Trouble using dp converted model

[rajnichahal]

Hi, I trained a deepmd model using deepmd version 1.2.1. In order to use the python API for deepmd version 2.1.1, I converted the model using following:
dp convert-from 1.2 -i old_frozen_model.pb -o new_frozen_model.pb

The model conversion was successful, but it is giving me the following error message when using Python API as below:
dp = DeepPot('new_frozen_model.pb')

Error message
/usr/local/envs/deepmd/lib/python3.10/site-packages/scipy/init.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.1
warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"
WARNING:tensorflow:From /usr/local/envs/deepmd/lib/python3.10/site-packages/tensorflow/python/compat/v2_compat.py:107: 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
WARNING:root:To get the best performance, it is recommended to adjust the number of threads by setting the environment variables OMP_NUM_THREADS, TF_INTRA_OP_PARALLELISM_THREADS, and TF_INTER_OP_PARALLELISM_THREADS.
WARNING:root:Environment variable KMP_BLOCKTIME is empty. Use the default value 0
WARNING:root:Environment variable KMP_AFFINITY is empty. Use the default value granularity=fine,verbose,compact,1,0
/usr/local/envs/deepmd/lib/python3.10/importlib/init.py:169: UserWarning: The NumPy module was reloaded (imported a second time). This can in some cases result in small but subtle issues and is discouraged.
_bootstrap._exec(spec, module)
Traceback (most recent call last):
File "/project/uml_stephen_lam/Rajni/pca_tsne/deepmd_data/get_data3_11_atomic.py", line 9, in
dp = DeepPot('new_frozen_model.pb')
File "/usr/local/envs/deepmd/lib/python3.10/site-packages/deepmd/infer/deep_pot.py", line 88, in init
DeepEval.init(
File "/usr/local/envs/deepmd/lib/python3.10/site-packages/deepmd/infer/deep_eval.py", line 41, in init
self.graph = self._load_graph(
File "/usr/local/envs/deepmd/lib/python3.10/site-packages/deepmd/infer/deep_eval.py", line 154, in _load_graph
graph_def.ParseFromString(f.read())
google.protobuf.message.DecodeError: Error parsing message with type 'tensorflow.GraphDef'

Please let me know how can I resolve this error. Thanks!

[njzjz]

Could you provide the model?

[rajnichahal]

1. Oh ok. Even if I use the Python API to convert, it is still writing .pbtxt file upon conversion:
   import os
   import textwrap
   from deepmd.env import tf
   from deepmd.utils import convert

convert.convert_13_to_21('Flinazr_train7_2.pb', 'output_model.pb')

2. Also, trying to convert .pbtxt to .pb as following--it is not qriting the output .pb file:
   import os
   import textwrap
   from deepmd.env import tf
   from deepmd.utils import convert

convert.convert_pbtxt_to_pb('frozen_model.pbtxt','new model.pb')

Please let me know if there is anything I am missing with convert_pbtxt_to_pb. Thanks

[njzjz]

If you never modified the following code (as_text=False), there may be a bug in the TensorFlow...

deepmd-kit/deepmd/utils/convert.py

Line 156 in 6e3d4a6

tf.train.write_graph(graph_def, './', pbfile, as_text=False)

[rajnichahal]

Hi, I did not modify the code.

Please let me know what should be my option here.

[njzjz]

If you get the same result by manually executing the following tensorflow codes, you can consider submitting an issue to tensorflow. There is nothing we can do here.

deepmd-kit/deepmd/utils/convert.py

Lines 151 to 156 in 6e3d4a6

	with tf.gfile.GFile(pbtxtfile, 'r') as f:
	graph_def = tf.GraphDef()
	file_content = f.read()
	# Merges the human-readable string in `file_content` into `graph_def`.
	text_format.Merge(file_content, graph_def)
	tf.train.write_graph(graph_def, './', pbfile, as_text=False)

[rajnichahal]

Hi,
I figured that my script was being killed and nothing was written because of less memory available. Converting as following is working now:
import os
import textwrap
from deepmd.env import tf
from deepmd.utils import convert
convert.convert_13_to_21('Flinazr_train7_2.pb', 'output_model.pb')

thanks for your help!