Merge pull request #318 from RichieHakim/dev

v1.3
RichieHakim · Oct 2, 2024 · 5727fcc · 5727fcc
2 parents b8ee25c + ddc703a
commit 5727fcc
Show file tree

Hide file tree

Showing 32 changed files with 2,689 additions and 1,055 deletions.
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
@@ -145,7 +145,8 @@ jobs:
       run: |
         # pip install tox tox-gh-actions
         pip install pytest pytest-cov
-        python -m pytest -rP --capture=tee-sys --cov=roicat --cov-report=xml:coverage.xml --color=yes --ignore=tests/test_packages.py --ignore=tests/test_interactive.py
+        # python -m pytest -v -rP --capture=tee-sys --cov=roicat --cov-report=xml:coverage.xml --color=yes --ignore=tests/test_packages.py --ignore=tests/test_interactive.py
+        python -m pytest -v -rP --capture=tee-sys --cov=roicat --cov-report=xml:coverage.xml --color=yes
 
     - name: Upload coverage reports to Codecov
       uses: codecov/codecov-action@v3  ## this is a public action recognized by GitHub Actions

diff --git a/README.md b/README.md
@@ -64,22 +64,22 @@ computer.
 
 ### TRACKING: 
 - [Interactive
-  notebook](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/tracking/tracking_interactive_notebook.ipynb)
+  notebook](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/tracking/1_tracking_interactive_notebook.ipynb)
 - [Google
-  CoLab](https://githubtocolab.com/RichieHakim/ROICaT/blob/main/notebooks/colab/tracking/tracking_interactive_notebook.ipynb)
+  CoLab](https://githubtocolab.com/RichieHakim/ROICaT/blob/main/notebooks/colab/tracking/1_tracking_interactive_notebook.ipynb)
 - [Command line interface script](https://github.com/RichieHakim/ROICaT/blob/main/scripts/demo_run_tracking_pipeline.py)
 
 ### CLASSIFICATION:
 - [Interactive notebook -
-  Drawing](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/classify_by_drawingSelection.ipynb)
+  Drawing](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/A1_classify_by_drawingSelection.ipynb)
 - [Google CoLab -
-  Drawing](https://githubtocolab.com/RichieHakim/ROICaT/blob/main/notebooks/colab/classification/classify_by_drawingSelection_colab.ipynb)
+  Drawing](https://githubtocolab.com/RichieHakim/ROICaT/blob/main/notebooks/colab/classification/A1_classify_by_drawingSelection_colab.ipynb)
 - [Interactive notebook -
-  Labeling](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/labeling_interactive.ipynb)
+  Labeling](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/B1_labeling_interactive.ipynb)
 - [Interactive notebook - Train
-  classifier](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/classifier_train_interactive.ipynb)
+  classifier](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/B2_classifier_train_interactive.ipynb)
 - [Interactive notebook - Inference with
-  classifier](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/classifier_inference_interactive.ipynb)
+  classifier](https://github.com/RichieHakim/ROICaT/blob/main/notebooks/jupyter/classification/B3_classifier_inference_interactive.ipynb)
 
 **OTHER:** 
 - [Custom data importing
@@ -236,12 +236,29 @@ intercompatible, so if you have CUDA 11.8 drivers, you can install
 
