From e87886f8781e54c0dde00a9e80d5da8c3c3a2b2e Mon Sep 17 00:00:00 2001 From: Evan Date: Sun, 4 Jan 2026 12:13:30 -0600 Subject: [PATCH 1/3] assignment 1 for sampling --- .../a1_sampling_and_reproducibility.ipynb | 33 +++++++++++++++---- 1 file changed, 26 insertions(+), 7 deletions(-) diff --git a/02_activities/assignments/a1_sampling_and_reproducibility.ipynb b/02_activities/assignments/a1_sampling_and_reproducibility.ipynb index 11852458..c6da55b2 100644 --- a/02_activities/assignments/a1_sampling_and_reproducibility.ipynb +++ b/02_activities/assignments/a1_sampling_and_reproducibility.ipynb @@ -16,7 +16,13 @@ "cell_type": "markdown", "id": "4ea73db3", "metadata": {}, - "source": [] + "source": [ + "The first several lines of code follow all of the imports is to set up a dataframe of what I think is people that attended the wedding and people that attended a brunch. The first part where you actually do sampling is infecting the random set of people. Although everyone in the data frame initially (that is everyone who went to the wedding and brunch is in your sampling frame) has an equal chance of being sampled, this is sampling without replacement, as your replace argument is set to FALSE. This makes sense, given the fact that if someone is infected, there is no point in reinfecting them. \n", + "\n", + "Following this, random sampling occurs again for the primary and secondary tracing. within your primary tracing, your sampling frame is people already infected, and so you are trying to decide which one of those you want to trace. You have zoomed in, moving away from the entire dataframe of people. In this case, your sample is likely much smaller as well. The code following that is not so much sampling, but just calculating propotions. \n", + "\n", + "Finally in the end, you created 1000 simulations of this. In terms of the 1000 simulations, you end up getting a red \"traced\" graph that approximates a normal distribution with a left tail, while the blue \"infect\" graph has less of a tail approximating a normal distribution" + ] }, { "cell_type": "markdown", @@ -30,7 +36,9 @@ "cell_type": "markdown", "id": "4cf5d993", "metadata": {}, - "source": [] + "source": [ + "I tried running simulation of 10, 100, and 1000. In the 10, it did not present any coherent distribution, but possible a square or uniform-type distribution given the small simulation size. In the 100 simulations, the graph started to look like the 1000. You will notice " + ] }, { "cell_type": "markdown", @@ -56,10 +64,21 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "id": "ab8587a0", "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "# Import necessary libraries\n", "import pandas as pd\n", @@ -130,8 +149,8 @@ "\n", " return p_wedding_infections, p_wedding_traces\n", "\n", - "# Run the simulation 1000 times\n", - "results = [simulate_event(m) for m in range(1000)]\n", + "# Run the simulation 100 times\n", + "results = [simulate_event(m) for m in range(100)]\n", "props_df = pd.DataFrame(results, columns=[\"Infections\", \"Traces\"])\n", "\n", "# Plotting the results\n", @@ -207,7 +226,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.13.0" + "version": "3.11.9" } }, "nbformat": 4, From 8f4a3040f5c0478cf478fac0e2796c3aa62b4e6e Mon Sep 17 00:00:00 2001 From: Evan Date: Thu, 8 Jan 2026 16:54:53 -0600 Subject: [PATCH 2/3] new pull request --- .../a1_sampling_and_reproducibility.ipynb | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/02_activities/assignments/a1_sampling_and_reproducibility.ipynb b/02_activities/assignments/a1_sampling_and_reproducibility.ipynb index c6da55b2..0cb5660b 100644 --- a/02_activities/assignments/a1_sampling_and_reproducibility.ipynb +++ b/02_activities/assignments/a1_sampling_and_reproducibility.ipynb @@ -21,7 +21,7 @@ "\n", "Following this, random sampling occurs again for the primary and secondary tracing. within your primary tracing, your sampling frame is people already infected, and so you are trying to decide which one of those you want to trace. You have zoomed in, moving away from the entire dataframe of people. In this case, your sample is likely much smaller as well. The code following that is not so much sampling, but just calculating propotions. \n", "\n", - "Finally in the end, you created 1000 simulations of this. In terms of the 1000 simulations, you end up getting a red \"traced\" graph that approximates a normal distribution with a left tail, while the blue \"infect\" graph has less of a tail approximating a normal distribution" + "Finally in the end, you created 1000 simulations of this. In terms of the 1000 simulations, you end up getting a red \"traced\" graph that approximates a normal distribution with a left tail, while the blue \"infect\" graph has less of a tail