".md" } \ No newline at end of file diff --git a/docs/build/json/index.fjson b/docs/build/json/index.fjson index 733be91..3935daa 100644 --- a/docs/build/json/index.fjson +++ b/docs/build/json/index.fjson @@ -1,8 +1,8 @@ { - "body": "
    \n

    Welcome to Keyword Extraction Service documentation!#

    \n
    \n

    Note

    \n

    If you\u2019d like to change the looks of things \ud83d\udc49 https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html

    \n
    \n
    \n

    Table of Contents#

    \n
    \n

    README

    \n\n
    \n
    \n

    Modules

    \n\n
    \n
    \n

    Indices and tables#

    \n\n
    \n
    \n
    \n", - "title": "Welcome to Keyword Extraction Service documentation!", + "body": "
    \n

    Welcome to CV Analysis Service documentation!#

    \n
    \n

    Note

    \n

    If you\u2019d like to change the looks of things \ud83d\udc49 https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html

    \n
    \n
    \n

    Table of Contents#

    \n
    \n

    README

    \n\n
    \n
    \n

    Modules

    \n\n
    \n
    \n

    Indices and tables#

    \n\n
    \n
    \n
    \n", + "title": "Welcome to CV Analysis Service documentation!", "sourcename": "index.rst.txt", "current_page_name": "index", - "toc": "\n", + "toc": "\n", "page_source_suffix": ".rst" } \ No newline at end of file diff --git a/docs/source/README.md b/docs/source/README.md index b1901d1..2c44979 100644 --- a/docs/source/README.md +++ b/docs/source/README.md @@ -1,203 +1,178 @@ -# Keyword-Service +# cv-analysis - Visual (CV-Based) Document Parsing -Service to get keywords of a paragraph or whole document. - - - -- [Keyword-Service](#keyword-service) - - [API](#api) - - [REST](#rest) - - [RabbitMQ](#rabbitmq) - - [Service Configuration](#service-configuration) - - [Language](#language) - - [Usage](#usage) - - [Run Docker Commands](#run-docker-commands) - - [Run locally](#run-locally) -- [Upload models to ML Flow](#upload-models-to-ml-flow) - - +parse_pdf() +This repository implements computer vision based approaches for detecting and parsing visual features such as tables or +previous redactions in documents. ## API -### REST - -The service provides endpoints to extract keywords from a text and to embed a text. For details, download -[OpenAPI schema](docs/openapi_redoc.html) and view it in a browser. - -### RabbitMQ - -The service listens to a queue and processes the messages. This method is ment to be used for extracting keywords from -whole documents. All RabbitMQ parameters including the queue names are set in environment variables, refer to the -service respective HELM chart for more information. - -The input message should be a JSON object with the following structure: +Input message: ```json { - "targetFilePath": string, - "responseFilePath": string + "targetFilePath": { + "pdf": "absolute file path", + "vlp_output": "absolute file path" + }, + "responseFilePath": "absolute file path", + "operation": "table_image_inference" } ``` -The service downloads the file specified in `targetFilePath`. Supported data structures for the target file are: +Response is uploaded to the storage as specified in the `responseFilePath` field. The structure is as follows: -- simplified text data (signifier key: `paragraphs`) -- structure object data (signifier key: `structureObjects`) - -As a response, the service uploads a JSON-structured file (as defined in `responseFilePath`) with the result under the -`data` key. The structure of the response file is as follows: - -```javascript +```json { - "targetFilePath" -: - string, - "responseFilePath" -: - string, - // and eventually further fields if present in the input message - "data" -: - [ + ..., + "data": [ + { + 'pageNum': 0, + 'bbox': { + 'x1': 55.3407, + 'y1': 247.0246, + 'x2': 558.5602, + 'y2': 598.0585 + }, + 'uuid': '2b10c1a2-393c-4fca-b9e3-0ad5b774ac84', + 'label': 'table', + 'tableLines': [ { - "keywords": Array[string], - "paragraphId": int, - "embedding": Array[float] // 384 dimensions - } - ] + 'x1': 0, + 'y1': 16, + 'x2': 1399, + 'y2': 16 + }, + ... + ], + 'imageInfo': { + 'height': 693, + 'width': 1414 + } + }, + ... + ] } -``` - -**Note** that - -- the `embedding` key is optional and can be omitted. The service will not calculate the embedding if the environment - variable `MODEL__COMPUTE_EMBEDDINGS` is set to `false`. -- the service also computes the keywords for the whole document. In this case, the `paragraphId` is set to `-1`. - -## Service