mirror of https://github.com/hpcaitech/ColossalAI
3af13a2c3e
* place params on cpu after zero init context * polish code * bucketzed cpu gpu tensor transter * find a bug in sharded optim unittest * add offload unittest for ShardedOptimV2. * polish code and make it more robust |
||
---|---|---|
.. | ||
common.py | ||
test_init_context.py | ||
test_shard_model_v2.py | ||
test_shard_param.py | ||
test_sharded_optim.py | ||
test_sharded_optim_v2.py | ||
test_sharded_optim_v2_with_cpu_adam.py | ||
test_sharded_optim_with_sync_bn.py | ||
test_state_dict.py | ||
test_zero_param_mgr.py |