Merge pull request #164 from geometric-intelligence/ninamiolane-clean-01

Clean title

.github/workflows/test.yml

@@ -46,11 +46,6 @@ jobs:
         run: |
           pip install -e .[${{ matrix.flavor }}]
 
-      - name: Typecheck [mypy]
-        run: |
-          mypy -p neurometry
-        continue-on-error: true
-
       - name: Run tests [pytest]
         run: |
           pytest --cov --cov-report=xml:coverage.xml

.gitignore

@@ -7,7 +7,7 @@ neurometry/wandb/*
 neurometry/datasets/rnn_grid_cells/Dual agent path integration high res/*
 neurometry/datasets/rnn_grid_cells/Single agent path integration high res/*
 neurometry/curvature/grid-cells-curvature/models/xu_rnn/results/*
-
+notebooks/
 
 *viewer*
 *vizcortex*

tests/test_tutorials.py

@@ -24,7 +24,8 @@ def _exec_tutorial(path):
 
 
 TUTORIALS_DIR = "tutorials"
-paths = [f"{TUTORIALS_DIR}/01_methods_create_synthetic_data.ipynb"]
+paths = [
+    f"{TUTORIALS_DIR}/01_methods_create_synthetic_data.ipynb"]
 
 
 @pytest.mark.parametrize("path", paths)

tutorials/01_methods_create_synthetic_data.ipynb

@@ -4,14 +4,16 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Create Synthetic Neural Manifolds"
+    "# Create Synthetic Neural Manifolds\n",
+    "\n",
+    "This notebook explains how to use the module `synthetic` to generate points on neural manifolds in neural state space."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Set-up + Imports"
+    "### Set-up"
    ]
   },
   {
@@ -107,11 +109,11 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Plot circle $\\mathcal{S}^1$ and $N=3$ encoding vectors\n",
-    "\n",
+    "## Ring $\\mathcal{S}^1$ in Neural State Space $\\mathbb{R}^3$\n",
     "\n",
+    "We use $N=3$ encoding vectors to represent the recording of $N=3$ neurons.\n",
     "\n",
-    "We will project latent manifold $\\mathcal{S}^1$ (minimal embedding dimension $d=2$) into $N$-dimensional neural state space ($N=3$) with a $d\\times N$ matrix $A$. The entries of $A$ are randomly sampled from a uniform distribution $U[-1,1]$ and its columns are random encoding vectors. "
+    "We will project the latent manifold $\\mathcal{S}^1$ (minimal embedding dimension $d=2$) into $N$-dimensional neural state space ($N=3$) with a $d\\times N$ matrix $A$. The entries of $A$ are randomly sampled from a uniform distribution $U[-1,1]$ and its columns are random encoding vectors. "
    ]
   },
   {
@@ -19486,7 +19488,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Visualize cylinder ($\\mathcal{S}^1 \\times [0,1]$)"
+    "## Cylinder ($\\mathcal{S}^1 \\times [0,1]$) in Neural State Space $\\mathbb{R}^3$"
    ]
   },
   {
@@ -23422,7 +23424,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Encode 2-dim Flat torus in $N$-dimensional neural state space (N=3)."
+    "## Flat torus $\\mathcal{T}^2$ in Neural State Space $\\mathbb{R}^3$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We propose to encode the 2-dimensional flat torus in $N$-dimensional neural state space, where $N = 3$ neurons."
    ]
   },
   {
@@ -27372,7 +27381,16 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Extrinsic dimension of curved low-dim manifold"
+    "## Estimate Extrinsic Dimension of Neural Manifold"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here, we use `skdim` to estimate the extrinsic dimension of a ring (circle) manifold embedded in neural state space $\\mathbb{R}^N$ where the number of neurons $N$ can vary.\n",
+    "\n",
+    "We observe whether the estimated extrinsic dimension varies with $N$."
    ]
   },
   {
@@ -43358,7 +43376,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Synthetic neural activity encoding Van der Pol Oscillator"
+    "## Synthetic Neural Activity Encoding: Van der Pol Oscillator"
    ]
   },
   {

tutorials/02_methods_estimate_manifold_dimension.ipynb

@@ -4,7 +4,11 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Estimate Neural Dimensions"
+    "# Estimate Neural Dimensions\n",
+    "\n",
+    "This notebook explains how to use the module `dimension` to estimate the dimension (extrinsic or intrinsic) of the neural manifold embedded in neural state space $\\mathbb{R}^N$.\n",
+    "\n",
+    "In particular, this notebook tests several existing methods of dimension estimation on the synthetic manifolds from the module `datasets.synthetic` to evaluate their performance."
    ]
   },
   {
@@ -84,6 +88,7 @@
     "    skdim_dimension_estimation,\n",
     ")\n",
     "\n",
+    "import torch\n",
     "os.environ[\"GEOMSTATS_BACKEND\"] = \"pytorch\"\n",
     "import geomstats.backend as gs"
    ]
@@ -4812,17 +4817,17 @@
     }
    ],
    "source": [
-    "n_components = 4\n",
+    "# n_components = 4\n",
     "\n",
-    "X_pls = pls_transformed_X[n_components - 1]\n",
+    "# X_pls = pls_transformed_X[n_components - 1]\n",
     "\n",
-    "X_pca = pca_transformed_X[n_components - 1]\n",
+    "# X_pca = pca_transformed_X[n_components - 1]\n",
     "\n",
-    "from gtda.homology import WeakAlphaPersistence\n",
+    "# from gtda.homology import WeakAlphaPersistence\n",
     "\n",
-    "wa_pers = WeakAlphaPersistence(homology_dimensions=(0, 1, 2))\n",
+    "# wa_pers = WeakAlphaPersistence(homology_dimensions=(0, 1, 2))\n",
     "\n",
-    "diagrams_wa_pers = wa_pers.fit_transform([X_pca, X_pls])"
+    "# diagrams_wa_pers = wa_pers.fit_transform([X_pca, X_pls])"
    ]
   }
  ],