Francisco Yirá
@franciscoyira.com
Data Scientist in Toronto 🇨🇦🍁. Learning in public. Python, data viz, ML/AI, causal inference. Ravenclaw 🦅. 🏳️🌈
www.franciscoyira.com
www.franciscoyira.com
J'ai commencé à écrire des articles en français sur mon blog de science des données 🙂
www.franciscoyira.com/fr/post/vali...
www.franciscoyira.com/fr/post/vali...
November 18, 2025 at 12:41 AM
J'ai commencé à écrire des articles en français sur mon blog de science des données 🙂
www.franciscoyira.com/fr/post/vali...
www.franciscoyira.com/fr/post/vali...
Je vous présente mon nouveau clavier portable ✨
October 28, 2025 at 1:03 PM
Je vous présente mon nouveau clavier portable ✨
Je suis très heureux parce que le week-end dernier, un ami proche du Chili m'a rendu visite à Toronto. Nous sommes allés dans beaucoup d’endroits touristiques de la ville, même quelques-uns que je n'avais pas encore visités en deux ans que j'habite ici!
October 25, 2025 at 1:05 PM
Je suis très heureux parce que le week-end dernier, un ami proche du Chili m'a rendu visite à Toronto. Nous sommes allés dans beaucoup d’endroits touristiques de la ville, même quelques-uns que je n'avais pas encore visités en deux ans que j'habite ici!
DuckDB est vraiment un des outils les plus puissants pour analyser des données. J’ai commencé à l’utiliser pour faire le suivi de mes dépenses bancaires: je télécharge simplement les exports CSV de mes banques et je peux exécuter des commandes SQL pour savoir combien je dépense dans chaque magasin
September 6, 2025 at 5:41 PM
DuckDB est vraiment un des outils les plus puissants pour analyser des données. J’ai commencé à l’utiliser pour faire le suivi de mes dépenses bancaires: je télécharge simplement les exports CSV de mes banques et je peux exécuter des commandes SQL pour savoir combien je dépense dans chaque magasin
Je suis très heureux de partager que je viens de renouveler ma certification AWS Cloud Practitioner 😄
August 18, 2025 at 9:45 PM
Je suis très heureux de partager que je viens de renouveler ma certification AWS Cloud Practitioner 😄
Je n'ai pas beaucoup de temps pour lire récemment mais je viens de trouver ce livre à ma bibliothèque publique la plus proche et je ne peux pas m'empêcher de le prendre pour apprendre plus de l'histoire du Canada
August 15, 2025 at 11:27 PM
Je n'ai pas beaucoup de temps pour lire récemment mais je viens de trouver ce livre à ma bibliothèque publique la plus proche et je ne peux pas m'empêcher de le prendre pour apprendre plus de l'histoire du Canada
tell me more about how AGI is supossed to be around the corner (this Claude Sonnet **4**)
June 15, 2025 at 7:03 PM
tell me more about how AGI is supossed to be around the corner (this Claude Sonnet **4**)
the amount of real state taken up by Copilot when opening Word web is borderline infuriating (and there is literally no option/toggle for 'I don't want to see any of this')
June 14, 2025 at 3:07 PM
the amount of real state taken up by Copilot when opening Word web is borderline infuriating (and there is literally no option/toggle for 'I don't want to see any of this')
Just pulled the trigger with studio ghibli avatar thing and I love it
May 14, 2025 at 2:51 AM
Just pulled the trigger with studio ghibli avatar thing and I love it
This week's data science flashcard: the Breusch-Pagan test, one of the most common ways to test for the homogeneity of variance assumption (homoskedasticity) in linear regression
![A data science flashcard titled "BREUSCH-PAGAN TEST" on a dark blue background. The card explains that this is a statistical test that checks whether the variance of the residuals in a regression model is constant (i.e., tests for heteroskedasticity).
The flashcard is divided into sections explaining how the test works:
1. First, we fit our regression model and compute the residuals, shown with mathematical equations.
2. Then an auxiliary regression is run using the squared residuals as response variable, while keeping the original predictors.