Configuration - -The service is configured via environment variables. The following variables are available: - -| Variable | Description | Default | -| ------------------------------------------ | ----------------------------------------------------------------------------------- | ------- | -| LOGGING__LEVEL | Logging level | INFO | -| MODEL__MAX_KEYWORDS_PER_PARAGRAPH | Maximum number of keywords per paragraph | 5 | -| MODLE__MAX_KEYWORDS_PER_DOCUMENT | Maximum number of keywords per document, when set to 0, no keywords are extracted | 0 | -| MODEL__COMPUTE_EMBEDDINGS | Whether to compute keyword embeddings or not | true | -| MODEL__PREPROCESSING__MIN_PARAGRAPH_LENGTH | Minimum number of characters in a paragraph to be considered for keyword extraction | 1 | -| MODEL__POSTPROCESSING__FILTER_SUBWORDS | Whether to filter out subwords from the keywords or not | true | - -**NOTE** that these variables are subject to change. For the most recent configuration, refer to the service respective -HELM chart. - -## Language - -Currently, there is an english, a german and a multi-language model for keyword extraction. The models are uploaded to -mlflow and can -be set in the Dockerfile when building the container: - -example for german model: -``` -ENV AZURE_RESOURCE_GROUP="mriedl" -ENV AZURE_AML_WORKSPACE="azureml-ws" -ENV AZURE_AML_MODEL_NAME="keyword-extraction-de" -ENV AZURE_AML_MODEL_VERSION="1" ``` -and example for english model: +## Installation -``` -ENV AZURE_RESOURCE_GROUP="mriedl" -ENV AZURE_AML_WORKSPACE="azureml-ws" -ENV AZURE_AML_MODEL_NAME="keyword-extraction-de" -ENV AZURE_AML_MODEL_VERSION="1" +```bash +git clone ssh://git@git.iqser.com:2222/rr/cv-analysis.git +cd cv-analysis + +python -m venv env +source env/bin/activate + +pip install -e . +pip install -r requirements.txt + +dvc pull ``` ## Usage -**Two Options:** +### As an API -1. REST: Send text per request to endpoint, endpoint returns keywords -2. Queue: Service gets text from queue, model calculates keywords, save keywords in queue +The module provided functions for the individual tasks that all return some kind of collection of points, depending on +the specific task. -To test the REST endpoint you have to set up an environment and do poetry install ( -see https://gitlab.knecon.com/knecon/research/template-python-project for details for setting up poetry) -Then run +#### Redaction Detection (API) -``` -python ./src/serve.py +The below snippet shows hot to find the outlines of previous redactions. + +```python +from cv_analysis.redaction_detection import find_redactions +import pdf2image +import numpy as np + +pdf_path = ... +page_index = ... + +page = pdf2image.convert_from_path(pdf_path, first_page=page_index, last_page=page_index)[0] +page = np.array(page) + +redaction_contours = find_redactions(page) ``` -You don't need to start a queue for that, just ignore the AMQP Error. -Port and host are set in settings.toml . -You can use the FastAPI under 127.0.0.1:8001/docs to send request to endpoint. +## As a CLI Tool -You can also test the service with docker: +Core API functionalities can be used through a CLI. -#### Run Docker Commands +### Table Parsing + +The tables parsing utility detects and segments tables into individual cells. ```bash -docker build -t ${IMAGE_NAME} -f Dockerfile --build-arg GITLAB_USER=${GITLAB_USER} \ - --build-arg GITLAB_ACCESS_TOKEN=${GITLAB_ACCESS_TOKEN} \ - --build-arg AZURE_TENANT_ID=${AZURE_TENANT_ID} \ - --build-arg AZURE_SUBSCRIPTION_ID=${AZURE_SUBSCRIPTION_ID} \ - --build-arg AZURE_CLIENT_ID=${AZURE_CLIENT_ID} \ - --build-arg AZURE_CLIENT_SECRET=${AZURE_CLIENT_SECRET} \ - --build-arg AZURE_AML_MODEL_VERSION=${AZURE_AML_MODEL_VERSION} \ - --build-arg AZURE_AML_MODEL_NAME=${AZURE_AML_MODEL_NAME} \ - --build-arg AZURE_RESOURCE_GROUP=${AZURE_RESOURCE_GROUP} \ - --build-arg AZURE_AML_WORKSPACE=${AZURE_AML_WORKSPACE} +python scripts/annotate.py data/test_pdf.pdf 7 --type table ``` +The below image shows a parsed table, where each table cell has been detected individually. + +![Table Parsing Demonstration](data/table_parsing.png) + +### Redaction Detection (CLI) + +The redaction detection utility detects previous redactions in PDFs (filled black rectangles). + ```bash -docker run --net=host -it --rm --name ${CONTAINER_NAME} ${IMAGE_NAME} +python scripts/annotate.py data/test_pdf.pdf 2 --type redaction ``` -#### Run locally +The below image shows the detected redactions with green outlines. -First you need to