Skip to content

Commit 97e22b1

Browse files
authored
Fixing horovod setup in makefile (#78)
* Fixing horovod setup in makefile * requirements file fix and readme fix
1 parent 2591db7 commit 97e22b1

File tree

3 files changed

+6
-23
lines changed

3 files changed

+6
-23
lines changed

nlp/bert/pytorch/README.md

Lines changed: 0 additions & 17 deletions
Original file line numberDiff line numberDiff line change
@@ -340,23 +340,6 @@ In order to re-generate the sample dataset, run the following script:
340340
python3 third_party/create_pretraining_data.py --input-file data/sample_text.txt --output-file data/sample_text.tfrecord --sequence-length 128 --mask-tokens 20 --duplication-factor 4 --do-lower-case --model bert-base-uncased
341341
```
342342

343-
344-
### Running and benchmarking
345-
346-
To run a tested and optimised configuration and to reproduce the performance shown on our [performance results page](https://www.graphcore.ai/performance-results), please follow the setup instructions in this README to setup the environment, and then use the `examples_utils` module (installed automatically as part of the environment setup) to run one or more benchmarks. For example:
347-
348-
```python
349-
python3 -m examples_utils benchmark --spec <path to benchmarks.yml file>
350-
```
351-
352-
Or to run a specific benchmark in the `benchmarks.yml` file provided:
353-
354-
```python
355-
python3 -m examples_utils benchmark --spec <path to benchmarks.yml file> --benchmark <name of benchmark>
356-
```
357-
358-
For more information on using the examples-utils benchmarking module, please refer to [the README](https://github.com/graphcore/examples-utils/blob/master/examples_utils/benchmarks/README.md).
359-
360343
### Troubleshooting
361344

362345
If Triton server tests fails with such error:

vision/cnns/pytorch/makefile

Lines changed: 6 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -3,9 +3,10 @@ all: install get-data install-turbojpeg
33
install-turbojpeg: get-turbojpeg build-turbojpeg
44

55
install:
6-
pip install -r requirements.txt
7-
pip uninstall pillow -y
8-
CC="cc -mavx2" pip install --no-cache-dir -U --force-reinstall pillow-simd
6+
pip3 install -r requirements.txt
7+
pip3 uninstall pillow -y
8+
CC="cc -mavx2" pip3 install --no-cache-dir -U --force-reinstall pillow-simd
9+
HOROVOD_WITH_PYTORCH=1 pip3 install --no-cache-dir 'horovod==0.24.3'
910

1011
get-turbojpeg:
1112
cd datasets && rm -rf libjpeg-turbo && git clone https://github.com/libjpeg-turbo/libjpeg-turbo.git
@@ -20,5 +21,5 @@ get-data:
2021
cd datasets && ./get_images.sh
2122

2223
test:
23-
cd tests && pytest -n 10 --forked
24-
cd tests_serial && pytest
24+
cd tests && python3 -m pytest -n 10 --forked
25+
cd tests_serial && python3 -m pytest

vision/cnns/pytorch/requirements.txt

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,6 @@ timm==0.5.4
1313
wandb==0.12.8
1414
pyyaml==5.4.1
1515
checksumdir==1.2.0
16-
horovod==0.24.3
1716
tritonclient[grpc]==2.16.0
1817
git+https://github.com/lilohuang/PyTurboJPEG.git@8706665787c7290397859075ae2f0bf35afeb41a
1918
git+https://github.com/graphcore/examples-utils@0f3024827451878955ed495daf2e1b552eaf749f#egg=examples-utils

0 commit comments

Comments
 (0)