hummingbird
/
fastNLP
Not watched
2
0
Fork 0
Code
Releases 13 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
2193 Commits
3 Branches
25 MB
728 Download
Tree: efaec85c57
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'efaec85c57'
${ noResults }
Commit Graph

26 Commits (efaec85c576e9af9691cc8bd4ebe72fd60b3bd6d)

Author SHA1 Message Date
  yh_cc 97b504c020 修改Trainer.save/Driver.save为Trainer.save_checkpoint/Driver.save_checkpoint 3 years ago
  x54-729 36ed894a77 Merge branch 'dev0.8.0' of github.com:fastnlp/fastNLP into dev0.8.0 4 years ago
  x54-729 f79ee04956 为torch测试例添加_NEED_IMPORT_TORCH 4 years ago
  YWMditto 1528107480 添加了 reproducible batch sampler 的测试 4 years ago
  yh_cc 5172b992cf Merge branch 'dev0.8.0' of github.com:fastnlp/fastNLP into dev0.8.0 4 years ago
  yh_cc 4ba0ff2902 fix test 4 years ago
  x54-729 5a1bc5235f Merge branch 'dev0.8.0' of github.com:fastnlp/fastNLP into dev0.8.0 4 years ago
  yh_cc 4ed012bf1a 更新cache_results的测试 4 years ago
  x54-729 2169974903 完善测试例的import 4 years ago
  x54-729 f2fa05d136 测试例添加 _NEED_IMPORT_PADDLE 4 years ago
  yh_cc b38fc1136e 1.修改ProgressBart在Trainer中的一个bug;2修复pytest的bug 4 years ago
  yh_cc f74b9b6bec 修改所有的 validate 为 evaluate ; 移动 callback.on_train_end() 4 years ago
  x54-729 705deeaea9 small 4 years ago
  YWMditto 3823fc557d 对 magic_argv_env_context 添加了 timeout 参数，测试函数超过一定时间后自动kill掉 4 years ago
  x54-729 cca265f99c Merge branch 'dev0.8.0' of github.com:fastnlp/fastNLP into dev0.8.0 4 years ago
  x54-729 d0f26c7c34 将validate_step替换为evaluate_step 4 years ago
  yh_cc 7c70874b4a 删除core.sampelrs.sampler.py；增加torch的clipgradient和warmupcallback 4 years ago
  yh_cc a4b2e0fac5 修复若干bug 4 years ago
  YWMditto 2924e2117f 删除了 driver 中的 **_step，使用 model_call 和 get_model_call_fn 来代替；删除了 driver 中的所有 dataloaders 4 years ago
  yh_cc 16a467393c 1.montior允许传入callable的对象进行选择; 2.解决Sampler中存在的循环引用问题 4 years ago
  YWMditto 3ee6fc66f5 添加了 on_after_optimizers_step 和 on_after_zero_grad 的callback接口 4 years ago
  yh_cc e8d11cd5a9 1. 修复torch 分布式在不同版本中group参数default值不一样的问题; 2. torch修复多卡时只有batchsampler evaluate会遇到bug的问题; 3。logger增加warning_once接口;4.增加callback相关文档 4 years ago
  YWMditto 5419b6a042 填了了关闭参数匹配的逻辑；添加了 trainer 中获取 driver 参数的接口 4 years ago
  x54-729 680421fe86 paddle分布式训练 4 years ago
  x54-729 6ef912675b 新的paddle测试数据集 4 years ago
  YWMditto fe270ef619 更新 callbacks, controllers, drivers, samplers 以及 tests 中的部分内容 4 years ago