Soporte cuda

.cog/openapi_schema.json

@@ -0,0 +1 @@
+{"components":{"schemas":{"HTTPValidationError":{"properties":{"detail":{"items":{"$ref":"#/components/schemas/ValidationError"},"title":"Detail","type":"array"}},"title":"HTTPValidationError","type":"object"},"Input":{"properties":{"audio":{"description":"Audio file (mp3 or wav).","format":"uri","title":"Audio","type":"string","x-order":0},"bars":{"default":50,"description":"Number of bars on the video at once.","title":"Bars","type":"integer","x-order":6},"color":{"default":"0.03,0.6,0.3","description":"Color of the bars as 'r,g,b' in [0,1].","title":"Color","type":"string","x-order":3},"color2":{"default":"0.5,0.3,0.6","description":"Color of the second waveform for stereo files.","title":"Color2","type":"string","x-order":4},"duration":{"description":"Duration in seconds from seek time.","title":"Duration","type":"number","x-order":13},"height":{"default":300,"description":"Height of the animation in pixels.","title":"Height","type":"integer","x-order":11},"oversample":{"default":4,"description":"Lower values will feel less reactive.","title":"Oversample","type":"number","x-order":7},"rate":{"default":60,"description":"Video framerate.","title":"Rate","type":"integer","x-order":1},"seek":{"description":"Seek to time in seconds in video.","title":"Seek","type":"number","x-order":12},"speed":{"default":4,"description":"Faster transitions between frames.","title":"Speed","type":"number","x-order":9},"stereo":{"default":false,"description":"Create 2 waveforms for stereo files.","title":"Stereo","type":"boolean","x-order":2},"time_param":{"default":0.4,"description":"Amount of audio shown at once on a frame.","title":"Time Param","type":"number","x-order":8},"white":{"default":false,"description":"Use white background.","title":"White","type":"boolean","x-order":5},"width":{"default":480,"description":"Width of the animation in pixels.","title":"Width","type":"integer","x-order":10}},"required":["audio"],"title":"Input","type":"object"},"Output":{"format":"uri","title":"Output","type":"string"},"PredictionRequest":{"properties":{"created_at":{"format":"date-time","title":"Created At","type":"string"},"id":{"title":"Id","type":"string"},"input":{"$ref":"#/components/schemas/Input"},"output_file_prefix":{"title":"Output File Prefix","type":"string"},"webhook":{"format":"uri","maxLength":65536,"minLength":1,"title":"Webhook","type":"string"},"webhook_events_filter":{"default":["start","output","logs","completed"],"items":{"$ref":"#/components/schemas/WebhookEvent"},"type":"array"}},"title":"PredictionRequest","type":"object"},"PredictionResponse":{"properties":{"completed_at":{"format":"date-time","title":"Completed At","type":"string"},"created_at":{"format":"date-time","title":"Created At","type":"string"},"error":{"title":"Error","type":"string"},"id":{"title":"Id","type":"string"},"input":{"$ref":"#/components/schemas/Input"},"logs":{"default":"","title":"Logs","type":"string"},"metrics":{"title":"Metrics","type":"object"},"output":{"$ref":"#/components/schemas/Output"},"started_at":{"format":"date-time","title":"Started At","type":"string"},"status":{"$ref":"#/components/schemas/Status"},"version":{"title":"Version","type":"string"}},"title":"PredictionResponse","type":"object"},"Status":{"description":"An enumeration.","enum":["starting","processing","succeeded","canceled","failed"],"title":"Status","type":"string"},"ValidationError":{"properties":{"loc":{"items":{"anyOf":[{"type":"string"},{"type":"integer"}]},"title":"Location","type":"array"},"msg":{"title":"Message","type":"string"},"type":{"title":"Error Type","type":"string"}},"required":["loc","msg","type"],"title":"ValidationError","type":"object"},"WebhookEvent":{"description":"An enumeration.","enum":["start","output","logs","completed"],"title":"WebhookEvent","type":"string"}}},"info":{"title":"Cog","version":"0.1.0"},"openapi":"3.0.2","paths":{"/":{"get":{"operationId":"root__get","responses":{"200":{"content":{"application/json":{"schema":{"title":"Response Root  Get"}}},"description":"Successful Response"}},"summary":"Root"}},"/health-check":{"get":{"operationId":"healthcheck_health_check_get","responses":{"200":{"content":{"application/json":{"schema":{"title":"Response Healthcheck Health Check Get"}}},"description":"Successful Response"}},"summary":"Healthcheck"}},"/predictions":{"post":{"description":"Run a single prediction on the model","operationId":"predict_predictions_post","parameters":[{"in":"header","name":"prefer","required":false,"schema":{"title":"Prefer","type":"string"}}],"requestBody":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/PredictionRequest"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/PredictionResponse"}}},"description":"Successful Response"},"422":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}},"description":"Validation Error"}},"summary":"Predict"}},"/predictions/{prediction_id}":{"put":{"description":"Run a single prediction on the model (idempotent creation).","operationId":"predict_idempotent_predictions__prediction_id__put","parameters":[{"in":"path","name":"prediction_id","required":true,"schema":{"title":"Prediction ID","type":"string"}},{"in":"header","name":"prefer","required":false,"schema":{"title":"Prefer","type":"string"}}],"requestBody":{"content":{"application/json":{"schema":{"allOf":[{"$ref":"#/components/schemas/PredictionRequest"}],"title":"Prediction Request"}}},"required":true},"responses":{"200":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/PredictionResponse"}}},"description":"Successful Response"},"422":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}},"description":"Validation Error"}},"summary":"Predict Idempotent"}},"/predictions/{prediction_id}/cancel":{"post":{"description":"Cancel a running prediction","operationId":"cancel_predictions__prediction_id__cancel_post","parameters":[{"in":"path","name":"prediction_id","required":true,"schema":{"title":"Prediction ID","type":"string"}}],"responses":{"200":{"content":{"application/json":{"schema":{"title":"Response Cancel Predictions  Prediction Id  Cancel Post"}}},"description":"Successful Response"},"422":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}},"description":"Validation Error"}},"summary":"Cancel"}},"/shutdown":{"post":{"operationId":"start_shutdown_shutdown_post","responses":{"200":{"content":{"application/json":{"schema":{"title":"Response Start Shutdown Shutdown Post"}}},"description":"Successful Response"}},"summary":"Start Shutdown"}}}}