3. The Breusch-Pagan statistic is just the F-statistic of this auxiliary regression, testing whether any predictors have explanatory power over the squared residuals. The F-statistic formula is provided.
4. This F-statistic is compared against a table to return a p-value. A value less than 0.05 usually results in rejecting the null hypothesis that the residuals have constant variance.
The card includes Python code showing how to implement the test:
```python
import statsmodels.api as sm
from statsmodels.stats.diagnostic import het_breuschpagan
model = sm.OLS(y, sm.add_constant(X)).fit()
bp_test = het_breuschpagan(model.resid, model.model.exog)
stat, pval = bp_test[0], bp_test[1]
print(f"BP stat: {stat:.2f}, p-value: {pval:.4f}")
```
A section titled "WHY IT MATTERS" explains that if residuals don't have constant variance, confidence intervals and p-values from standard OLS regression can't be trusted. The Breusch-Pagan test helps catch heteroskedasticity early and apply corrective measures like weighted least squares or robust standard errors.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ub5ufkyiedg3bxsix6nkp3gj/bafkreifiyc4w6aej6ozxg663g5vdixpcm2hzm7x2elojq65vujl3j6bgzi@jpeg)
May 11, 2025 at 1:21 PM
This week's data science flashcard: the Breusch-Pagan test, one of the most common ways to test for the homogeneity of variance assumption (homoskedasticity) in linear regression
I did it. I turned my nutrition into a linear optimization problem
May 4, 2025 at 4:30 PM
I did it. I turned my nutrition into a linear optimization problem
Hol Food macros look better - especially having zero trans fats, more protein and less sugar (previously I though soylent was sugar free but it has a bunch of maltodextrin, which kind of worse?).
I think I'll suplement it with avocado oil for a lower-carb ratio
I think I'll suplement it with avocado oil for a lower-carb ratio
April 22, 2025 at 12:14 AM
Hol Food macros look better - especially having zero trans fats, more protein and less sugar (previously I though soylent was sugar free but it has a bunch of maltodextrin, which kind of worse?).
I think I'll suplement it with avocado oil for a lower-carb ratio
I think I'll suplement it with avocado oil for a lower-carb ratio
I'm switching from Soylent to Hol Food (aka Canadian Soylent), in part because of the Buy Canadian cause, but also because Soylent has become totally unreliable with order fulfillment
so, obviously, I had to run a micronutrient comparison between the two
so, obviously, I had to run a micronutrient comparison between the two
April 21, 2025 at 11:59 PM
I'm switching from Soylent to Hol Food (aka Canadian Soylent), in part because of the Buy Canadian cause, but also because Soylent has become totally unreliable with order fulfillment
so, obviously, I had to run a micronutrient comparison between the two
so, obviously, I had to run a micronutrient comparison between the two
Mood: studying for the Databricks certification at McDonald’s (loving their new Oat milk option for coffee!)
April 20, 2025 at 4:57 PM
Mood: studying for the Databricks certification at McDonald’s (loving their new Oat milk option for coffee!)
I think I have a love/hate relationship with the (objectively ugly?) artwork at Union Station
April 15, 2025 at 1:26 AM
I think I have a love/hate relationship with the (objectively ugly?) artwork at Union Station
Another nice Sunday morning tinkering with my Flask app for statistical process control :) (while enjoying a maple latte 🍁☕️)
Today I added a toolbar with user-defined specification limits and cut-off period to compare before/after metrics
Today I added a toolbar with user-defined specification limits and cut-off period to compare before/after metrics
April 13, 2025 at 4:30 PM
Another nice Sunday morning tinkering with my Flask app for statistical process control :) (while enjoying a maple latte 🍁☕️)
Today I added a toolbar with user-defined specification limits and cut-off period to compare before/after metrics
Today I added a toolbar with user-defined specification limits and cut-off period to compare before/after metrics
I’m starting a series where I share data science flashcards from my personal study practice that I find particularly relevant/useful.
