Adding api-inference-community to huggingface_hub.

.github/workflows/python-api-quality.yaml

@@ -0,0 +1,20 @@
+name: Check code quality
+
+on: push
+
+jobs:
+  run_tests:
+    runs-on: ubuntu-18.04
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up Python
+        uses: actions/setup-python@v1
+        with:
+          python-version: 3.8
+      - name: Install dependencies
+        run: |
+          pip install black isort flake8 mypy
+      - name: Make quality
+        workdir: api-inference-community
+        run: |
+          make quality

.github/workflows/python-api-tests.yaml

@@ -0,0 +1,33 @@
+name: Python-tests
+
+on:
+  push:
+    branches:
+      - "*"
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.8"]
+
+    steps:
+      - run: |
+          sudo apt-get install ffmpeg
+
+      - uses: actions/checkout@v2
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install dependencies
+        workdir: api-inference-community
+        run: |
+          pip install --upgrade pip
+          pip install pytest pillow httpx
+          pip install -e .
+      - run: make test
+        workdir: api-inference-community

README.md

@@ -140,6 +140,10 @@ We add a custom "Use in Asteroid" button.
 
 When clicked you get a library-specific code sample that you'll be able to specify. 🔥
 
+## API integration into the huggingface.co hub
+
+In order to get functional widgets on the hub for your models check out this [doc](https://github.com/huggingface/huggingface_hub/tree/master/api-inference-community)
+
 <br>
 
 ## Feedback (feature requests, bugs, etc.) is super welcome 💙💚💛💜♥️🧡

api-inference-community/.gitignore

@@ -0,0 +1,9 @@
+__pycache__/
+*.py[cod]
+
+# Speechbrain artefact.
+pretrained_checkpoints
+
+*.egg-info/
+build/
+dist/

api-inference-community/.pre-commit-config.yaml

@@ -0,0 +1,64 @@
+repos:
+  - repo: https://github.com/ambv/black
+    rev: 20.8b1
+    hooks:
+      - id: black
+  - repo: https://gitlab.com/pycqa/flake8
+    rev: 3.8.3
+    hooks:
+      - id: flake8
+  - repo: https://github.com/pre-commit/mirrors-isort
+    rev: v5.7.0 # Use the revision sha / tag you want to point at
+    hooks:
+      - id: isort
+  - repo: https://github.com/pre-commit/mirrors-mypy
+    rev: "b84ce099a2fd3c5216b6ccf3fd176c3828b075fb" # Use the sha / tag you want to point at
+    hooks:
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/common/
+        entry: mypy docker_images/common/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/speechbrain/
+        entry: mypy docker_images/speechbrain/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/kasteroid/
+        entry: mypy docker_images/asteroid/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/allennlp/
+        entry: mypy docker_images/allennlp/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/espnet/
+        entry: mypy docker_images/espnet/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/timm/
+        entry: mypy docker_images/timm/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/flair/
+        entry: mypy docker_images/flair/
+        pass_filenames: false
+      - id: mypy
+        args: [--ignore-missing-imports]
+        additional_dependencies: [tokenize-rt==3.2.0]
+        files: ^docker_images/sentence_transformers/
+        entry: mypy docker_images/sentence_transformers/
+        pass_filenames: false

api-inference-community/MANIFEST.in

@@ -0,0 +1 @@
+include README.md requirements.txt

api-inference-community/Makefile

@@ -0,0 +1,20 @@
+.PHONY: quality style
+
+
+check_dirs := .
+
+
+
+quality:
+	black --check $(check_dirs)
+	isort --check-only $(check_dirs)
+	flake8 $(check_dirs)
+
+style:
+	black $(check_dirs)
+	isort $(check_dirs)
+
+
+test:
+	pytest -sv tests/
+

api-inference-community/README.md

@@ -0,0 +1,59 @@
+
+This repositories enable third-party libraries integrated with [huggingface_hub](https://github.com/huggingface/huggingface_hub/) to create
+their own docker so that the widgets on the hub can work as the `transformers` one do.
+
+The hardware to run the API will be provided by Hugging Face for now.
+
+The `docker_images/common` folder is intended to be a starter point for all new libs that 
+want to be integrated.
+
+### Adding a new container from a new lib.
+
+
+1. Copy the `docker_images/common` folder into your library's name `docker_images/example`.
+2. Edit:
+    - `docker_images/example/requirements.txt`
+    - `docker_images/example/app/main.py`
+    - `docker_images/example/app/pipelines/{task_name}.py` 
+    to implement the desired functionnality. All required code is marked with `IMPLEMENT_THIS` markup.
+3. Feel free to customize anything required by your lib everywhere you want. The only real requirements, are to honor the HTTP endpoints, in the same fashion as the `common` folder for all your supported tasks.
+4. Edit `example/tests/test_api.py` to add TESTABLE_MODELS.
+5. Pass the test suite `pytest -sv --rootdir docker_images/example/ docker_images/example/`
+6. Enjoy !
+
+### Developping while updating `api-inference-community`.
+
+If you ever come across a bug within `api-inference-community` or want to update it
+the developpement process is slightly more involved.
+
+- First, make sure you need to change this package, each framework is very autonomous
+ so if your code can get away by being standalone go that way first as it's much simpler.
+- If you can make the change only in `api-inference-community` without depending on it
+that's also a great option. Make sure to add the proper tests to your PR.
+- Finally, the best way to go is to develop locally using `manage.py` command:
+- Do the necessary modifications within `api-inference-community` first.
+- Install it locally in your environment with `pip install -e .`
+- Install your package dependencies locally.
+- Run your webserver locally: `./manage.py start --framework example --task audio-source-separation --model-id MY_MODEL`
+- When everything is working, you will need to split your PR in two, 1 for the `api-inference-community` part.
+  The second one will be for your package specific modifications and will only land once the `api-inference-community`
+  tag has landed.
+- This workflow is still work in progress, don't hesitate to ask questions to maintainers.
+
+Another similar command `./manage.py docker --framework example --task audio-source-separation --model-id MY_MODEL`
+Will launch the server, but this time in a protected, controlled docker environment making sure the behavior
+will be exactly the one in the API.
+
+
+
+### Available tasks
+
+- **Automatic speech recognition**: Input is a file, output is a dict of understood words being said within the file
+- **Text generation**: Input is a text, output is a dict of generated text
+- **Image recognition**: Input is an image, output is a dict of generated text
+- **Question answering**: Input is a question + some context, output is a dict containing necessary information to locate the answer to the `question` within the `context`.
+- **Audio source separation**: Input is some audio, and the output is n audio files that sum up to the original audio but contain individual soures of sound (either speakers or instruments for instant).
+- **Token classification**: Input is some text, and the output is a list of entities mentionned in the text. Entities can be anything remarquable like locations, organisations, persons, times etc...
+- **Text to speech**: Input is some text, and the output is an audio file saying the text...
+- **Sentence Similarity**: Input is some sentence and a list of reference sentences, and the list of similarity scores.
+

