docs: no mps training; how change cache dir

README.md

@@ -289,6 +289,13 @@ docker run -it --gpus all -v $HOME/.cache/huggingface:/root/.cache/huggingface -
 ## Running on Google Colab
 [Example Colab](https://colab.research.google.com/drive/1rOvQNs0Cmn_yU1bKWjCOHzGVDgZkaTtO?usp=sharing)
 
+## Q&A
+
+**Q**: How do I change the cache directory for where models are stored?
+
+**A**: Set the `HUGGINGFACE_HUB_CACHE` environment variable. 
+
+
 ## ChangeLog
 
 - perf: cli now has minimal overhead such that `aimg --help` runs in ~650ms instead of ~3400ms

imaginairy/cmds.py

@@ -875,6 +875,15 @@ def train_concept(
 
     You can find a lot of relevant instructions here: https://github.com/JoePenna/Dreambooth-Stable-Diffusion
     """
+    from imaginairy.utils import get_device
+
+    if "mps" in get_device():
+        click.secho(
+            "⚠️  MPS (MacOS) is not supported for training. Please use a GPU or CPU.",
+            fg="yellow",
+        )
+        return
+
     import os.path
 
     from imaginairy.train import train_diffusion_model