From 71d60daf3a5ca8e780635c5c75501cf3a57660e9 Mon Sep 17 00:00:00 2001 From: John Doe Date: Thu, 7 Jul 2022 18:08:33 -0400 Subject: [PATCH 1/3] Add timestamps for Luciano --- videos-list/29-luciano.md | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/videos-list/29-luciano.md b/videos-list/29-luciano.md index 6d7414f..60e089c 100644 --- a/videos-list/29-luciano.md +++ b/videos-list/29-luciano.md @@ -14,8 +14,25 @@ https://discourse.pymc.io/t/posterior-predictive-sampling-in-pymc3-by-luciano-pa ## Timestamps - 0:00 Start of event -- x:xx -- x:xx +- 0:22 Background on PyMC3 common workflow and posterior predictive distribution +- 2:56 What is this presentation about? +- 3:19 Simple model without posterior predictive problems +- 4:27 Create train and test data +- 5:12 Translate the math into PyMC3 +- 6:52 Visualize and plot the model and predictions +- 8:20 Make predictions on the new data +- 9:29 Shape problems +- 12:46 A simple extension of linear regression +- 17:26 View the results of the extension +- 17:50 Error on the test data +- 19:15 What happens if we try to marginalize label out of the model? +- 21:38 Shape problem with the test data +- 22:55 Solution, a functional approach +- 27:50 Factory functions aren’t a silver bullet +- 28:12 Manually trim the inferred posterior +- 29:22 Challenges with sampling +- 32:10 How can PyMC help? +- 32:45 Conclusion ## Note: help us add timestamps here https://github.com/pymc-devs/video-timestamps From a2f95603e8574621d7c0c1bc57c9c5aa312a3a34 Mon Sep 17 00:00:00 2001 From: Reshama Shaikh Date: Sun, 10 Jul 2022 11:42:15 -0400 Subject: [PATCH 2/3] add leading zero --- videos-list/29-luciano.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/videos-list/29-luciano.md b/videos-list/29-luciano.md index 60e089c..05cc067 100644 --- a/videos-list/29-luciano.md +++ b/videos-list/29-luciano.md @@ -13,7 +13,7 @@ Discourse Discussion https://discourse.pymc.io/t/posterior-predictive-sampling-in-pymc3-by-luciano-paz/6028 ## Timestamps -- 0:00 Start of event +00:00 Start of event - 0:22 Background on PyMC3 common workflow and posterior predictive distribution - 2:56 What is this presentation about? - 3:19 Simple model without posterior predictive problems From 8e254a64211b1f096318c7e899db88aa0615f15f Mon Sep 17 00:00:00 2001 From: Reshama Shaikh Date: Sun, 10 Jul 2022 11:43:30 -0400 Subject: [PATCH 3/3] formatting of time --- videos-list/29-luciano.md | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/videos-list/29-luciano.md b/videos-list/29-luciano.md index 05cc067..48a9b03 100644 --- a/videos-list/29-luciano.md +++ b/videos-list/29-luciano.md @@ -14,25 +14,25 @@ https://discourse.pymc.io/t/posterior-predictive-sampling-in-pymc3-by-luciano-pa ## Timestamps 00:00 Start of event -- 0:22 Background on PyMC3 common workflow and posterior predictive distribution -- 2:56 What is this presentation about? -- 3:19 Simple model without posterior predictive problems -- 4:27 Create train and test data -- 5:12 Translate the math into PyMC3 -- 6:52 Visualize and plot the model and predictions -- 8:20 Make predictions on the new data -- 9:29 Shape problems -- 12:46 A simple extension of linear regression -- 17:26 View the results of the extension -- 17:50 Error on the test data -- 19:15 What happens if we try to marginalize label out of the model? -- 21:38 Shape problem with the test data -- 22:55 Solution, a functional approach -- 27:50 Factory functions aren’t a silver bullet -- 28:12 Manually trim the inferred posterior -- 29:22 Challenges with sampling -- 32:10 How can PyMC help? -- 32:45 Conclusion +00:22 Background on PyMC3 common workflow and posterior predictive distribution +02:56 What is this presentation about? +03:19 Simple model without posterior predictive problems +04:27 Create train and test data +05:12 Translate the math into PyMC3 +06:52 Visualize and plot the model and predictions +08:20 Make predictions on the new data +09:29 Shape problems +12:46 A simple extension of linear regression +17:26 View the results of the extension +17:50 Error on the test data +19:15 What happens if we try to marginalize label out of the model? +21:38 Shape problem with the test data +22:55 Solution, a functional approach +27:50 Factory functions aren’t a silver bullet +28:12 Manually trim the inferred posterior +29:22 Challenges with sampling +32:10 How can PyMC help? +32:45 Conclusion ## Note: help us add timestamps here https://github.com/pymc-devs/video-timestamps