 
 # TODO:
-- Some more integration tests
-- Switch to ONNX for ROINet
-- Add more documentation / tutorials
-- Make a GUI
-- Finish ROIextractors integration
-- Make a Docker container
-- Make a standard classifier
-- Write the paper
-- Make tweet about it
+#### algorithmic improvements:
+- [ ] Add in method to use more similarity metrics for tracking
+- [ ] Coordinate descent on each similarity metric
+- [ ] Add F and Fneu to data_roicat, dFoF and trace quality metric functions
+- [ ] Add in notebook for demonstrating using temporal similarity metrics (SWT on dFoF)
+- [ ] Make a standard classifier
+- [ ] Try other clustering methods
+- [ ] Make image aligner based on image similarity + RANSAC of centroids or s_SF
+- [ ] Better post-hoc curation metrics and visualizations
+#### code improvements:
+- [ ] Update automatic regression module (make new repo for it)
+- [ ] Switch to ONNX for ROINet
+- [ ] Some more integration tests
+- [ ] Add more documentation / tutorials
+- [ ] Make a GUI
+- [ ] Finish ROIextractors integration
+- [ ] Make a Docker container
+- [ ] Make colab demo notebook not require user data
+#### other:
+- [ ] Write the paper
+- [ ] Make tweet about it
+- [ ] Make a video or two on how to use it
+- [ ] Maybe use lightthetorch for torch installation
+- [ ] Better Readme
+- [ ] More documentation
+- [ ] Make a regression model for in-plane-ness
diff --git a/conftest.py b/conftest.py
@@ -52,7 +52,7 @@ def download_data_test_zip(directory):
         check_local_first=True, 
         check_hash=True, 
         hash_type='MD5', 
-        hash_hex=r'aec4284b7f0a25e5e8eec85ed659d532',
+        hash_hex=r'2fcd64902d3c71eb0a85bbdb15a7d68e',
         mkdir=True,
         allow_overwrite=True,
         write_mode='wb',

diff --git a/docs/source/inputsAndOutputs.rst b/docs/source/inputsAndOutputs.rst
@@ -14,8 +14,10 @@ Inputs
 Outputs
 #######
 
-The outputs of ROICaT are encapsulated in a ``results.pkl`` file, which is a
-Python dictionary containing the following fields:
+The outputs of ROICaT tracking are encapsulated in a set of files. The main file
+is a ``...tracking.results_all.richfile`` directory that can be loaded using
+``roicat.util.RichFile_ROICaT``. It is a python dictionary with the following
+fields:
 
 Clusters
 ~~~~~~~~
@@ -30,6 +32,8 @@ Clusters
 - **labels_bool_bySession:** Same as ``labels_bool``, but by session.
 - **labels_dict:** Dictionary mapping UCIDs to ROI indices. Keys are UCIDs,
   values are lists of ROI indices.
+- **quality_metrics:** Dictionary of quality metrics for each cluster. See below
+  section `Quality Metrics <quality-metrics>`_ for more details.
 
 ROIs
 ~~~~

