Fix: trl==0.15.2 (has GRPO, no vllm/FSDP dep)

Dockerfile

@@ -20,16 +20,17 @@ RUN pip install --no-cache-dir \
     torch --index-url https://download.pytorch.org/whl/cu121
 
 RUN pip install --no-cache-dir \
-    "transformers==4.46.3" \
-    "trl==0.12.2" \
-    "peft==0.13.2" \
+    "transformers>=4.46,<4.48" \
+    "trl==0.15.2" \
+    "peft>=0.13,<0.15" \
     "datasets>=2.18" \
-    "accelerate==0.34.2" \
+    "accelerate>=0.34,<0.36" \
     huggingface_hub
 
-# Verify imports work
-RUN python3 -c "from trl import SFTTrainer, SFTConfig, GRPOConfig, GRPOTrainer; print('TRL imports OK')"
-RUN python3 -c "import torch; print('CUDA available:', torch.cuda.is_available())"
+# Verify all imports work at build time
+RUN python3 -c "from trl import SFTTrainer, SFTConfig, GRPOConfig, GRPOTrainer; print('TRL OK')"
+RUN python3 -c "from peft import LoraConfig; print('PEFT OK')"
+RUN python3 -c "import torch; print('CUDA:', torch.cuda.is_available())"
 
 COPY train_on_hf.py .
 RUN chmod -R 777 /app