Andrew Kane

@baarkerlounger The end goal is to make data analysis and machine learning as enjoyable as possible in Ruby. The main difference from Daru is its based on Numo, for both...

@arbox I'm happy to chat on specific features / ideas, but want to give this a fresh look (and think there's limited benefit to reusing the codebase).

Wanted to voice my support for this as well. Python currently won't output on assignment. ```py a = 1 + 2 # no output 1 + 2 # output ```...

I have a branch to support this once #276 is merged. The basic code is: ```ruby module IRuby class PlainBackend def eval(code, store_history) @irb.context.evaluate(code, 0) @irb.context.last_value unless assignment_expression?(code) end def...

Hey @MatheusRich, thanks! Can you explain more about the 2nd part?

I haven't tried it with other languages, but if there are [models](https://huggingface.co/models) for the supported tasks, it should be possible (if a model uses a different tokenizer, may need to...

Hey @mmontagna, thanks for the PR! It's an interesting idea. Have you tried benchmarking it to see how much impact it has on performance?

Great, just tried it with pgbench locally and also see a significant improvement. ```sh pgbench -i -s 10 pgsync_bench1 pgbench -i -s 10 pgsync_bench2 pgsync pgbench_accounts --from pgsync_bench1 --to pgsync_bench2...

Good catch. I think it'll be an issue with any column types that don't support the equality operator (including custom types). I also think there may be an issue with...

Sounds good. fwiw, I think the null issue will affect correctness/cause it to sync less rows than desired, so you'll probably want to test different combinations of null in the...