Add TF weights
#4
by Rocketknight1 HF Staff - opened
Model converted by the transformers' pt_to_tf CLI. All converted model outputs and hidden layers were validated against its PyTorch counterpart.
Maximum crossload output difference=3.433e-05; Maximum crossload hidden layer difference=5.722e-06;
Maximum conversion output difference=3.433e-05; Maximum conversion hidden layer difference=5.722e-06;
Hi @emilyalsentzer - this is an automated TF port of the Bio_ClinicalBERT weights. Outputs have been tested and are equivalent up to normal float32 tolerances!
emilyalsentzer changed pull request status to merged