Update documentation #705

neuml · Jun 4, 2024 · f28225b · f28225b
1 parent 68fdc9c
commit f28225b
Show file tree

Hide file tree

Showing 3 changed files with 21 additions and 2 deletions.
diff --git a/docs/embeddings/configuration/general.md b/docs/embeddings/configuration/general.md
@@ -9,13 +9,17 @@ keyword: boolean
 
 Enables sparse keyword indexing for this embeddings.
 
+When enabled, this parameter creates a BM25 index for full text search. It also implicitly disables the [defaults](../vectors/#defaults) setting for vector search.
+
 ## hybrid
 ```yaml
 hybrid: boolean
 ```
 
 Enables hybrid (sparse + dense) indexing for this embeddings.
 
+When enabled, this parameter creates a BM25 index for full text search. It has no effect on the [defaults](../vectors/#defaults) or [path](../vectors/#path) settings.
+
 ## indexes
 ```yaml
 indexes: dict

diff --git a/docs/embeddings/configuration/vectors.md b/docs/embeddings/configuration/vectors.md
@@ -10,6 +10,13 @@ path: string
 Sets the path for a vectors model. When using a transformers/sentence-transformers model, this can be any model on the
 [Hugging Face Hub](https://huggingface.co/models) or a local file path. Otherwise, it must be a local file path to a word embeddings model.
 
+## defaults
+```yaml
+defaults: boolean
+```
+
+Uses default vector model path when enabled (default setting is True) and `path` is not provided. See [this link](../) for an example.
+
 ## method
 ```yaml
 method: transformers|sentence-transformers|llama.cpp|litellm|external|words

diff --git a/docs/faq.md b/docs/faq.md
@@ -91,7 +91,7 @@ SQLError: no such function: json_extract
 
 __Solution__
 
-Upgrade Python version as it doesn't have SQLite support for json_extract
+Upgrade Python version as it doesn't have SQLite support for `json_extract`
 
 ----------
 
@@ -101,7 +101,15 @@ Segmentation faults and similar errors on macOS
 
 __Solution__
 
-Disable OpenMP threading via the environment variable `export OMP_NUM_THREADS=1` or downgrade PyTorch to <= 1.12. See issue [#377](https://github.com/neuml/txtai/issues/377) for more.
+Set the following environment parameters.
+
+- Disable OpenMP threading via the environment variable `export OMP_NUM_THREADS=1`
+- Disable PyTorch MPS device via `export PYTORCH_MPS_DISABLE=1`
+- Disable llama.cpp metal via `export LLAMA_NO_METAL=1`
+
+If the issue is with a training pipeline.
+
+- Disable GPU training via `export ACCELERATE_USE_CPU=1`
 
 ----------