p83571920
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!4857 Deepfm, add 8p learing rate value to README.md 

Merge pull request !4857 from tom_chen/deepfm_lr
This commit is contained in:
mindspore-ci-bot 2020-08-20 21:40:11 +08:00 committed by Gitee
parent b4925d3e1b 447be49322
commit 153ed6f602
1 changed files with 19 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
model_zoo/official/recommend/deepfm/README.md
View File

@ -28,6 +28,7 @@ The overall network architecture of DeepFM is show below:
├── README.md
├── scripts
│ ├──run_distribute_train.sh
│ ├──run_distribute_train_gpu.sh
│ ├──run_standalone_train.sh
│ ├──run_eval.sh
├── src
@ -44,18 +45,21 @@ The overall network architecture of DeepFM is show below:
### Usage
- sh run_train.sh [DEVICE_NUM] [DATASET_PATH] [RANK_TABLE_FILE]
- python train.py --dataset_path [DATASET_PATH]
- sh run_distribute_train.sh [DEVICE_NUM] [DATASET_PATH] [RANK_TABLE_FILE]
- sh run_distribute_train_gpu.sh [DEVICE_NUM] [DATASET_PATH]
- sh run_standalone_train.sh [DEVICE_ID] [DEVICE_TARGET] [DATASET_PATH]
- python train.py --dataset_path [DATASET_PATH] --device_target [DEVICE_TARGET]
### Launch
```
# distribute training example
sh scripts/run_distribute_train.sh 8 /opt/dataset/criteo /opt/mindspore_hccl_file.json
sh scripts/run_distribute_train_gpu.sh 8 /opt/dataset/criteo
# standalone training example
sh scripts/run_standalone_train.sh 0 /opt/dataset/criteo
sh scripts/run_standalone_train.sh 0 Ascend /opt/dataset/criteo
or
python train.py --dataset_path /opt/dataset/criteo > output.log 2>&1 &
python train.py --dataset_path /opt/dataset/criteo --device_target Ascend > output.log 2>&1 &
```
### Result
@ -71,13 +75,13 @@ and eval log will be redirected to `./auc.log` by default.
### Usage
- sh run_eval.sh [DEVICE_ID] [DATASET_PATH] [CHECKPOINT_PATH]
- sh run_eval.sh [DEVICE_ID] [DEVICE_TARGET] [DATASET_PATH] [CHECKPOINT_PATH]
### Launch
```
# infer example
sh scripts/run_eval.sh 0 ~/criteo/eval/ ~/train/deepfm-15_41257.ckpt
sh scripts/run_eval.sh 0 Ascend ~/criteo/eval/ ~/train/deepfm-15_41257.ckpt
```
> checkpoint can be produced in training process.
@ -92,6 +96,15 @@ Inference result will be stored in the example path, you can find result like th
# Model description
## Learning Rate
| Number of Devices | Learning Rate |
| ---------------------- | ------------------ |
| 1 | 1e-5 |
| 8 | 1e-4 |
> Change the learning rate at src/config.py accordingly.
## Performance
### Training Performance