Docs

docs/Tutorials.rst

@@ -0,0 +1,13 @@
+Tutorials
+========================================================================================
+
+These notebooks serve as a tutorial for the core aspects of Deepdisc.
+
+.. toctree::
+
+    Installation <Tutorials/Installation>
+    Preprocesing <Tutorials/Preprocessing>
+    Understanding Deepdisc Configs <Tutorials/configs>
+    Loading Data with Deepdisc <Tutorials/data_loading>
+    Training with Deepdisc <Tutorials/training>
+    Predictions with Deepdisc <Tutorials/predictions>

docs/Tutorials/Preprocessing.ipynb

@@ -0,0 +1,41 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "04413870-779a-4fef-9eaa-76233ccdcbaf",
+   "metadata": {},
+   "source": [
+    "# Preprocessing"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "40d97c38-22c5-49e9-b849-e40fbd2ed1a8",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.18"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

docs/notebooks/data_loading.ipynb → docs/Tutorials/data_loading.ipynb

@@ -15,21 +15,28 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
   "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
    "name": "python",
-   "version": "3.8.9"
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.18"
   },
-  "orig_nbformat": 4,
   "vscode": {
    "interpreter": {
     "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
    }
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }

docs/notebooks/predictions.ipynb → docs/Tutorials/predictions.ipynb

@@ -1,7 +1,6 @@
 {
  "cells": [
   {
-   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
@@ -11,26 +10,42 @@
   {
    "cell_type": "markdown",
    "metadata": {},
+   "source": [
+    "test"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": []
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
   "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
    "name": "python",
-   "version": "3.8.9"
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.18"
   },
-  "orig_nbformat": 4,
   "vscode": {
    "interpreter": {
     "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
    }
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }

docs/Tutorials/training.ipynb

@@ -0,0 +1,37 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Training with Deepdisc "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.18"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}

docs/index.rst

@@ -47,4 +47,4 @@ Notes:
 
    Home page <self>
    API Reference <autoapi/index>
-   Notebooks <notebooks>
+   Tutorials <Tutorials>

src/deepdisc/astrodet/detectron.py

@@ -17,7 +17,6 @@
 import detectron2.data as data
 import detectron2.data.transforms as T
 
-# Yufeng Dec21 more import
 import matplotlib.pyplot as plt
 import torch
 from detectron2.data import build_detection_train_loader
@@ -391,7 +390,7 @@ def _setup_augs(self, augs, k: int):
         if self.cropaug is None:
             pass
         else:
-            augl.insert(0, self.cropaug)
+            augl.insert(-1, self.cropaug)
 
         return augl
 

src/deepdisc/data_format/augment_image.py

@@ -208,8 +208,8 @@ def dc2_train_augs(image):
 
         ],
         k=-1,
-        #cropaug=None,
-        cropaug=T.RandomCrop("relative", (0.5, 0.5))
+        cropaug=None,
+        #cropaug=T.RandomCrop("relative", (0.5, 0.5))
     )
     return augs
 

src/deepdisc/inference/match_objects.py

@@ -77,11 +77,13 @@ def get_object_coords(dataset_dict, outputs):
     ys = []
 
     for box in pred_boxes:
-        x,y = box[0], box[1]
+        w = box[2]-box[0]
+        h = box[3]-box[1]
+        x = box[0]+w//2
+        y = box[1]+h//2
         xs.append(x)
         ys.append(y)
 
-
     coords = wcs.pixel_to_world(xs,ys)
     ras = coords.ra.degree
     decs = coords.dec.degree

src/deepdisc/model/models.py

@@ -310,15 +310,15 @@ def _forward_redshift(self, features, instances):
 
             fcs = self.redshift_conv(features)
             pdfs = self.output_pdf(fcs)
-            zs = torch.tensor(np.linspace(0, 5, 200)).to(fcs.device)
+            zs = torch.tensor(np.linspace(0, 3, 300)).to(fcs.device)
             nin = torch.as_tensor(np.array([num_instances_per_img]))
             #probs = torch.zeros((num_instances_per_img[0], 200)).to(fcs.device)
 
             inds = np.cumsum(num_instances_per_img)
 
 
 
-            probs = torch.zeros((torch.sum(nin), 200)).to(fcs.device)
+            probs = torch.zeros((torch.sum(nin), 300)).to(fcs.device)
             for j, z in enumerate(zs):
                 probs[:, j] = pdfs.log_prob(z)
 
@@ -609,15 +609,15 @@ def _forward_redshift(self, features, instances):
 
             fcs = self.redshift_fc(features)
             pdfs = self.output_pdf(fcs)
-            zs = torch.tensor(np.linspace(0, 5, 200)).to(fcs.device)
+            zs = torch.tensor(np.linspace(0, 3, 300)).to(fcs.device)
             nin = torch.as_tensor(np.array([num_instances_per_img]))
             #probs = torch.zeros((num_instances_per_img[0], 200)).to(fcs.device)
 
             inds = np.cumsum(num_instances_per_img)
 
 
 
-            probs = torch.zeros((torch.sum(nin), 200)).to(fcs.device)
+            probs = torch.zeros((torch.sum(nin), 300)).to(fcs.device)
             for j, z in enumerate(zs):
                 probs[:, j] = pdfs.log_prob(z)
 
@@ -762,7 +762,7 @@ def _forward_redshift(self, features, instances):
 
             fcs = self.redshift_fc(features)
             pdfs = self.output_pdf(fcs)
-            zs = torch.tensor(np.linspace(0, 5, 200)).to(fcs.device)
+            zs = torch.tensor(np.linspace(-1, 5, 200)).to(fcs.device)
             nin = torch.as_tensor(np.array([num_instances_per_img]))
             #probs = torch.zeros((num_instances_per_img[0], 200)).to(fcs.device)