.dockerignore

@@ -0,0 +1,17 @@
+# The .dockerignore file excludes files from the container build process.
+#
+# https://docs.docker.com/engine/reference/builder/#dockerignore-file
+
+# Exclude Git files
+**/.git
+**/.github
+**/.gitignore
+
+# Exclude Python cache files
+__pycache__
+.mypy_cache
+.pytest_cache
+.ruff_cache
+
+# Exclude Python virtual environment
+/venv

.github/workflows/push.yaml

@@ -0,0 +1,54 @@
+name: Push to Replicate
+
+on:
+  # Workflow dispatch allows you to manually trigger the workflow from GitHub.com
+  # Go to your repo, click "Actions", click "Push to Replicate", click "Run workflow"
+  workflow_dispatch:
+    inputs:
+      model_name:
+        description: 'Enter the model name, like "alice/bunny-detector". If unset, this will default to the value of `image` in cog.yaml.'
+  # # Uncomment these lines to trigger the workflow on every push to the main branch
+  # push:
+  #   branches:
+  #     - main
+
+jobs:
+  push_to_replicate:
+    name: Push to Replicate
+
+    # If your model is large, the default GitHub Actions runner may not 
+    # have enough disk space. If you need more space you can set up a 
+    # bigger runner on GitHub.
+    runs-on: ubuntu-latest
+
+    steps:
+      # This action cleans up disk space to make more room for your
+      # model code, weights, etc.
+      - name: Free disk space
+        uses: jlumbroso/[email protected]
+        with:
+          tool-cache: false
+          docker-images: false
+
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      # This action installs Docker buildx and Cog (and optionally CUDA)
+      - name: Setup Cog
+        uses: replicate/setup-cog@v2
+        with:
+          # If you set REPLICATE_API_TOKEN in your GitHub repository secrets,
+          # the action will authenticate with Replicate automatically so you
+          # can push your model
+          token: ${{ secrets.REPLICATE_API_TOKEN }}
+
+      # If you trigger the workflow manually, you can specify the model name.
+      # If you leave it blank (or if the workflow is triggered by a push), the 
+      # model name will be derived from the `image` value in cog.yaml.
+      - name: Push to Replicate
+        run: |
+          if [ -n "${{ inputs.model_name }}" ]; then
+            cog push r8.im/${{ inputs.model_name }}
+          else
+            cog push
+          fi

Dockerfile

@@ -0,0 +1,37 @@
+FROM nvidia/cuda:12.1.1-runtime-ubuntu20.04
+
+# Configurar entorno no interactivo
+ENV DEBIAN_FRONTEND=noninteractive
+
+# Instalar dependencias del sistema
+RUN apt-get update && apt-get install -y \
+    python3.9 \
+    python3.9-dev \
+    libcairo2-dev \
+    libjpeg-dev \
+    libgif-dev \
+    libpng-dev \
+    meson \
+    ninja-build \
+    build-essential \
+    ffmpeg \
+    git \
+    && rm -rf /var/lib/apt/lists/*
+
+# Crear y establecer el directorio de trabajo
+WORKDIR /app
+
+# Copiar archivos del proyecto
+COPY . /app
+
+# Actualizar pip e instalar dependencias
+RUN python3.9 -m pip install --upgrade pip && \
+    python3.9 -m pip install -r requirements.txt && \
+    python3.9 -m pip uninstall -y pycairo && \
+    python3.9 -m pip install pycairo \
+    python3.9 -m pip uninstall cog -y \
+    python3.9 -m pip pip install cog
+
+# Comando de entrada
+ENTRYPOINT ["python3"]
+

