MrYxJ/calculate-flops.pytorch

• highreadme#1

  Reposition the README Introduction to highlight LLM/Hugging Face focus

  原因：

  当前

  This tool(calflops) is designed to compute the theoretical amount of FLOPs(floating-point operations)、MACs(multiply-add operations) and Parameters in all various neural networks, such as Linear、 CNN、 RNN、 GCN、**Transformer(Bert、LlaMA etc Large Language Model)**，even including **any custom models** via ```torch.nn.function.*``` as long as based on the Pytorch implementation.

  复制粘贴的修复

  复制

  Calflops is the easiest and most convenient tool for calculating FLOPs, MACs, and Parameters in PyTorch models, especially for large language models (LLMs) and models from the Hugging Face platform. It supports various neural networks including Linear, CNN, RNN, GCN, and Transformers (like Bert, LLaMA), as well as any custom models based on PyTorch.

  待办完成忽略
• mediumabout#2

  Update the repository description to emphasize LLM and Hugging Face support

  原因：

  当前

  The calflops is designed to calculate FLOPs、MACs and Parameters in all various neural networks, such as Linear、 CNN、 RNN、 GCN、Transformer(Bert、LlaMA etc Large Language Model)

  复制粘贴的修复

  复制

  Calflops is an easy-to-use tool for calculating FLOPs, MACs, and Parameters in PyTorch neural networks, with a special focus on large language models (LLMs) and Hugging Face models. It supports various architectures like Linear, CNN, RNN, GCN, and Transformers (e.g., Bert, LLaMA).

  待办完成忽略
• mediumtopics#3

  Add `huggingface` and `llm` to repository topics

  原因：

  当前

  calflops, flops-counter, large-language-models, pytorch

  复制粘贴的修复

  复制

  calflops, flops-counter, large-language-models, pytorch, huggingface, llm

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  How to accurately measure computational cost and parameters for PyTorch neural network models?

  你：未被推荐
  AI 推荐顺序：

  1. PyTorch-OpCounter (torchstat) (sovrasov/flops-counter.pytorch)
  2. thop (Lyken116/pytorch-OpCounter)
  3. fvcore (from Detectron2) (facebookresearch/fvcore)
  4. torchinfo (TylerYep/torchinfo)
  5. ptflops (sukhov-alex/ptflops)
  6. DeepSpeed (flops_profiler) (microsoft/DeepSpeed)

  AI 推荐了 6 个替代方案，却始终没点名 MrYxJ/calculate-flops.pytorch。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Need a tool to analyze FLOPs and parameter usage in large language models built with PyTorch.

  你：未被推荐
  AI 推荐顺序：

  1. PyTorch profiler
  2. thop
  3. fvcore
  4. torchinfo
  5. deepspeed.profiler
  6. NVIDIA Nsight Systems

  AI 推荐了 6 个替代方案，却始终没点名 MrYxJ/calculate-flops.pytorch。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  pass

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of MrYxJ/calculate-flops.pytorch?

  pass
  ⚠AI 未点名 MrYxJ/calculate-flops.pytorch —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts MrYxJ/calculate-flops.pytorch in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 MrYxJ/calculate-flops.pytorch

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo MrYxJ/calculate-flops.pytorch solve, and who is the primary audience?

  pass
  ⚠AI 未点名 MrYxJ/calculate-flops.pytorch —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？