Skip to content

Commit 0b285bd

Browse files
authored
Merge pull request #235 from jrzaurin/mps_backened_support
Mps backened support
2 parents 9df6585 + 278c72f commit 0b285bd

File tree

188 files changed

+51888
-20991
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

188 files changed

+51888
-20991
lines changed

.coveragerc

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,11 @@ omit =
55
pytorch_widedeep/bayesian_models/bayesian_nn/modules/*
66

77
[report]
8+
exclude_lines =
9+
pragma: no cover
10+
def __repr__
11+
class GroupWideDeepDataset
12+
def group_collate_fn
813
exclude_also =
914
def __repr__
1015
omit =

.gitignore

Lines changed: 10 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -60,4 +60,13 @@ checkpoints
6060

6161
# wnb
6262
wandb/
63-
wandb_api.key
63+
wandb_api.key
64+
65+
# pytest
66+
pytest.ini
67+
68+
# code style
69+
code_style.sh
70+
71+
# isort
72+
.isort.cfg

.isort.cfg

Lines changed: 0 additions & 4 deletions
This file was deleted.

.readthedocs.yaml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ version: 2
99
build:
1010
os: ubuntu-20.04
1111
tools:
12-
python: "3.9"
12+
python: "3.10"
1313

1414
mkdocs:
1515
configuration: mkdocs/mkdocs.yml

.travis.yml

Lines changed: 0 additions & 34 deletions
This file was deleted.

README.md

Lines changed: 11 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11

22
<p align="center">
3-
<img width="300" src="docs/figures/widedeep_logo.png">
3+
<img width="300" src="mkdocs/sources/docs/figures/widedeep_logo.png">
44
</p>
55

66
[![PyPI version](https://badge.fury.io/py/pytorch-widedeep.svg)](https://pypi.org/project/pytorch-widedeep/)
@@ -60,7 +60,7 @@ Figure below:
6060

6161

6262
<p align="center">
63-
<img width="750" src="docs/figures/widedeep_arch_new.png">
63+
<img width="750" src="mkdocs/sources/docs/figures/widedeep_arch_new.png">
6464
</p>
6565

6666

@@ -69,7 +69,7 @@ In math terms, and following the notation in the
6969
without a ``deephead`` component can be formulated as:
7070

7171
<p align="center">
72-
<img width="500" src="docs/figures/architecture_1_math.png">
72+
<img width="500" src="mkdocs/sources/docs/figures/architecture_1_math.png">
7373
</p>
7474

7575

@@ -164,7 +164,7 @@ Face models.
164164

165165

166166
<p align="center">
167-
<img width="400" src="docs/figures/arch_1.png">
167+
<img width="400" src="mkdocs/sources/docs/figures/arch_1.png">
168168
</p>
169169

170170

@@ -210,7 +210,7 @@ trainer.fit(
210210
**2. Tabular and Text data**
211211

212212
<p align="center">
213-
<img width="400" src="docs/figures/arch_2.png">
213+
<img width="400" src="mkdocs/sources/docs/figures/arch_2.png">
214214
</p>
215215

216216

@@ -262,7 +262,7 @@ trainer.fit(
262262
in `WideDeep`**
263263

264264
<p align="center">
265-
<img width="400" src="docs/figures/arch_3.png">
265+
<img width="400" src="mkdocs/sources/docs/figures/arch_3.png">
266266
</p>
267267

268268
```python
@@ -313,7 +313,7 @@ trainer.fit(
313313
`WideDeep`**
314314

315315
<p align="center">
316-
<img width="500" src="docs/figures/arch_4.png">
316+
<img width="500" src="mkdocs/sources/docs/figures/arch_4.png">
317317
</p>
318318

319319
```python
@@ -375,7 +375,7 @@ trainer.fit(
375375
`ModelFuser` class**
376376

377377
<p align="center">
378-
<img width="500" src="docs/figures/arch_5.png">
378+
<img width="500" src="mkdocs/sources/docs/figures/arch_5.png">
379379
</p>
380380

381381
```python
@@ -454,7 +454,7 @@ property and avoids some typing errors internally.
454454

455455

456456
<p align="center">
457-
<img width="600" src="docs/figures/arch_6.png">
457+
<img width="600" src="mkdocs/sources/docs/figures/arch_6.png">
458458
</p>
459459

460460

@@ -596,7 +596,7 @@ target). We can create a two-tower model where the user and item features are
596596
passed through two separate models and then "fused" via a dot product.
597597

598598
<p align="center">
599-
<img width="350" src="docs/figures/arch_7.png">
599+
<img width="350" src="mkdocs/sources/docs/figures/arch_7.png">
600600
</p>
601601

602602

@@ -692,7 +692,7 @@ This one is "a bonus" to illustrate the use of multi-target losses, more than
692692
actually a different architecture.
693693

694694
<p align="center">
695-
<img width="200" src="docs/figures/arch_8.png">
695+
<img width="200" src="mkdocs/sources/docs/figures/arch_8.png">
696696
</p>
697697

698698

VERSION

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
1.6.4
1+
1.6.5

code_style.sh

Lines changed: 0 additions & 8 deletions
This file was deleted.

docs/Makefile

Lines changed: 0 additions & 21 deletions
This file was deleted.

docs/_static/custom.css

Lines changed: 0 additions & 45 deletions
This file was deleted.
-99.7 KB
Binary file not shown.

docs/bayesian_models.rst

Lines changed: 0 additions & 15 deletions
This file was deleted.

docs/callbacks.rst

Lines changed: 0 additions & 29 deletions
This file was deleted.

0 commit comments

Comments
 (0)