fixed dotend req

Dockerfile

@@ -5,6 +5,10 @@ ARG DEBIAN_FRONTEND=noninteractive
 ENV PYTHONUNBUFFERED=1
 ENV NODE_MAJOR=20
 
+# Expose Flask port
+ENV FAST_API_PORT=7860
+EXPOSE 7860
+
 RUN apt-get update && apt-get install --no-install-recommends -y \
     build-essential \
     python3.9 \
@@ -23,15 +27,11 @@ RUN curl -fsSL https://deb.nodesource.com/gpgkey/nodesource-repo.gpg.key | gpg -
 RUN echo "deb [signed-by=/etc/apt/keyrings/nodesource.gpg] https://deb.nodesource.com/node_${NODE_MAJOR}.x nodistro main" | tee /etc/apt/sources.list.d/nodesource.list > /dev/null
 RUN apt-get update && apt-get install nodejs -y
 
-# Expose Flask port
-ENV FAST_API_PORT=7860
-EXPOSE 7860
-
 # Install Python dependencies
 COPY ./requirements.txt /code/requirements.txt
 RUN pip3 install --no-cache-dir --upgrade -r /code/requirements.txt
 
-# Copy package.json and package-lock.json (or yarn.lock)
+# Copy package.json and package-lock.json
 COPY frontend/package*.json frontend/
 RUN cd frontend && npm install
 
@@ -44,5 +44,4 @@ COPY app/ app/
 COPY server.py server.py
 
 #ENV LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libtcmalloc.so.4
-#CMD ["python3", "server.py"]
-CMD ["python3", "app/pipeline_test.py"]
+CMD ["python3", "server.py"]

app/bot.py

@@ -86,7 +86,7 @@ class DailyVision(EventHandler):
             self.__camera = Daily.create_camera_device("camera",
                                                        width = video_frame.width,
                                                        height = video_frame.height,
-                                                       color_format="RGB")
+                                                       color_format="RGBA")
             self.__client.update_inputs({
                 "camera": {
                     "isEnabled": True,
@@ -115,16 +115,15 @@ class DailyVision(EventHandler):
 
                 if video_frame:
                     image = Image.frombytes("RGBA", (video_frame.width, video_frame.height), video_frame.buffer)
-                    result_image = self.__pipeline.predict(params, image).convert("RGB")
+                    result_image = self.__pipeline.predict(params, image)
                     self.__camera.write_frame(result_image.tobytes())
                     #pil = Image.fromarray(result.render()[0], mode="RGB").tobytes()
-                    #self.__camera.write_frame(result_image.tobytes())
             except queue.Empty:
                 pass
 
     def on_video_frame(self, participant_id, video_frame):
         # Process ~15 frames per second (considering incoming frames at 30fps).
-        if time.time() - self.__time > 1: #0.05:
+        if time.time() - self.__time > 5: #0.05:
             self.__time = time.time()
             self.setup_camera(video_frame)
             self.__queue.put(video_frame)    

app/pipeline.py

@@ -181,8 +181,8 @@ class Pipeline:
             
             self.pipe(
                 prompt="warmup",
-                image=[Image.new("RGB", (640, 480))],
-                control_image=[Image.new("RGB", (640, 480))],
+                image=[Image.new("RGBA", (640, 480))],
+                control_image=[Image.new("RGBA", (640, 480))],
         )
 
     def predict(self, params: "Pipeline.InputParams", image) -> Image.Image:

requirements.txt

@@ -10,7 +10,7 @@ pillow
 pydantic
 utils
 psutil
-dotenv
+python-dotenv
 
 transformers==4.35.2
 torch==2.1.1