api-inference-community/api_inference_community/routes.py

@@ -0,0 +1,113 @@
+import logging
+import os
+import time
+from typing import Any, Dict
+
+from api_inference_community.validation import ffmpeg_convert, normalize_payload
+from pydantic import ValidationError
+from starlette.requests import Request
+from starlette.responses import JSONResponse, Response
+
+
+HF_HEADER_COMPUTE_TIME = "x-compute-time"
+HF_HEADER_COMPUTE_TYPE = "x-compute-type"
+HF_HEADER_COMPUTE_CHARACTERS = "x-compute-characters"
+COMPUTE_TYPE = os.getenv("COMPUTE_TYPE", "cpu")
+
+logger = logging.getLogger(__name__)
+
+
+async def pipeline_route(request: Request) -> Response:
+    start = time.time()
+    payload = await request.body()
+    task = os.environ["TASK"]
+    pipe = request.app.pipeline
+    try:
+        sampling_rate = pipe.sampling_rate
+    except Exception:
+        sampling_rate = None
+    try:
+        inputs, params = normalize_payload(payload, task, sampling_rate=sampling_rate)
+    except ValidationError as e:
+        errors = []
+        for error in e.errors():
+            errors.append(f'{error["msg"]}: `{error["loc"][0]}` in `parameters`')
+        return JSONResponse({"error": errors}, status_code=400)
+    except (EnvironmentError, ValueError) as e:
+        return JSONResponse({"error": str(e)}, status_code=400)
+    except Exception as e:
+        return JSONResponse({"error": str(e)}, status_code=500)
+
+    return call_pipe(pipe, inputs, params, start)
+
+
+def call_pipe(pipe: Any, inputs, params: Dict, start: float) -> Response:
+    root_logger = logging.getLogger()
+    warnings = set()
+
+    class RequestsHandler(logging.Handler):
+        def emit(self, record):
+            """Send the log records (created by loggers) to
+            the appropriate destination.
+            """
+            warnings.add(record.getMessage())
+
+    handler = RequestsHandler()
+    handler.setLevel(logging.WARNING)
+    root_logger.addHandler(handler)
+    for _logger in logging.root.manager.loggerDict.values():  # type: ignore
+        try:
+            _logger.addHandler(handler)
+        except Exception:
+            pass
+
+    status_code = 200
+    n_characters = 0
+    try:
+        outputs = pipe(inputs)
+        n_characters = get_input_characters(inputs)
+    except (AssertionError, ValueError) as e:
+        outputs = {"error": str(e)}
+        status_code = 400
+    except Exception as e:
+        outputs = {"error": "unknown error"}
+        status_code = 500
+        logger.error(f"There was an inference error: {e}")
+
+    if warnings and isinstance(outputs, dict):
+        outputs["warnings"] = list(sorted(warnings))
+
+    compute_type = COMPUTE_TYPE
+    headers = {
+        HF_HEADER_COMPUTE_TIME: "{:.3f}".format(time.time() - start),
+        HF_HEADER_COMPUTE_TYPE: compute_type,
+        # https://stackoverflow.com/questions/43344819/reading-response-headers-with-fetch-api/44816592#44816592
+        "access-control-expose-headers": f"{HF_HEADER_COMPUTE_TYPE}, {HF_HEADER_COMPUTE_TIME}",
+    }
+    if status_code == 200:
+        headers[HF_HEADER_COMPUTE_CHARACTERS] = f"{n_characters}"
+        if os.getenv("TASK") in {"text-to-speech", "audio-source-separation"}:
+            # Special case, right now everything is flac audio we can output
+            waveform, sampling_rate = outputs
+            data = ffmpeg_convert(waveform, sampling_rate)
+            headers["content-type"] = "audio/flac"
+            return Response(data, headers=headers, status_code=status_code)
+    return JSONResponse(
+        outputs,
+        headers=headers,
+        status_code=status_code,
+    )
+
+
+def get_input_characters(inputs) -> int:
+    if isinstance(inputs, str):
+        return len(inputs)
+    elif isinstance(inputs, (tuple, list)):
+        return sum(get_input_characters(input_) for input_ in inputs)
+    elif isinstance(inputs, dict):
+        return sum(get_input_characters(input_) for input_ in inputs.values())
+    return -1
+
+
+async def status_ok(request):
+    return JSONResponse({"ok": "ok"})