Skip to content

Added Gloo as a backend option #203

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Jan 26, 2022
Merged

Added Gloo as a backend option #203

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7d0b2ad
Added Gloo as a backend option
sayantan1410 Jan 25, 2022
8cb8985
ran prettier and changed backend from nccl
sayantan1410 Jan 25, 2022
9f0ab87
corrected code for adding Gloo as a backend
sayantan1410 Jan 26, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 12 additions & 2 deletions src/components/TabTraining.vue
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,9 +6,16 @@
:label="deterministic.description"
:saveKey="deterministic.name"
/>
<h2 class="training">Distributed Training (NCCL backend)</h2>
<h2 class="training">Distributed Training</h2>
<FormCheckbox label="Use distributed training" saveKey="use_dist" />
<div v-show="store.config.use_dist">
<h2>Choose a Backend</h2>
<FormSelect
required
:saveKey="backend.name"
:label="backend.description"
:options="backend.options"
/>
<FormRadio :options="[launch, spawn]" saveKey="dist" defaultV="launch" />
<FormInput
:label="nproc_per_node.description"
Expand Down Expand Up @@ -46,13 +53,15 @@ import { training } from '../metadata/metadata.json'
import FormCheckbox from './FormCheckbox.vue'
import FormInput from './FormInput.vue'
import FormRadio from './FormRadio.vue'
import FormSelect from './FormSelect.vue'
import { store } from '../store.js'

export default {
components: { FormCheckbox, FormInput, FormRadio },
components: { FormCheckbox, FormInput, FormRadio, FormSelect },
setup() {
const {
deterministic,
backend,
launch,
spawn,
nproc_per_node,
Expand All @@ -69,6 +78,7 @@ export default {
return {
store,
deterministic,
backend,
launch,
spawn,
nproc_per_node,
Expand Down
5 changes: 5 additions & 0 deletions src/metadata/metadata.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,11 @@
"description": "Master node port for torch native backends (mandatory if you have filled number of nodes)",
"min": 0,
"default": 8080
},
"backend": {
"name": "backend",
"description": "Choose a Backend",
"options": ["NCCL", "Gloo"]
}
},
"handlers": {
Expand Down
20 changes: 10 additions & 10 deletions src/templates/template-common/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,26 +9,26 @@

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -39,7 +39,7 @@ python -m torch.distributed.launch \
python -m torch.distributed.launch \
--nproc_per_node #:::= it.nproc_per_node :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand All @@ -56,24 +56,24 @@ python -m torch.distributed.launch \

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -83,7 +83,7 @@ python main.py \
```sh
python main.py \
--nproc_per_node #:::= it.nproc_per_node :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand Down
20 changes: 10 additions & 10 deletions src/templates/template-text-classification/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,26 +25,26 @@ pip install -r requirements.txt --progress-bar off -U

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -55,7 +55,7 @@ python -m torch.distributed.launch \
python -m torch.distributed.launch \
--nproc_per_node #:::= it.nproc_per_node :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand All @@ -72,24 +72,24 @@ python -m torch.distributed.launch \

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -99,7 +99,7 @@ python main.py \
```sh
python main.py \
--nproc_per_node #:::= it.nproc_per_node :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand Down
20 changes: 10 additions & 10 deletions src/templates/template-vision-classification/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,26 +25,26 @@ pip install -r requirements.txt --progress-bar off -U

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -55,7 +55,7 @@ python -m torch.distributed.launch \
python -m torch.distributed.launch \
--nproc_per_node #:::= it.nproc_per_node :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand All @@ -72,24 +72,24 @@ python -m torch.distributed.launch \

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -99,7 +99,7 @@ python main.py \
```sh
python main.py \
--nproc_per_node #:::= it.nproc_per_node :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand Down
20 changes: 10 additions & 10 deletions src/templates/template-vision-dcgan/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,26 +25,26 @@ pip install -r requirements.txt --progress-bar off -U

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python -m torch.distributed.launch \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -55,7 +55,7 @@ python -m torch.distributed.launch \
python -m torch.distributed.launch \
--nproc_per_node #:::= it.nproc_per_node :::# \
--use_env main.py \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand All @@ -72,24 +72,24 @@ python -m torch.distributed.launch \

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank 0 \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

- Execute on worker nodes

```sh
python main.py \
--nproc_per_node #:::= nproc_per_node :::# \
--nproc_per_node #:::= it.nproc_per_node :::# \
--nnodes #:::= it.nnodes :::# \
--node_rank <node_rank> \
--master_addr #:::= it.master_addr :::# \
--master_port #:::= it.master_port :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } else { :::#
Expand All @@ -99,7 +99,7 @@ python main.py \
```sh
python main.py \
--nproc_per_node #:::= it.nproc_per_node :::# \
--backend nccl
--backend #:::= it.backend :::#
```

#::: } :::#
Expand Down
Loading