# FROM argilla/argilla-hf-spaces:v2.8.0

# Copy the auth config section
# COPY .oauth.yaml /home/argilla/

# Comment this line to disable annotation progress sharing feature
# ENV ARGILLA_ENABLE_SHARE_YOUR_PROGRESS=1

# Uncoment this line to remove the persistence storage warning
#ENV ARGILLA_SHOW_HUGGINGFACE_SPACE_PERSISTENT_STORAGE_WARNING=false

FROM pytorch/pytorch:2.1.0-cuda11.8-cudnn8-runtime

# Install basic tools
RUN apt-get update && apt-get install -y git curl

# Install Python packages
RUN pip install --upgrade pip
RUN pip install \
    datasets \
    transformers \
    accelerate \
    peft \
    trl \
    spacy \
    chromadb \
    huggingface_hub \
    tensorboard

# Download spaCy model
RUN python -m spacy download en_core_web_sm

# Copy all files into the container
COPY . /workspace
WORKDIR /workspace

# Set the command to run your training script
CMD ["python", "train.py"]