Merge pull request #1 from mlucy/feature/add-normalization-and-pca

Feature/add normalization and pca

.gitignore

@@ -0,0 +1 @@
+basilica/docs/_build/

basilica/basilica/__init__.py

@@ -101,8 +101,16 @@ def embed_images(self, images, model='generic', version='default',
         :type batch_size: int
         :param opts: Options specific to the model/version you chose.
         :type opts: Dict[str, Any]
+        :param opts["dimensions"]: Number of dimensions to return.  PCA will be used to reduce the number of dimensions with minimal information loss.
+        :type opts["dimensions"]: int
+        :param opts["normalize_l2"]: Whether or not each instance should be scaled to have unit L2 norm.  (This is sometimes useful for instance retrieval tasks.)  Defaults to False.
+        :type opts["normalize_l2"]: bool
+        :param opts["normalize_mean"]: Whether or not to normalize each feature in the embedding to have mean 0 across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_mean"]: bool
+        :param opts["normalize_variance"]: Whether or not to normalize each feature in the embedding to have unit variance across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_variance"]: bool
         :param timeout: HTTP timeout for request.
-        :type opts: int
+        :type timeout: int
         :returns: A generator of embeddings.
         :rtype: Generator[List[float]]
 
@@ -132,8 +140,16 @@ def embed_image(self, image, model='generic', version='default',
         :type version: str
         :param opts: Options specific to the model/version you chose.
         :type opts: Dict[str, Any]
+        :param opts["dimensions"]: Number of dimensions to return.  PCA will be used to reduce the number of dimensions with minimal information loss.
+        :type opts["dimensions"]: int
+        :param opts["normalize_l2"]: Whether or not each instance should be scaled to have unit L2 norm.  (This is sometimes useful for instance retrieval tasks.)  Defaults to False.
+        :type opts["normalize_l2"]: bool
+        :param opts["normalize_mean"]: Whether or not to normalize each feature in the embedding to have mean 0 across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_mean"]: bool
+        :param opts["normalize_variance"]: Whether or not to normalize each feature in the embedding to have unit variance across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_variance"]: bool
         :param timeout: HTTP timeout for request.
-        :type opts: int
+        :type timeout: int
         :returns: An embedding.
         :rtype: List[float]
 
@@ -159,8 +175,16 @@ def embed_image_files(self, image_files, model='generic', version='default',
         :type batch_size: int
         :param opts: Options specific to the model/version you chose.
         :type opts: Dict[str, Any]
+        :param opts["dimensions"]: Number of dimensions to return.  PCA will be used to reduce the number of dimensions with minimal information loss.
+        :type opts["dimensions"]: int
+        :param opts["normalize_l2"]: Whether or not each instance should be scaled to have unit L2 norm.  (This is sometimes useful for instance retrieval tasks.)  Defaults to False.
+        :type opts["normalize_l2"]: bool
+        :param opts["normalize_mean"]: Whether or not to normalize each feature in the embedding to have mean 0 across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_mean"]: bool
+        :param opts["normalize_variance"]: Whether or not to normalize each feature in the embedding to have unit variance across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_variance"]: bool
         :param timeout: HTTP timeout for request.
-        :type opts: int
+        :type timeout: int
         :returns: A generator of embeddings.
         :rtype: Generator[List[float]]
 
@@ -189,8 +213,17 @@ def embed_image_file(self, image_file, model='generic', version='default',
         :type version: str
         :param opts: Options specific to the model/version you chose.
         :type opts: Dict[str, Any]
+        :param opts["dimensions"]: Number of dimensions to return.  PCA will be used to reduce the number of dimensions with minimal information loss.
+        :type opts["dimensions"]: int
+        :param opts["normalize_l2"]: Whether or not each instance should be scaled to have unit L2 norm.  (This is sometimes useful for instance retrieval tasks.)  Defaults to False.
+        :type opts["normalize_l2"]: bool
+        :param opts["normalize_mean"]: Whether or not to normalize each feature in the embedding to have mean 0 across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_mean"]: bool
+        :param opts["normalize_variance"]: Whether or not to normalize each feature in the embedding to have unit variance across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_variance"]: bool
+
         :param timeout: HTTP timeout for request.
-        :type opts: int
+        :type timeout: int
         :returns: An embedding.
         :rtype: List[float]
 
@@ -216,8 +249,16 @@ def embed_sentences(self, sentences, model='english', version='default',
         :type batch_size: int
         :param opts: Options specific to the model/version you chose.
         :type opts: Dict[str, Any]
+        :param opts["dimensions"]: Number of dimensions to return.  PCA will be used to reduce the number of dimensions with minimal information loss.
+        :type opts["dimensions"]: int
+        :param opts["normalize_l2"]: Whether or not each instance should be scaled to have unit L2 norm.  (This is sometimes useful for instance retrieval tasks.)  Defaults to False.
+        :type opts["normalize_l2"]: bool
+        :param opts["normalize_mean"]: Whether or not to normalize each feature in the embedding to have mean 0 across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_mean"]: bool
+        :param opts["normalize_variance"]: Whether or not to normalize each feature in the embedding to have unit variance across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_variance"]: bool
         :param timeout: HTTP timeout for request.
-        :type opts: int
+        :type timeout: int
         :returns: A generator of embeddings.
         :rtype: Generator[List[float]]
 
@@ -243,8 +284,16 @@ def embed_sentence(self, sentence, model='english', version='default',
         :type version: str
         :param opts: Options specific to the model/version you chose.
         :type opts: Dict[str, Any]
+        :param opts["dimensions"]: Number of dimensions to return.  PCA will be used to reduce the number of dimensions with minimal information loss.
+        :type opts["dimensions"]: int
+        :param opts["normalize_l2"]: Whether or not each instance should be scaled to have unit L2 norm.  (This is sometimes useful for instance retrieval tasks.)  Defaults to False.
+        :type opts["normalize_l2"]: bool
+        :param opts["normalize_mean"]: Whether or not to normalize each feature in the embedding to have mean 0 across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_mean"]: bool
+        :param opts["normalize_variance"]: Whether or not to normalize each feature in the embedding to have unit variance across our sample dataset.  Defaults to True when `dimensions` is set, or False otherwise.
+        :type opts["normalize_variance"]: bool
         :param timeout: HTTP timeout for request.
-        :type opts: int
+        :type timeout: int
         :returns: An embedding.
         :rtype: List[float]