cog.yaml

@@ -0,0 +1,31 @@
+build:
+  python_version: "3.9"  # Define la versión de Python
+  gpu: true  # Habilitar soporte para GPU
+  cuda: "12.1"  # Especificar la versión de CUDA compatible con cupy-cuda12x
+  python_packages:
+    - cupy-cuda12x==11.6.0
+    - fastrlock==0.8.2
+    - imageio==2.31.2
+    - imageio-ffmpeg==0.5.1
+    - numpy==1.21.6
+    - Pillow==9.5.0
+    - psutil==6.1.0
+    - tqdm==4.67.0
+  system_packages:
+    - libcairo2-dev  # Necesario para pycairo
+    - libjpeg-dev
+    - libgif-dev
+    - libpng-dev
+    - meson
+    - ninja-build
+    - build-essential
+    - ffmpeg
+    - git
+  run:
+    - pip install --upgrade pip
+    - pip uninstall -y pycairo  # Reinstalar pycairo para asegurar compatibilidad
+    - pip install pycairo
+    - pip uninstall cog -y
+    - pip install cog
+
+predict: "predict.py:Predictor"  # Punto de entrada para predicciones

predict.py

@@ -0,0 +1,70 @@
+import os
+import subprocess
+from cog import BasePredictor, Input, Path
+
+class Predictor(BasePredictor):
+    def predict(
+        self,
+        audio: Path = Input(description="Audio file (mp3 or wav)."),
+        rate: int = Input(description="Video framerate.", default=60),
+        stereo: bool = Input(description="Create 2 waveforms for stereo files.", default=False),
+        color: str = Input(description="Color of the bars as 'r,g,b' in [0,1].", default="0.03,0.6,0.3"),
+        color2: str = Input(description="Color of the second waveform for stereo files.", default="0.5,0.3,0.6"),
+        white: bool = Input(description="Use white background.", default=False),
+        bars: int = Input(description="Number of bars on the video at once.", default=50),
+        oversample: float = Input(description="Lower values will feel less reactive.", default=4.0),
+        time_param: float = Input(description="Amount of audio shown at once on a frame.", default=0.4),
+        speed: float = Input(description="Faster transitions between frames.", default=4.0),
+        width: int = Input(description="Width of the animation in pixels.", default=480),
+        height: int = Input(description="Height of the animation in pixels.", default=300),
+        seek: float = Input(description="Seek to time in seconds in video.", default=None),
+        duration: float = Input(description="Duration in seconds from seek time.", default=None),
+    ) -> Path:
+        """
+        Procesa el archivo de audio subido y devuelve un archivo MP4.
+        """
+        # Ruta de salida para el archivo generado
+        output_path = Path("/src/out.mp4")
+        os.makedirs(os.path.dirname(output_path), exist_ok=True)
+
+        # Construir el comando como una lista
+        command = [
+            "/root/.pyenv/shims/python3",
+            os.path.join('/src/seewav.py'),
+            "-r", str(rate),
+            "-c", color,
+            "-c2", color2,
+            "-B", str(bars),
+            "-O", str(oversample),
+            "-T", str(time_param),
+            "-S", str(speed),
+            "-W", str(width),
+            "-H", str(height),
+            str(audio),
+            str(output_path)
+        ]
+
+        if stereo:
+            command.append("--stereo")
+        if white:
+            command.append("--white")
+        if seek is not None:
+            command.extend(["-s", str(seek)])
+        if duration is not None:
+            command.extend(["-d", str(duration)])
+
+        # Imprimir el comando para depuración
+        print("Comando a ejecutar:", " ".join(command))
+
+        # Ejecutar el comando y capturar la salida
+        result = subprocess.run(command, check=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+
+        # Imprimir la salida y errores para depuración
+        print("STDOUT:", result.stdout)
+        print("STDERR:", result.stderr)
+
+        # Verificar si el archivo de salida se ha generado correctamente
+        if not os.path.exists(output_path) or os.path.getsize(output_path) == 0:
+            raise RuntimeError("El archivo de salida no se ha generado correctamente.")
+
+        return output_path  # Devolver la ruta del archivo de salida

requirements.txt

@@ -1,3 +1,9 @@
-numpy
-pycairo
-tqdm
+cupy-cuda12x==11.6.0
+fastrlock==0.8.2
+imageio==2.31.2
+imageio-ffmpeg==0.5.1
+numpy==1.21.6
+Pillow==9.5.0
+psutil==6.1.0
+pycairo==1.23.0
+tqdm==4.67.0