approximating a normal distribution. When you run 1000, you also notice that the red graph approximates the blue one. I can only imagine that if you run 10,000 simulations, this approximation would become closer. " ] }, { @@ -37,7 +37,7 @@ "id": "4cf5d993", "metadata": {}, "source": [ - "I tried running simulation of 10, 100, and 1000. In the 10, it did not present any coherent distribution, but possible a square or uniform-type distribution given the small simulation size. In the 100 simulations, the graph started to look like the 1000. You will notice " + "I tried running simulation of 10, 100, and 1000. In the 10, it did not present any coherent distribution, but possible a square or uniform-type distribution given the small simulation size. In the 100 simulations, the graph started to look like the 1000. You will notice particularly in the traced \"red\" part that there is a distribution approximating a normal distribution with a skew and tail to the left. The f ew results in this tail essentially saw, the lower the frequency, the lower the proportion of cases! As you increase the \"n\" number of simulations, red becomes a closer approximation of the blue. The traced v. infected in other words. Lastly, I will mention that the results are not quite repoducible, because there is no seed set! So each time you run it, the mean of the histograms may shift, or the spreadness and kurtosis of the distributions can vary! " ] }, { @@ -52,7 +52,9 @@ "cell_type": "markdown", "id": "77613cc3", "metadata": {}, - "source": [] + "source": [ + "I simply added \"np.random.seed(123)\". By adding the seed, it ensures that each time we are running a similar simulation producing similar results. When i run the code multiple times, it presents the same output! Try it for yourself! " + ] }, { "cell_type": "markdown", @@ -64,13 +66,13 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 12, "id": "ab8587a0", "metadata": {}, "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -150,6 +152,7 @@ " return p_wedding_infections, p_wedding_traces\n", "\n", "# Run the simulation 100 times\n", + "np.random.seed(123) # For reproducibility\n", "results = [simulate_event(m) for m in range(100)]\n", "props_df = pd.DataFrame(results, columns=[\"Infections\", \"Traces\"])\n", "\n", From 1a1610c61c55cf0638fb528f9a9c538d7254bf54 Mon Sep 17 00:00:00 2001 From: Evan Date: Thu, 8 Jan 2026 16:59:45 -0600 Subject: [PATCH 3/3] another pull request since previous changes did not get uploaded --- .../a1_sampling_and_reproducibility.ipynb | 32 ++++++++++++++++--- 1 file changed, 28 insertions(+), 4 deletions(-) diff --git a/02_activities/assignments/a1_sampling_and_reproducibility.ipynb b/02_activities/assignments/a1_sampling_and_reproducibility.ipynb index 0cb5660b..e9786172 100644 --- a/02_activities/assignments/a1_sampling_and_reproducibility.ipynb +++ b/02_activities/assignments/a1_sampling_and_reproducibility.ipynb @@ -17,9 +17,33 @@ "id": "4ea73db3", "metadata": {}, "source": [ - "The first several lines of code follow all of the imports is to set up a dataframe of what I think is people that attended the wedding and people that attended a brunch. The first part where you actually do sampling is infecting the random set of people. Although everyone in the data frame initially (that is everyone who went to the wedding and brunch is in your sampling frame) has an equal chance of being sampled, this is sampling without replacement, as your replace argument is set to FALSE. This makes sense, given the fact that if someone is infected, there is no point in reinfecting them. \n", + "I am splitting my answer into 4 steps with detailed answers for each subsection: \n", "\n", - "Following this, random sampling occurs again for the primary and secondary tracing. within your primary tracing, your sampling frame is people already infected, and so you are trying to decide which one of those you want to trace. You have zoomed in, moving away from the entire dataframe of people. In this case, your sample is likely much smaller as well. The code following that is not so much sampling, but just calculating propotions. \n", + "Step 1: Setting Up the Dataframe \n", + "Function Used: pd.DataFrame\n", + "Sample Size: 1000 individuals\n", + "Sampling Frame: Everyone in attendance (whole population) of the two events, brunches and weddings \n", + "Procedure: The first several lines of code following all of the imports are to set up a dataframe of what I think is people that attended the wedding and people that attended a brunch. The script creates n= 1000, with 800 individuals at the brunch and 200 at the wedding \n", + "Underlying Distribution: deterministic allocation (200 weddings, 800 brunches) \n", + "\n", + "\n", + "Step 2: Infecting a Random Subset