...
 
Commits (22)
    https://gitcode.net/greenplum/pytorch-widedeep/-/commit/60326795a8438f2fa4bb8693dd9b1d15ac0eaecf added oversample_mul to the dataloader_params to fix issue #153 2023-04-25T14:39:50+02:00 Javier jrzaurin@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/eb67b60b5c54ff9bb39a96977dcfe77da83ac2db Merge pull request #155 from jrzaurin/add-oversample-mul 2023-04-25T14:48:07+02:00 Javier jrzaurin@gmail.com added oversample_mul to the dataloader_params to fix issue #153 https://gitcode.net/greenplum/pytorch-widedeep/-/commit/29c2b9455dbb5472b52222c55a0a404bf97286dc Added Citations file 2023-05-31T17:01:16+02:00 Javier jrzaurin@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/05e007e9913f22e34c4606b14e6f6bb30ca2fbc1 removed citations file 2023-05-31T17:06:46+02:00 Javier jrzaurin@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/ccca550f538f8db51ec1080bbc136ea616dc130a add default vision models 2023-06-12T14:27:37+01:00 Krish krishandavda92@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/71dbca8c5149bfff29a7dcd29a008d5bdff66210 newline 2023-06-12T14:29:01+01:00 Krish krishandavda92@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/efff12b7563a3204e69aa9e075551174ff6fea52 fix style 2023-06-12T18:39:15+01:00 Krish krishandavda92@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/883ae44e4f038393d0e01b15f22d9087e4ab5e50 update test 2023-06-12T18:41:06+01:00 Krish krishandavda92@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/bda30f03de07a244b14a840eb7221fc91e440d26 fix black style 2023-06-12T18:44:10+01:00 Krish krishandavda92@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/9acfceca015a7553f181fb86f99a0d8ed7b5e798 Merge pull request #171 from jrzaurin/default-vision-models 2023-06-15T10:48:06+02:00 Javier jrzaurin@gmail.com Default vision models https://gitcode.net/greenplum/pytorch-widedeep/-/commit/8ce56d63f6a8a0405dac410c51b75e701cd20f20 Fix: EarlyStopping does not store and restore the model #175 2023-07-05T23:41:09+08:00 LuoXueling sharrin@sjtu.edu.cn https://gitcode.net/greenplum/pytorch-widedeep/-/commit/46e68b6b87ac5ff895fee19a990643af98a58bf3 The unit test for the restored state of EarlyStopping. 2023-07-06T01:47:06+08:00 LuoXueling sharrin@sjtu.edu.cn https://gitcode.net/greenplum/pytorch-widedeep/-/commit/29972a970cbeb1018e0c3e1a6a69f2e3630d3913 Update build.yml 2023-07-07T09:59:20+01:00 Krishan Davda krishandavda92@gmail.com Get fork PR tests running https://gitcode.net/greenplum/pytorch-widedeep/-/commit/30e8033e017f325a6ff168a803b04c3306516f88 Update build.yml 2023-07-07T10:01:22+01:00 Krishan Davda krishandavda92@gmail.com manual trigger on fork PR https://gitcode.net/greenplum/pytorch-widedeep/-/commit/b28426ca59a7b3d18b0f0d2ee643cc91cfdf134d Merge pull request #176 from LuoXueling/master 2023-07-07T10:20:18+01:00 Javier jrzaurin@gmail.com Fix: EarlyStopping does not store and restore the model #175 https://gitcode.net/greenplum/pytorch-widedeep/-/commit/26e1985921e63c44a9cf47338a5fccf7f1d81ec1 Update build.yml 2023-07-07T10:26:37+01:00 Krishan Davda krishandavda92@gmail.com revert https://gitcode.net/greenplum/pytorch-widedeep/-/commit/890917aa310c6960e5bdcc870e77935ee60ebcf9 Fix #175 early stopping and model checkpoint restoring weights. Also fix #174... 2023-07-07T17:21:46+01:00 Javier jrzaurin@gmail.com Fix #175 early stopping and model checkpoint restoring weights. Also fix #174 the finetunner single batch run https://gitcode.net/greenplum/pytorch-widedeep/-/commit/8406813ce4742cfd17c9225fc73f467a26700982 fixed unit test to run on GPU and fix a little bug for custom data loaders 2023-07-14T16:46:47+01:00 Javier jrzaurin@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/42cfe5cc3daf2992f07893d11d4a497fdca198d6 Merge pull request #177 from jrzaurin/fix_restore_best_weights 2023-07-14T18:48:29+01:00 Javier jrzaurin@gmail.com Fix #175 early stopping and model checkpoint restoring weights. https://gitcode.net/greenplum/pytorch-widedeep/-/commit/4002fdb7d980dce5a8969e4639d5c3d550bab874 Added a notebook to illustrate the use of feature importance. And also fixed ... 2023-07-20T19:37:16+02:00 Javier jrzaurin@gmail.com Added a notebook to illustrate the use of feature importance. And also fixed a little bug related to the calling of the callbacks. https://gitcode.net/greenplum/pytorch-widedeep/-/commit/ce6c70a2ea5f7add30f23e5317afa07fc478bb7f Merge branch 'master' into feature-importance 2023-07-20T20:13:33+02:00 Javier jrzaurin@gmail.com https://gitcode.net/greenplum/pytorch-widedeep/-/commit/fb307463a04b8b1ed66e98fa3c2095e58e3adebc bumped version to 1.3 2023-07-20T20:28:37+02:00 Javier jrzaurin@gmail.com
......@@ -83,4 +83,4 @@ jobs:
- name: upload coverage to Codecov
uses: codecov/codecov-action@v1
with:
fail_ci_if_error: true
\ No newline at end of file
fail_ci_if_error: true
......@@ -655,6 +655,7 @@ class EarlyStopping(Callback):
if self.monitor_op(current - self.min_delta, self.best):
self.best = current
self.wait = 0
self.best_epoch = epoch
if self.restore_best_weights:
self.state_dict = copy.deepcopy(self.model.state_dict())
else:
......@@ -665,7 +666,9 @@ class EarlyStopping(Callback):
def on_train_end(self, logs: Optional[Dict] = None):
if self.stopped_epoch > 0 and self.verbose > 0:
print("Epoch %05d: early stopping" % (self.stopped_epoch + 1))
print(
f"Best Epoch: {self.best_epoch + 1}. Best {self.monitor}: {self.best:.5f}"
)
if self.restore_best_weights and self.state_dict is not None:
if self.verbose > 0:
print("Restoring model weights from the end of the best epoch")
......
......@@ -524,7 +524,7 @@ class Trainer(BaseTrainer):
self.callback_container.on_epoch_end(epoch, epoch_logs, on_epoch_end_metric)
if self.early_stop:
self.callback_container.on_train_end(epoch_logs)
# self.callback_container.on_train_end(epoch_logs)
break
if self.model.with_fds:
......
__version__ = "1.2.3"
__version__ = "1.3.0"
......@@ -21,6 +21,7 @@ dev_status = {
"1.0": "Development Status :: 5 - Production/Stable", # v1.0 - most functionality + doc + test # noqa
"1.1": "Development Status :: 5 - Production/Stable", # v1.1 - new functionality
"1.2": "Development Status :: 5 - Production/Stable", # v1.2 - new functionality
"1.3": "Development Status :: 5 - Production/Stable", # v1.3 - new functionality
"2.0": "Development Status :: 6 - Mature", # v2.0 - new functionality?
}
......