You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Move framework-independent codes to the `deepmd_utils` module without
modification, as a step of
#3118.
---------
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
njzjz
added a commit
to njzjz/deepmd-kit
that referenced
this issue
Jan 13, 2024
See #3118 and
dptech-corp/deepmd-pytorch#137. Subclass needs
to implement `is_gpu_available` and `is_oom_error`.
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Summary
Move backend-independent utility codes in the
deepmd
package to thedeepmd_utils
package so other backends can use them.Detailed Description
Compatibility should be kept.
Move without modification
deepmd_utils
(without modifaction) #3140deepmd.common.add_data_requirement
deepmd.common.select_idx_map
deepmd.common.j_must_have
deepmd.common.j_loader
deepmd.common.expand_sys_str
deepmd.common.get_np_precision
deepmd.utils.argcheck
deepmd.utils.compat
deepmd.utils.data
deepmd.utils.data_system
deepmd.utils.pair_tab
deepmd.utils.path
deepmd.utils.plugin
deepmd.utils.random
deepmd.utils.weight_avg
deepmd.model.model_stat
deepmd.loggers
Move with minor modification
deepmd.utils.batch_size
add cross-platform AutoBatchSize #3143Two things are different: (1) is_gpu_available; (2) OOM exception.
deepmd.model_devi
Move model deviation and ase calculator todeepmd_utils
#3173A universal
DeepPot
class is required.deepmd.calculator
Move model deviation and ase calculator todeepmd_utils
#3173A universal
DeepPot
class is required.deepmd.entrypoints
move deepmd.entrypoints.{doc,gui} to deepmd_utils.entrypoints.{doc,gui} #3144dropdeepmd.cluster
deepmd.tf.cluster.slurm
#3239The way to get GPUs is different.
Further Information, Files, and Links
No response
The text was updated successfully, but these errors were encountered: