Mentions légales du service

Skip to content
Snippets Groups Projects
Commit f9daded9 authored by BIGAUD Nathan's avatar BIGAUD Nathan
Browse files

Correcting tf file typos

parent 6227db0a
Branches
No related tags found
1 merge request!32Add support for Jax / Haiku
...@@ -17,8 +17,8 @@ ...@@ -17,8 +17,8 @@
"""Unit tests for TensorflowModel.""" """Unit tests for TensorflowModel."""
import warnings
import sys import sys
import warnings
from typing import Any, List, Literal from typing import Any, List, Literal
import numpy as np import numpy as np
...@@ -38,7 +38,7 @@ from declearn.utils import set_device_policy ...@@ -38,7 +38,7 @@ from declearn.utils import set_device_policy
# dirty trick to import from `model_testing.py`; # dirty trick to import from `model_testing.py`;
# pylint: disable=wrong-import-order, wrong-import-position # pylint: disable=wrong-import-order, wrong-import-position
sys.path.append(".") sys.path.append(".")
from model_testing import ModelTestSuite, ModelTestCase from model_testing import ModelTestCase, ModelTestSuite
class TensorflowTestCase(ModelTestCase): class TensorflowTestCase(ModelTestCase):
...@@ -51,7 +51,7 @@ class TensorflowTestCase(ModelTestCase): ...@@ -51,7 +51,7 @@ class TensorflowTestCase(ModelTestCase):
16-neurons fully-connected layer with ReLU 16-neurons fully-connected layer with ReLU
1 output neuron with sigmoid activation 1 output neuron with sigmoid activation
* "MLP-tune": * "MLP-tune":
- same as NLP, but freeze the first layer of the stack - same as MLP, but freeze the first layer of the stack
* "RNN": * "RNN":
- input: 128-tokens-sequence in a 100-tokens-vocabulary - input: 128-tokens-sequence in a 100-tokens-vocabulary
- stack: 32-dimensional embedding matrix - stack: 32-dimensional embedding matrix
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment