Andrea Costantino

Is there any news on this? [Here](https://docs.fast.ai/distributed.html) on the fastai docs, they say that it is possible to use accelerate with fastai. I tried it and it works. I am...

Thanks for your prompt reply. I didn't mention the notebook. I am training on command line with accelerate, and on IDE (Spyder) without accelerate, using the exact same learning rate....

``` from fastai.vision.all import * from fastai.distributed import * from fastai.vision.models.xresnet import * seed = 42 np.random.seed(seed) torch.manual_seed(seed) set_seed(seed) path = rank0_first(untar_data, URLs.IMAGENETTE_160) dls = DataBlock( blocks=(ImageBlock, CategoryBlock), splitter=GrandparentSplitter(valid_name='val'), get_items=get_image_files,...

- `Accelerate` version: 0.10.0 - Platform: Linux-4.15.0-76-generic-x86_64-with-glibc2.27 - Python version: 3.9.12 - Numpy version: 1.22.3 - PyTorch version (GPU?): 1.11.0 (True) - `Accelerate` default config: - compute_environment: LOCAL_MACHINE - distributed_type:...

This is the output from CLI with fp16 set to no: > epoch train_loss valid_loss accuracy top_k_accuracy time > 0 2.037301 2.129598 0.251592 0.782786 00:09 > 1 1.847155 1.756253 0.390374...

Another try. Here I had to set batch_size = 128. I removed `to_fp16()` from the script, and `fp16` from accelerate config. Output from IDE: > epoch train_loss valid_loss accuracy top_k_accuracy...

Yep, I also did some testing at the time and I realized the batch size problem. But thanks for looking into this! Indeed, I guess that more clear information about...

If anyone incurs in the same problem, this is the code I used to fix the offending files: ``` import json import glob from shutil import copyfile import os, stat...

I'm encountering a persistent memory problem, similar to one reported years ago in this issue, where the `pylsp` process progressively consumes all the RAM (64 GB) when using Spyder IDE,...

I am having a similar error (see [here](https://neurostars.org/t/node-error-on-mriqc-wf-dwimriqc-computeiqms-datasink/29188/3) for the crash report and error), but I do not have DWI scans. The problem for me seems to be the `'summary_bg_k':...