Thomas Capelle comments

Results 169 comments of


                                            Thomas Capelle

Big Refactor

I am writing an article based on your code and video: https://wandb.ai/capecape/train_sd/reports/Training-a-Conditional-Diffusion-model-from-scratch--VmlldzoyODMxNjE3

Create a `HOWTO` doc

I would also would know how to use scalene... Is there a jupyter notebook with examples?

Create a `HOWTO` doc

Thanks, please put his line on the readme: ``` # Load Scalene %load_ext scalene ```

I would like to collaborate with your project.

Actually, my model get's worse. But don't really know what params have to give to the attention layers, how many heads is reasonable.

I would like to collaborate with your project.

Not really, I am using sequences of images encoded by a CNN to generate a multi-variable timeseries.

How can I add a button

wow man, what a wonderful answer. Wow, i did exactly as you described, compiled the `TypeScript` (without knowing nothing about ts or js, just doing `tsc main_labeller.ts` and IT WORKED!!!!)...

UnstructuredMarkdownLoader strips Markdown formatting from documents, rendering MarkdownTextSplitter non-functional

What "kind of" works for me is doing: ```python from langchain.document_loaders import TextLoader from langchain.text_splitter import MarkdownTextSplitter # just ingest the Markdown file raw data = TextLoader(one_file) # split using...

Organise steps logic

Are these the types of QPS you were expecting @msaroufim ?

Organise steps logic

Running this command: ```bash tune run full_finetune_single_device --config my_dev/tiny_llama.yaml \ log_every_n_steps=5 \ max_steps_per_epoch=10 \ epochs=2 ``` you get a progress bar with only 10 steps: ``` INFO:torchtune.utils.logging:Dataset and Sampler are...

Organise steps logic

Hello guys, what else do we need here? Can I add these changes to the other recipes?