diff --git a/...cation/classify_by_drawingSelection.ipynb → ...ion/A1_classify_by_drawingSelection.ipynb b/...cation/classify_by_drawingSelection.ipynb → ...ion/A1_classify_by_drawingSelection.ipynb
@@ -65,7 +65,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 14,
    "id": "aaeb2017-7988-47b1-96d6-cc9b0c68edf3",
    "metadata": {},
    "outputs": [],
@@ -87,7 +87,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 15,
    "id": "9904f613-2094-48bd-8265-cfd7e0aed001",
    "metadata": {},
    "outputs": [],
@@ -226,7 +226,8 @@
     "    ROI_images=data.ROI_images,  ## Input images of ROIs\n",
     "    um_per_pixel=data.um_per_pixel,  ## Resolution of FOV\n",
     "    pref_plot=False,  ## Whether or not to plot the ROI sizes\n",
-    ");"
+    ");\n",
+    "images = roinet.ROI_images_rs"
    ]
   },
   {
@@ -246,7 +247,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "roicat.visualization.display_toggle_image_stack(roinet.ROI_images_rs[:1000], image_size=(200,200))"
+    "roicat.visualization.display_toggle_image_stack(images[:1000], image_size=(200,200))"
    ]
   },
   {
@@ -286,14 +287,14 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "umap = UMAP(\n",
+    "model_umap = UMAP(\n",
     "    n_neighbors=25,\n",
     "    n_components=2,\n",
     "    n_epochs=400,\n",
     "    verbose=True,\n",
     "    densmap=False,\n",
     ")\n",
-    "emb = umap.fit_transform(roinet.latents)"
+    "emb = model_umap.fit_transform(roinet.latents)"
    ]
   },
   {
@@ -314,7 +315,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 75,
    "id": "802791ab-e0fc-424a-8320-d6a1e641f61d",
    "metadata": {},
    "outputs": [],
@@ -326,7 +327,7 @@
     "#     border_frac=0.05,\n",
     ")\n",
     "\n",
-    "images_overlay = roinet.ROI_images_rs[idx_images_overlay]"
+    "images_overlay = images[idx_images_overlay]"
    ]
   },
   {
@@ -374,33 +375,32 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 77,
    "id": "9005978f-a9b4-4a56-87d1-fe6844e238b9",
    "metadata": {},
    "outputs": [],
    "source": [
+    "## Save the layout to an SVG file if desired\n",
     "# roicat.helpers.export_svg_hv_bokeh(layout, '/home/rich/Desktop/umap_with_labels_dotsOnly.svg')"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 78,
    "id": "b9b82100",
    "metadata": {},
    "outputs": [],
    "source": [
     "n_sessions = len(data.ROI_images)\n",
     "idx_session_cat = np.concatenate([[ii]*data.ROI_images[ii].shape[0] for ii in range(n_sessions)])\n",
     "bool_good_cat = roicat.helpers.idx2bool(fn_get_indices(), length=len(idx_session_cat))\n",
-    "preds_good_sessions = [np.int64((bool_good_cat * (idx_session_cat==ii))[idx_session_cat==ii]) for ii in range(data.n_sessions)]\n",
+    "preds_good_sessions = roicat.util.JSON_List([[int(l) for l in labels] for labels in roicat.util.labels_to_labelsBySession(bool_good_cat, data.n_roi)])\n",
     "\n",
-    "classification_output = {\n",
-    "    'preds': preds_good_sessions,\n",
-    "    'spatialFootprints': data.spatialFootprints,\n",
-    "    'FOV_height': data.FOV_height,\n",
-    "    'FOV_width': data.FOV_width,\n",
-    "    'input_data': data.paths_stat,\n",
-    "    'um_per_pixel': data.um_per_pixel,\n",
+    "run_data = {\n",
+    "    \"data\": data.__dict__,\n",
+    "    \"roinet\": roinet.__dict__,\n",
+    "    \"umap_embedding\": emb,\n",
+    "    \"preds\": preds_good_sessions,\n",
     "}"
    ]
   },
@@ -420,7 +420,7 @@
    "outputs": [],
    "source": [
     "print(f\"Number of 'good' and 'bad' ROIs from each session:\")\n",
-    "print([f\"good: {p.sum()} / bad: {(p!=1).sum()}\" for p in preds_good_sessions])"
+    "print([f\"good: {np.array(p).sum()} / bad: {(np.array(p)!=1).sum()}\" for p in preds_good_sessions])"
    ]
   },
   {
@@ -465,27 +465,21 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "fe17605e-96da-4d5d-b111-0d96bb0888b4",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "mouse = 'mouse_0916N'"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "ca2f1539",
+   "id": "028db758",
    "metadata": {},
    "outputs": [],
    "source": [
-    "dir_save = f'/media/rich/bigSSD/analysis_data/face_rhythm/{mouse}/'\n",
-    "filename_save = mouse\n",
+    "## Define the directory to save the results to\n",
+    "dir_save = '/media/rich/bigSSD/data_tmp/test_data/'\n",
+    "name_save = 'mouse_1'\n",
     "\n",
-    "path_save = str(Path(dir_save).resolve() / (filename_save + '.ROICaT.classification_drawn.results' + '.pkl'))\n",
-    "print(f'path_save: {path_save}')\n",
+    "paths_save = {\n",
+    "    'preds': str(Path(dir_save) / f'{name_save}.classification_drawn.preds.json'),\n",
+    "    'run_data': str(Path(dir_save) / f'{name_save}.classification_drawn.run_data.richfile'),\n",
+    "}\n",
     "\n",
-    "roicat.helpers.pickle_save(classification_output, path_save, mkdir=True)"
+    "roicat.helpers.json_save(obj=preds_good_sessions, filepath=paths_save['preds'])\n",
+    "roicat.util.RichFile_ROICaT(path=paths_save['run_data']).save(run_data, overwrite=True)"
    ]
   },
   {
@@ -522,7 +516,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.3"
+   "version": "3.11.9"
   }
  },
  "nbformat": 4,