π Data Science Riddle
You're building a chatbot but it gives generic answers. What's the root issue?
You're building a chatbot but it gives generic answers. What's the root issue?
Anonymous Quiz
10%
Model is too deep
66%
Training data lacks context
11%
Wrong loss function
13%
Poor tokenization
π Data Science Riddle
Model Accuracy improves after dropping half the features. Why?
Model Accuracy improves after dropping half the features. Why?
Anonymous Quiz
13%
Model became smaller
70%
Overfitting reduced
12%
Data size shrank
6%
Training faster
β€3
Understanding the Forecast Statistics and Four Moments (4P).pdf
181.8 KB
Statistical Moments (M1, M2) for Data Analysis
Here are 5 curated PDFs diving into the mean (M1), variance (M2), and their applications in crafting research questions and sourcing data.
A channel member requested resources on this topic and we delivered.
If you have a topic you want resources on let us know, and weβll make it happen!
@datascience_bds
Here are 5 curated PDFs diving into the mean (M1), variance (M2), and their applications in crafting research questions and sourcing data.
A channel member requested resources on this topic and we delivered.
If you have a topic you want resources on let us know, and weβll make it happen!
@datascience_bds
β€8
π Data Science Riddle
Why do we use Batch Normalization?
Why do we use Batch Normalization?
Anonymous Quiz
31%
Speeds up training
40%
Prevents overfitting
9%
Adds non-linearity
20%
Reduces dataset size
β€3
π Data Science Riddle
Your object detection model misses small objects. Easiest fix?
Your object detection model misses small objects. Easiest fix?
Anonymous Quiz
24%
Use larger input images
27%
Add more classes
34%
Reduce learning rate
15%
Train longer
π€ AI that creates AI: ASI-ARCH finds 106 new SOTA architectures
ASI-ARCH β experimental ASI that autonomously researches and designs neural nets. It hypothesizes, codes, trains & tests models.
π‘ Scale:
1,773 experiments β 20,000+ GPU-hours.
Stage 1 (20M params, 1B tokens): 1,350 candidates beat DeltaNet.
Stage 2 (340M params): 400 models β 106 SOTA winners.
Top 5 trained on 15B tokens vs Mamba2 & Gated DeltaNet.
π Results:
PathGateFusionNet: 48.51 avg (Mamba2: 47.84, Gated DeltaNet: 47.32).
BoolQ: 60.58 vs 60.12 (Gated DeltaNet).
Consistent gains across tasks.
π Insights:
Prefers proven tools (gating, convs), refines them iteratively.
Ideas come from: 51.7% literature, 38.2% self-analysis, 10.1% originality.
SOTA share: self-analysis β to 44.8%, literature β to 48.6%.
@datascience_bds
ASI-ARCH β experimental ASI that autonomously researches and designs neural nets. It hypothesizes, codes, trains & tests models.
π‘ Scale:
1,773 experiments β 20,000+ GPU-hours.
Stage 1 (20M params, 1B tokens): 1,350 candidates beat DeltaNet.
Stage 2 (340M params): 400 models β 106 SOTA winners.
Top 5 trained on 15B tokens vs Mamba2 & Gated DeltaNet.
π Results:
PathGateFusionNet: 48.51 avg (Mamba2: 47.84, Gated DeltaNet: 47.32).
BoolQ: 60.58 vs 60.12 (Gated DeltaNet).
Consistent gains across tasks.
π Insights:
Prefers proven tools (gating, convs), refines them iteratively.
Ideas come from: 51.7% literature, 38.2% self-analysis, 10.1% originality.
SOTA share: self-analysis β to 44.8%, literature β to 48.6%.
@datascience_bds
β€4
π Databricks Tip: REPLACE vs MERGE
When updating Delta tables, youβve got two powerful options:
πΉ REPLACE TABLE β¦ ON
π Like throwing away the entire library and rebuilding it.
- Drops the old table & recreates it.
- Schema + data = fully replaced.
- β‘ Super fast but destructive (old data gone).
- β Best for full refreshes or schema changes.
πΉ MERGE
π Like updating only the books that changed.
- Works row by row.
- Updates, inserts, or deletes specific records.
- π Preserves unchanged data.
- β Best for incremental updates or CDC (Change Data Capture).
βοΈ Key Difference
- REPLACE = Start fresh with a new table.
- MERGE = Surgically update rows without losing the rest.
π Rule of thumb:
Use REPLACE for full rebuilds,
Use MERGE for incremental upserts.
#Databricks #DeltaLake
When updating Delta tables, youβve got two powerful options:
πΉ REPLACE TABLE β¦ ON
π Like throwing away the entire library and rebuilding it.
- Drops the old table & recreates it.
- Schema + data = fully replaced.
- β‘ Super fast but destructive (old data gone).
- β Best for full refreshes or schema changes.
πΉ MERGE
π Like updating only the books that changed.
- Works row by row.
- Updates, inserts, or deletes specific records.
- π Preserves unchanged data.
- β Best for incremental updates or CDC (Change Data Capture).
βοΈ Key Difference
- REPLACE = Start fresh with a new table.
- MERGE = Surgically update rows without losing the rest.
π Rule of thumb:
Use REPLACE for full rebuilds,
Use MERGE for incremental upserts.
#Databricks #DeltaLake
β€3