Add TF weights
#3
by
neggles
- 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=3.719e-05; Maximum crossload hidden layer difference=1.831e-04;
Maximum conversion output difference=3.719e-05; Maximum conversion hidden layer difference=1.831e-04;
CAUTION: The maximum admissible error was manually increased to 0.1!
Note: Actual output differences are:
List of maximum output differences above the threshold (1e-05):
logits: 1.001e-05
List of maximum hidden layer differences above the threshold (1e-05):
hidden_states[2]: 1.335e-05
hidden_states[3]: 1.144e-04
hidden_states[4]: 3.004e-05
Minor error in conversion code when I created this. See GH PR for details.
lysandre
changed pull request status to
merged