babysor / babysor/MockingBird

loss 值没有下降

Open
#773 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
36.9k
Forks
5.2k
PR merge metrics
No merged PRs in 30d

Description

数据6万条,49小时,Step: 229k的时候Loss: 0.2852 ,一直没有下降,这个根数据集的大小有多大关系?

{| Epoch: 36/84 (1926/1926) | Loss: 0.2852 | 0.87 steps/s | Step: 229k | }

4444

Contributor guide

No contributing guide indexed for this repository

Research direction

No source file, test, or entry point is named. Start by reviewing the training configuration and logs for the reported 60,000-example run, then compare the loss behavior with a controlled dataset-size run; done means determining whether dataset size explains the plateau and documenting the evidence.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
ai, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.