download the model from mlflow. This can be done with the *"src/ml_flow/download_model.py"* script. -This scripts downloads a model and copies config and model data to the specific locations, such that the model can -be loaded. +![Redaction Detection Demonstration](data/redaction_detection.png) -For running/testing the keyword extraction locally you can use the *src/tests/test_process.py* script. +### Layout Parsing -Model ist stored and loaded via DVC, you need the connection string under -https://portal.azure.com/#@knecon.com/resource/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourceGroups/taas-rg/providers/Microsoft.Storage/storageAccounts/taassaracer/keys +The layout parsing utility detects elements such as paragraphs, tables and figures. -# Upload models to ML Flow - -To upload the models to mlflow, you can use following script: src/mlflow/upload_model.py -For authentication following environment variables need to be set: - -``` -#AZURE_TENANT_ID="" -#AZURE_SUBSCRIPTION_ID="" -#AZURE_CLIENT_ID="" -#AZURE_CLIENT_SECRET="" +```bash +python scripts/annotate.py data/test_pdf.pdf 7 --type layout ``` -Additional settings (resource group, experiment name, etc.) can be specified in the config ( -*./src/mlflow/config/azure_config.toml*). -The *upload_model.py* has the following parameters: +The below image shows the detected layout elements on a page. -``` -options: - -h, --help show this help message and exit - -a AZURE_CONFIG, --azure_config AZURE_CONFIG - Location of the configuration file for Azure (default: src/mlflow/config/azure_config.toml) - -b BASE_CONFIG, --base_config BASE_CONFIG - Location of the basic training configuration (default: src/mlflow/config/settings_de.toml) - +![Layout Parsing Demonstration](data/layout_parsing.png) +### Figure Detection + +The figure detection utility detects figures specifically, which can be missed by the generic layout parsing utility. + +```bash +python scripts/annotate.py data/test_pdf.pdf 3 --type figure ``` -the base config contains all information for the models used. Examples for German and -English are placed in */src/mlflow/config/* +The below image shows the detected figure on a page. -Note: Multi-language model tracking does not work for now. After the upload script reports an error, you have to -manually track the -model [here](https://ml.azure.com/experiments?wsid=/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourcegroups/fforesight-rg/providers/Microsoft.MachineLearningServices/workspaces/ff-aml-main&tid=b44be368-e4f2-4ade-a089-cd2825458048) -where you can find the run. Adhere to the naming conventions for the model name and versions, -see [here](https://ml.azure.com/model/list?wsid=/subscriptions/4b9531fc-c5e4-4b11-8492-0cc173c1f97d/resourcegroups/fforesight-rg/providers/Microsoft.MachineLearningServices/workspaces/ff-aml-main&tid=b44be368-e4f2-4ade-a089-cd2825458048) +![Figure Detection Demonstration](data/figure_detection.png) +## Running as a service + +### Building + +Build base image + +```bash +bash setup/docker.sh +``` + +Build head image + +```bash +docker build -f Dockerfile -t cv-analysis . --build-arg BASE_ROOT="" +``` + +### Usage (service) + +Shell 1 + +```bash +docker run --rm --net=host --rm cv-analysis +``` + +Shell 2 + +```bash +python scripts/client_mock.py --pdf_path /path/to/a/pdf +``` diff --git a/docs/source/data/.gitignore b/docs/source/data/.gitignore new file mode 100644 index 0000000..8f6ae38 --- /dev/null +++ b/docs/source/data/.gitignore @@ -0,0 +1,5 @@ +/test_pdf.pdf +/figure_detection.png +/layout_parsing.png +/redaction_detection.png +/table_parsing.png diff --git a/docs/source/data/data/.gitignore b/docs/source/data/data/.gitignore new file mode 100644 index 0000000..8f6ae38 --- /dev/null +++ b/docs/source/data/data/.gitignore @@ -0,0 +1,5 @@ +/test_pdf.pdf +/figure_detection.png +/layout_parsing.png +/redaction_detection.png +/table_parsing.png diff --git a/docs/source/index.rst b/docs/source/index.rst index e356eff..05aa9b4 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -3,9 +3,9 @@ You can adapt this file completely to your liking, but it should at least contain the root `toctree` directive. -==================================================== -Welcome to Keyword Extraction Service documentation! -==================================================== +============================================= +Welcome to CV Analysis Service documentation! +============================================= .. note::