Tal

a merge request for adding TResNet: High Performance GPU-Dedicated Architecture https://github.com/mrT23/TResNet

Is the operation in SGE-Block equivalent to GroupNorm ?

1

Hi. I have two questions: Question 1: ``` t = t - t.mean(dim=1, keepdim=True) std = t.std(dim=1, keepdim=True) + 1e-5 t = t / std t = t.view(b, self.groups, h,...

Hi I developed a new family of GPU-dedicated models called TResNet https://arxiv.org/pdf/2003.13630.pdf https://github.com/mrT23/TResNet TResNet models give significantly better speed-accuracy tradeoff than other models out there like Resent, EfficientNet and MixNet...

(1) Use 'csv.Dialect.fieldnames' instead of hardcoding delimiter and quotechar. Why: makes the code more readable and easier to modify and extend. (2) Add error handling for missing ‘source’ column. Why: ...

Target is not converted to 'Long' error

7

i am trying to run simple PETS dataset training. when i use the following simple learner with label smooth: `learn = cnn_learner(dbunch, resnet34, metrics=[accuracy, top_k_accuracy], loss_func=LabelSmoothingCrossEntropy()) learn.fit_one_cycle(4)` i get the...

QLoRA and DDP

3

Thanks for the great repo i have two questions about training the models (specifically WizardCoder): 1. have you tried training with QLoRa, and not just LoRa ? are you considering...

custom_llm_provider is not working with acompletion, but is working with completion

3

Addresses the bug described here: https://github.com/BerriAI/litellm/issues/3480 I am not sure I understand the logic there and why 'completion' and 'acompletion' have a different treatment for the `custom_llm_provider` parameter, but this...

### What happened? When I give the `completion` endpoint a parameter: `kwargs["custom_llm_provider"] ='openai'` it uses it as expected (I am working with vllm) However, `acompletion` ignores this parameter, and crashes....

hi @balancap, thanks for the excellent repo. i want to recreate the training the lead to mobilenetV2 SOTA results. did you use for SOTA training the parameters in: https://github.com/balancap/tf-imagenet/blob/test_nets/docs/mobilenet_v2.md ?...

Tr/show config

3

### **PR Type** enhancement ___ ### **Description** - Added a new function `show_relevant_configurations` in `utils.py` to generate markdown text for relevant configurations. - Integrated this function across multiple tools (`pr_code_suggestions.py`,...