OSError: results_gen_sym.dat not found.

[mbennun]

I am trying the example using:
aifeynman.run_aifeynman("./example_data/", "example1.txt", 60, "14ops.txt", polyfit_deg=3, NN_epochs=500)

It runs and then fails with this error:

Checking for generalized symmetry...
identify_decompositions ./example_data/ example1.txt_train
Trying (0, 1)
(1.2643647381029814, (0, 1))
Trying (0, 2)
(3.397227321254669, (0, 2))
Trying (1, 2)
(3.3632891097948683, (1, 2))
mask [ True False True]
Trying to solve mysteries with brute force...
Trying to solve results/gradients_gen_sym_example1.txt_train
Traceback (most recent call last):
File "", line 1, in
File "/usr/local/lib/python3.9/site-packages/aifeynman/S_run_aifeynman.py", line 274, in run_aifeynman
PA = run_AI_all(pathdir,filename+"_train",BF_try_time,BF_ops_file_type, polyfit_deg, NN_epochs, PA=PA)
File "/usr/local/lib/python3.9/site-packages/aifeynman/S_run_aifeynman.py", line 156, in run_AI_all
bf_all_output = np.loadtxt("results_gen_sym.dat", dtype="str")
File "/usr/local/lib/python3.9/site-packages/numpy/lib/npyio.py", line 1067, in loadtxt
fh = np.lib._datasource.open(fname, 'rt', encoding=encoding)
File "/usr/local/lib/python3.9/site-packages/numpy/lib/_datasource.py", line 193, in open
return ds.open(path, mode, encoding=encoding, newline=newline)
File "/usr/local/lib/python3.9/site-packages/numpy/lib/_datasource.py", line 533, in open
raise IOError("%s not found." % path)
OSError: results_gen_sym.dat not found.

[kvyaswanth]

results_gen_sym.dat was to be generated by this file "feynman_sr_mdl5"

I suggest checking if the file is executable, or try using print statement here to check if the above mentioned code is running

[toontran]

Apparently if you want to use AI Feynman 2.0, it's better to use a historical version of this repo, according to this article

[Rafail-Karkanis]

I am getting the same error.
@kvyaswanth This returns a [Errno 2] No such file or directory: 'feynman_sr_mdl5'
@toontran I tried to use dcshapiro's fork, but it returns another error upon executing the example.py
Is there any commit hash that actually works?