The dp compress with pytorch backend can not do the compression (finetune-->distillation based on deepmd-kit-v3-b3, dpgen2) #274
Description
we use the Pytorch backend for the distillation process, and download the model at /prep-run-train/output/models/task.0000 by "dpgen2 download ..." command, then get a model: model.ckpt.pt, and frozen it by the " dp --pt freeze -o model.pth" command (need a manual add checkpoint file) and obtain the model.pth. However, this model.pth cannot compress by the Pytorch compression command "dp compress -i model.pth -o model-compress.pth", and it gives the following error message:
root@bohrium-12166-1204587:/personal/dpa2_hea/version10/distill_stable/iter-000002/prep-run-train/output/models/task.0000# dp compress -i model.pth -o model_compress.pth
2024-11-12 14:48:51.537628: E tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:9342] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2024-11-12 14:48:51.537679: E tensorflow/compiler/xla/stream_executor/cuda/cuda_fft.cc:609] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2024-11-12 14:48:51.537696: E tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:1518] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2024-11-12 14:48:51.544735: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE4.1 SSE4.2 AVX AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
WARNING:tensorflow:From /opt/deepmd-kit-3.0.0b3/lib/python3.10/site-packages/tensorflow/python/compat/v2_compat.py:108: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
Instructions for updating:
non-resource variables are not supported in the long term
To get the best performance, it is recommended to adjust the number of threads by setting the environment variables OMP_NUM_THREADS, DP_INTRA_OP_PARALLELISM_THREADS, and DP_INTER_OP_PARALLELISM_THREADS. See https://deepmd.rtfd.io/parallelism/ for more information.
WARNING:tensorflow:disable_mixed_precision_graph_rewrite() called when mixed precision is already disabled.
Traceback (most recent call last):
File "/opt/deepmd-kit-3.0.0b3/bin/dp", line 10, in
sys.exit(main())
File "/opt/deepmd-kit-3.0.0b3/lib/python3.10/site-packages/deepmd/main.py", line 923, in main
deepmd_main(args)
File "/opt/deepmd-kit-3.0.0b3/lib/python3.10/site-packages/deepmd/tf/entrypoints/main.py", line 81, in main
compress(**dict_args)
File "/opt/deepmd-kit-3.0.0b3/lib/python3.10/site-packages/deepmd/tf/entrypoints/compress.py", line 98, in compress
graph, _ = load_graph_def(input)
File "/opt/deepmd-kit-3.0.0b3/lib/python3.10/site-packages/deepmd/tf/utils/graph.py", line 42, in load_graph_def
graph_def.ParseFromString(f.read())
google.protobuf.message.DecodeError: Error parsing message with type 'tensorflow.GraphDef'