First up: AdaBoost vs. Gradient Boosting. Can you recall what sets them apart? Find the answer in the thread :)
#datascience #learninginpublic
First up: AdaBoost vs. Gradient Boosting. Can you recall what sets them apart? Find the answer in the thread :)
#datascience #learninginpublic
April 12, 2025 at 3:43 PM
I’m starting a series where I share data science flashcards from my personal study practice that I find particularly relevant/useful.
First up: AdaBoost vs. Gradient Boosting. Can you recall what sets them apart? Find the answer in the thread :)
#datascience #learninginpublic
First up: AdaBoost vs. Gradient Boosting. Can you recall what sets them apart? Find the answer in the thread :)
#datascience #learninginpublic
Just got back from #TechTO, and the energy was truly something else. There's something magnetic about being in a room full of builders willing bold tech ideas into existence and giving their whole soul to it
This kind of passion is exactly why I chose this city to grow and build my career
This kind of passion is exactly why I chose this city to grow and build my career
April 8, 2025 at 2:47 AM
Just got back from #TechTO, and the energy was truly something else. There's something magnetic about being in a room full of builders willing bold tech ideas into existence and giving their whole soul to it
This kind of passion is exactly why I chose this city to grow and build my career
This kind of passion is exactly why I chose this city to grow and build my career
Enjoyed some “vibe-coding” this Saturday by continuing working on my Flask Python app to detect and show unusual variations in Statistical Process Control charts 📈📊 (applying what I learnt in the Lean Six Sigma training I just completed). Hope to deploy and share soon! 😃
March 23, 2025 at 1:43 AM
Enjoyed some “vibe-coding” this Saturday by continuing working on my Flask Python app to detect and show unusual variations in Statistical Process Control charts 📈📊 (applying what I learnt in the Lean Six Sigma training I just completed). Hope to deploy and share soon! 😃
Happy to share that my sleep schedule is now a 2-sigma process
March 17, 2025 at 10:22 PM
Happy to share that my sleep schedule is now a 2-sigma process
Decided to give Cursor a try and built this quick prototype of Control Charts generation with Flask+Dash, despite never having used those frameworks before
(End goal is to have a public facing tool that checks for the 8 rules of “in-control processes” according to Lean Six Sigma )
(End goal is to have a public facing tool that checks for the 8 rules of “in-control processes” according to Lean Six Sigma )
March 16, 2025 at 3:29 PM
Decided to give Cursor a try and built this quick prototype of Control Charts generation with Flask+Dash, despite never having used those frameworks before
(End goal is to have a public facing tool that checks for the 8 rules of “in-control processes” according to Lean Six Sigma )
(End goal is to have a public facing tool that checks for the 8 rules of “in-control processes” according to Lean Six Sigma )
My favourite part of going to the Reference Library is stumbling upon random books and sections while heading to the washroom
March 15, 2025 at 6:46 PM
My favourite part of going to the Reference Library is stumbling upon random books and sections while heading to the washroom
Chuffed to start a new meetup in #Toronto: 𝐒𝐡𝐮𝐭 𝐔𝐩 𝐚𝐧𝐝 𝐂𝐨𝐝𝐞! It will be a bi-weekly session for focused, distraction-free coding at Second Cup Café, followed by networking and project sharing.
More details here: meetu.ps/e/NX8yk/xTng...
More details here: meetu.ps/e/NX8yk/xTng...
March 12, 2025 at 10:10 PM
Chuffed to start a new meetup in #Toronto: 𝐒𝐡𝐮𝐭 𝐔𝐩 𝐚𝐧𝐝 𝐂𝐨𝐝𝐞! It will be a bi-weekly session for focused, distraction-free coding at Second Cup Café, followed by networking and project sharing.
More details here: meetu.ps/e/NX8yk/xTng...
More details here: meetu.ps/e/NX8yk/xTng...
Trying out Miznon, the new pita/mediterranean restaurant at Union Station (got the eggs-no-steak pita sandwich 🥙 😋)
March 3, 2025 at 5:49 PM
Trying out Miznon, the new pita/mediterranean restaurant at Union Station (got the eggs-no-steak pita sandwich 🥙 😋)