Upload folder using huggingface_hub

Dockerfile

@@ -1,37 +1,48 @@
-FROM pytorch/pytorch:2.4.0-cuda12.4-cudnn9-runtime
+FROM nvidia/cuda:12.1.1-devel-ubuntu22.04
 
 ENV DEBIAN_FRONTEND=noninteractive
-RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*
 
-# 1. Environment Hardening
-ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:$LD_LIBRARY_PATH
-ENV PATH=/home/user/.local/bin:$PATH
+# Install python3.10 and git
+RUN apt-get update && apt-get install -y \
+    python3.10 \
+    python3-pip \
+    python3.10-dev \
+    git \
+    && rm -rf /var/lib/apt/lists/*
 
-# 3. User Mapping (Part 1: Create user)
+# Make python3.10 the default python3
+RUN update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.10 1
+
+# Set up user
 RUN useradd -m -u 1000 user
 USER user
+ENV PATH="/home/user/.local/bin:${PATH}"
 WORKDIR /home/user/app
 
-# 2. Pip Setup (Part 1: Upgrade pip)
-RUN python3 -m pip install --no-cache-dir --upgrade pip
+# PIP Installation Sequence (Crucial)
+# a. pip install --upgrade pip
+RUN python3 -m pip install --upgrade pip
+
+# b. pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 --index-url https://download.pytorch.org/whl/cu121
+RUN python3 -m pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 --index-url https://download.pytorch.org/whl/cu121
+
+# c. pip uninstall -y torchao (ensure it is gone)
+RUN python3 -m pip uninstall -y torchao
 
-# Install Unsloth and training dependencies
-# 2. Pip Setup (Part 2: Use python3 -m pip)
-# 4. Dependency Protection (trl and xformers with --no-deps)
+# d. pip install --no-deps "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
+RUN python3 -m pip install --no-deps "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
+
+# e. pip install --no-deps xformers==0.0.27 trl==0.11.0 peft accelerate bitsandbytes
+RUN python3 -m pip install --no-deps xformers==0.0.27 trl==0.11.0 peft accelerate bitsandbytes
+
+# Install other requirements
 COPY --chown=user:user requirements.txt .
-RUN python3 -m pip install --no-cache-dir \
-    "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git" \
-    && python3 -m pip install --no-cache-dir --no-deps xformers \
-    && python3 -m pip install --no-cache-dir --no-deps trl \
-    && python3 -m pip install --no-cache-dir -r requirements.txt \
-    && python3 -m pip install --no-cache-dir \
-    peft accelerate bitsandbytes
-
-# 3. User Mapping (Part 2: chown for all COPY)
+RUN python3 -m pip install -r requirements.txt
+
+# Copy application code
 COPY --chown=user:user train.py .
 COPY --chown=user:user aegis_training_data_500.json .
 
 EXPOSE 7860
 
-# -u for unbuffered stdout so logs appear in real time in HF Space console
 CMD ["python3", "-u", "train.py"]

requirements.txt

@@ -1,8 +1,7 @@
-transformers>=4.45.0
+transformers==4.44.2
 huggingface_hub
 sentencepiece
 protobuf
 datasets
 scipy
 gradio
-safetensors