of Individuals\n", + "Functions used: np.random.choice\n", + "Sample Size: Attack RATE = 0.10, so 0.10 *1000 = 100 individuals sampled (or infected)\n", + "Sampling Frame: Everyone in attendance (whole population) of the two events could have been infected. Although everyone in the data frame initially (that is everyone who went to the wedding and brunch is in your sampling frame) has an equal chance of being sampled, this is sampling without replacement, as your replace argument is set to FALSE. This makes sense, given the fact that if someone is infected, there is no point in reinfecting them. \n", + "Underlying distribution: Technically uniform, and could vary given that the code does not specify a specific proportion in wedding/brunch infection. \n", + "\n", + "\n", + "Step 3: Primary Contact Tracing \n", + "Functions used: np.random.rand\n", + "Sample Size/Sampling Frame: Every 100 infected individual gets sampled, given that we are assigning a 0/1 to each of them. The 0.20 probability helps inform us which ones are traced and which ones are not! \n", + "Within your primary tracing, your sampling frame is people already infected, and so you are trying to decide which one of those you want to trace. You have zoomed in, moving away from the entire dataframe of people. In this case, your sample is likely much smaller as well. \n", + "\n", + "Step 4: Secondary contact tracing \n", + "Functions used: value_counts() - used to count event types!\n", + "Sample Size/Sampling: our sampling are all of those that got traced. I estimate this sample size to be ~20, though of course 0.2 is just a probability, so depending on our run, we could have 18 or 19 or 21 people traced! \n", + "\n", + "The code following that is not so much sampling, but just calculating propotions. \n", "\n", "Finally in the end, you created 1000 simulations of this. In terms of the 1000 simulations, you end up getting a red \"traced\" graph that approximates a normal distribution with a left tail, while the blue \"infect\" graph has less of a tail approximating a normal distribution. When you run 1000, you also notice that the red graph approximates the blue one. I can only imagine that if you run 10,000 simulations, this approximation would become closer. " ] @@ -37,7 +61,7 @@ "id": "4cf5d993", "metadata": {}, "source": [ - "I tried running simulation of 10, 100, and 1000. In the 10, it did not present any coherent distribution, but possible a square or uniform-type distribution given the small simulation size. In the 100 simulations, the graph started to look like the 1000. You will notice particularly in the traced \"red\" part that there is a distribution approximating a normal distribution with a skew and tail to the left. The f ew results in this tail essentially saw, the lower the frequency, the lower the proportion of cases! As you increase the \"n\" number of simulations, red becomes a closer approximation of the blue. The traced v. infected in other words. Lastly, I will mention that the results are not quite repoducible, because there is no seed set! So each time you run it, the mean of the histograms may shift, or the spreadness and kurtosis of the distributions can vary! " + "I tried running simulation of 10, 100, and 1000. In the 10, it did not present any coherent distribution, but possible a square or uniform-type distribution given the small simulation size. In the 100 simulations, the graph started to look like the 1000. You will notice particularly in the traced \"red\" part that there is a distribution approximating a normal distribution with a skew and tail to the left. The few results in this tail essentially saw, the lower the frequency, the lower the proportion of cases! As you increase the \"n\" number of simulations, red becomes a closer approximation of the blue. The traced v. infected in other words. Lastly, I will mention that the results are not quite repoducible, because there is no seed set! So each time you run it, the mean of the histograms may shift, or the spreadness and kurtosis of the distributions can vary! Edit: In the first pull request, the answer got cut off " ] }, { @@ -53,7 +77,7 @@ "id": "77613cc3", "metadata": {}, "source": [ - "I simply added \"np.random.seed(123)\". By adding the seed, it ensures that each time we are running a similar simulation producing similar results. When i run the code multiple times, it presents the same output! Try it for yourself! " + "I simply added \"np.random.seed(123)\". By adding the seed, it ensures that each time we are running a similar simulation producing similar results. When i run the code multiple times, it presents the same output! Edited: For some reason, this answer did not show up on the first pull request, even though I wrote it! " ] }, {