JeremyXU

you can use the following command : pip install --user biaoxu4(your username) and then run pip install -- user . it works as follows 

The tutorial is from https://nb.bohrium.dp.tech/detail/18475433825, in the part of "DP-Gen based on a DPA-2 pretrained model"

The submit comand is dpgen2 submit input.json. and The all input files in the directory is as follows, if you need any files, please let me know. drwxr-xr-x 4 root...

**dp --pt compress -i model.pth -o compress_model.pth** **also not work** root@bohrium-12166-1204587:/personal/dpa2_hea/version10/distill_stable/iter-000002/prep-run-train/output/models/task.0000# ls checkpoint model.ckpt.pt model.pth root@bohrium-12166-1204587:/personal/dpa2_hea/version10/distill_stable/iter-000002/prep-run-train/output/models/task.0000# **dp --pt compress -i model.pth -o compress_model.pth** To get the best performance, it is...

**The setting of descriptor during distillation with the "attn_layer": 0** "descriptor": { "type": "se_atten_v2", "sel": 120, "rcut_smth": 0.50, "rcut": 6.00, "neuron": [ 25, 50, 100 ], "resnet_dt": false, "axis_neuron": 16,...

Whether the b4 supports compression with pytorch backend?