the Jupytext pairing disappeared somehow?

bzinberg · bzinberg · commit 17cb3f4a0b8c · 2020-04-02T12:39:25.000-04:00
diff --git a/tutorials/A Bottom-Up Introduction to Gen.ipynb b/tutorials/A Bottom-Up Introduction to Gen.ipynb
@@ -1148,6 +1148,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/A Bottom-Up Introduction to Gen.jl b/tutorials/A Bottom-Up Introduction to Gen.jl
@@ -1,15 +1,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # A Bottom-Up Introduction to Gen
@@ -265,7 +266,7 @@ trace[:initial_n]
 # +
 using Gen: get_choices
 
-display(get_choices(trace))
+get_choices(trace)
 # -
 
 # Gen also stores the arguments on which the function was called:
@@ -360,7 +361,7 @@ end
 
 using Gen: get_score
 trace = simulate(foo, (0.3,))
-display(get_choices(trace))
+get_choices(trace)
 println("log probability: $(get_score(trace))")
 
 # Check this value against the hand-computed value in our table above.
@@ -387,8 +388,6 @@ println("log probability: $(get_score(trace))")
 using Gen: choicemap
 
 constraints = choicemap((:a, true), (:c, false))
-
-display(constraints)
 # -
 
 # The `choicemap` constructor above took two elements of the form `(address, value`). This is equivalent to constructing an empty choice map and then populating it:
@@ -411,7 +410,7 @@ using Gen: generate
 #
 # Let's check that the trace actually agrees with our constraints:
 
-display(get_choices(trace))
+get_choices(trace)
 
 # We can also check the return value:
 
diff --git a/tutorials/Data-Driven Proposals in Gen.ipynb b/tutorials/Data-Driven Proposals in Gen.ipynb
@@ -1491,6 +1491,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/Data-Driven Proposals in Gen.jl b/tutorials/Data-Driven Proposals in Gen.jl
@@ -1,15 +1,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Tutorial: Data-Driven Proposals in Gen
@@ -280,7 +281,7 @@ end;
 # We can propose values of random choices from the proposal function using [`Gen.propose`](https://probcomp.github.io/Gen/dev/ref/gfi/#Gen.propose). This method returns the choices, as well as some other information, which we won't need for our purposes. For now, you can think of `Gen.propose` as similar to `Gen.initialize` except that it does not produce a full execution trace (only the choices), and it does not accept constraints. We can see the random choices for one run of the proposal on our data set:
 
 (proposed_choices, _, _) = Gen.propose(custom_dest_proposal, (measurements, scene))
-display(proposed_choices)
+proposed_choices
 
 # The function below runs the proposal 1000 times. For each run, it generates a trace of the model where the `:dest_x` and `:dest_y` choices are constrained to the proposed values, and then visualizes the resulting traces. We make the proposal a parameter of the function because we will be visualizing the output distribution of various proposals later in the notebook.
 
diff --git a/tutorials/Introduction to Modeling in Gen.ipynb b/tutorials/Introduction to Modeling in Gen.ipynb
@@ -1827,6 +1827,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/Introduction to Modeling in Gen.jl b/tutorials/Introduction to Modeling in Gen.jl
@@ -1,15 +1,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Tutorial: Introduction to Modeling in Gen
@@ -127,7 +128,7 @@ Gen.get_args(trace)
 
 # The trace also contains the value of the random choices, stored in a map from address to value called a *choice map*. This map is available through the API method [`get_choices`]():
 
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 # We can pull out individual values from this map using Julia's subscripting syntax `[...]`:
 
@@ -539,12 +540,12 @@ end;
 # We first show the addresses sampled by `bar`:
 
 trace = Gen.simulate(bar, ())
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 # And the addresses sampled by `bar_using_namespace`:
 
 trace = Gen.simulate(bar_using_namespace, ())
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 # Using `@trace` with a namespace can help avoid address collisions for complex models.
 
diff --git a/tutorials/Iterative inference in Gen.ipynb b/tutorials/Iterative inference in Gen.ipynb
@@ -1198,7 +1198,8 @@
  ],
  "metadata": {
   "jupytext": {
-   "encoding": "# -*- coding: utf-8 -*-"
+   "encoding": "# -*- coding: utf-8 -*-",
+   "formats": "ipynb,jl:light"
   },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
diff --git a/tutorials/Iterative inference in Gen.jl b/tutorials/Iterative inference in Gen.jl
@@ -2,15 +2,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Tutorial: Basics of Iterative Inference Programming in Gen
diff --git a/tutorials/Modeling with Black-Box Julia Code.ipynb b/tutorials/Modeling with Black-Box Julia Code.ipynb
@@ -678,6 +678,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/Modeling with Black-Box Julia Code.jl b/tutorials/Modeling with Black-Box Julia Code.jl
@@ -1,15 +1,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Cora Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: cora-julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Modeling with black-box Julia code 
@@ -166,7 +167,6 @@ end;
 planner_params = PlannerParams(300, 3.0, 2000, 1.)
 (trace, _) = Gen.generate(agent_model, (scene, dt, num_ticks, planner_params));
 choices = Gen.get_choices(trace)
-display(choices)
 
