Commenting out nevergrad-sweeper as it's causing problems again. Upgr…

…ading other things

README.md

@@ -5,11 +5,12 @@ High order and piecewise networks for natural language generation (see [here](ht
 layers here is a fully connected network where each link has multiple segments.  Only one segment in
 a link is active for each input so the network sparsity is determined by the number of segments. Although it looks like a standard MLP, the structure is more complicated and is a form of routing network with piecewise polynomials.
 
-
 ![image](images/language-interpolation-drawing.png)
 
 I'm interested in creating larger language models from an ensemble of smaller models.  This would give better flexibility in adding or removing specific sources.
 
+Currently working on sparse high-order transformers
+
 # Dataset
 
 Data from project Gutenberg are used, either single or multiple books.  Training is done on the character level.  A pytorch lightning data module for project Gutenberg has been implemented for data loading.

config/high_order_interpolation.yaml

@@ -13,7 +13,6 @@ data:
   characters_per_feature: 10
   max_features: 100
 
-
   train:
     filenames: null # ["books/the_raven.txt", "books/frankenstein.txt"]
 

examples/high_order_interpolation.py

@@ -27,7 +27,7 @@
 logging.getLogger().setLevel(logging.DEBUG)
 
 
-@hydra.main(config_path="../config", config_name="high_order_interpolation")
+@hydra.main(config_path="../config", config_name="high_order_interpolation", version_base="1.3")
 def run_language_interpolation(cfg: DictConfig):
     logger.info(OmegaConf.to_yaml(cfg))
     logger.info("Working directory : {}".format(os.getcwd()))

pyproject.toml

@@ -9,15 +9,15 @@ license = "MIT"
 python = ">=3.9,<4.0"
 torchsummary = "^1.5.1"
 omegaconf = "^2.1.1"
-hydra-core = "^1.2.0"
 gutenbergpy = {git = "https://github.com/jloveric/gutenbergpy.git"}
 torch-intermediate-layer-getter = "^0.1.post1"
 torch-optimizer = "^0.3.0"
-hydra-nevergrad-sweeper = "1.3.0.dev0"
+#hydra-nevergrad-sweeper = "1.3.0.dev0"
 pytorch-lightning = "^2.0.0"
 high-order-layers-torch = "^2.0.0"
 langchain = "^0.0.334"
 torch = "2.0.0"
+hydra-core = "^1.3.2"
 
 [tool.poetry.group.dev.dependencies]
 black = "^23.11.0"