 # Next we explore the assumptions of the model by sampling many traces from the generative function and visualizing them. We have created a visualization specialized for this generative function for use with the `GenViz` package, in the directory `../inverse-planning/grid-viz/dist`. We have also defined a `trace_to_dict` method to convert the trace into a value that can be automatically serialized into a JSON string for use by the visualization:
 
diff --git a/tutorials/Modeling with TensorFlow Code.ipynb b/tutorials/Modeling with TensorFlow Code.ipynb
@@ -526,6 +526,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/Modeling with TensorFlow Code.jl b/tutorials/Modeling with TensorFlow Code.jl
@@ -1,15 +1,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Modeling with TensorFlow code
@@ -78,7 +79,7 @@ println(size(probs))
 
 #  Note that generative functions constructed using GenTF do not make random choices:
 
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 # The return value is the Julia value corresponding to the Tensor `y`:
 
@@ -107,7 +108,7 @@ end;
 
 # We see that the `net` generative function does not make any random choices. The only random choices are the digit labels for each input input:
 
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 # Before the `digit_model` will be useful for anything, it needs to be trained. We load some code for loading batches of MNIST training data.
 
diff --git a/tutorials/Particle Filtering in Gen.ipynb b/tutorials/Particle Filtering in Gen.ipynb
@@ -723,7 +723,8 @@
  ],
  "metadata": {
   "jupytext": {
-   "encoding": "# -*- coding: utf-8 -*-"
+   "encoding": "# -*- coding: utf-8 -*-",
+   "formats": "ipynb,jl:light"
   },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
diff --git a/tutorials/Particle Filtering in Gen.jl b/tutorials/Particle Filtering in Gen.jl
@@ -2,15 +2,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Tutorial: Particle Filtering in Gen
@@ -339,7 +340,7 @@ Gen.load_generated_functions()
 # We can understand the behavior of `chain` by getting a trace of it and printing the random choices:
 
 trace = Gen.simulate(chain, (4, State(0., 0., 0., 0.), 0.01, 0.01))
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 # We now write a new version of the generative model that invokes `chain` instead of using the Julia `for` loop:
 
@@ -375,7 +376,7 @@ Gen.load_generated_functions()
 # Let's generate a trace of this new model program to understand its structure:
 
 (trace, _) = Gen.generate(unfold_model, (4,))
-display(Gen.get_choices(trace))
+Gen.get_choices(trace)
 
 function unfold_particle_filter(num_particles::Int, zs::Vector{Float64}, num_samples::Int)
     init_obs = Gen.choicemap((:z0, zs[1]))    
diff --git a/tutorials/Reasoning About Regenerate.ipynb b/tutorials/Reasoning About Regenerate.ipynb
@@ -326,6 +326,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/Reasoning About Regenerate.jl b/tutorials/Reasoning About Regenerate.jl
@@ -1,15 +1,16 @@
 # ---
 # jupyter:
 #   jupytext:
+#     formats: ipynb,jl:light
 #     text_representation:
 #       extension: .jl
 #       format_name: light
 #       format_version: '1.5'
 #       jupytext_version: 1.3.3
 #   kernelspec:
-#     display_name: Julia 1.1.1
+#     display_name: Julia 1.4.0
 #     language: julia
-#     name: julia-1.1
+#     name: julia-1.4
 # ---
 
 # # Reasoning About Regenerate
@@ -73,7 +74,7 @@ using Gen: regenerate, select, NoChange
 # +
 using Gen: get_choices
 
-display(get_choices(trace))
+get_choices(trace)
 # -
 
 # Re-run the regenerate command until you get a trace where `a` is `false`. Note that the address `b` doesn't appear in the resulting trace. Then, run the command again until you get a trace where `a` is `true`. Note that now there is a value for `b`. This value of `b` was sampled along with the new value for `a`---`regenerate` will regenerate new values for the selected adddresses, but also any new addresses that may be introduced as a consequence of stochastic control flow.
@@ -205,7 +206,7 @@ log((0.7 * 0.9)/(0.3 * 0.4 * 0.2)) + log((0.3 * 0.4)/(0.7))
 
 trace, weight = generate(foo, (0.3,), choicemap((:a, true), (:b, false), (:c, true)));
 (trace, weight, retdiff) = regenerate(trace, (0.3,), (NoChange(),), select(:a));
-display(get_choices(trace))
+get_choices(trace)
 println("weight: $weight");
 
 # Aren't we glad this is automated by Gen!
diff --git a/tutorials/Scaling with Combinators and the Static Modeling Language.ipynb b/tutorials/Scaling with Combinators and the Static Modeling Language.ipynb
@@ -795,6 +795,9 @@
   }
  ],
  "metadata": {
+  "jupytext": {
+   "formats": "ipynb,jl:light"
+  },
   "kernelspec": {
    "display_name": "Julia 1.4.0",
    "language": "julia",
diff --git a/tutorials/Scaling with Combinators and the Static Modeling Language.jl b/tutorials/Scaling with Combinators and the Static Modeling Language.jl
