maxmustermann/DeepLearningExamples
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

Merge pull request #659 from hXl3s/RN50/readme-update

Browse source 
[ConvNets/TF] Document synthetic dataset options
This commit is contained in:
nv-kkudrynski 2020-08-20 16:45:53 +02:00 committed by GitHub
parent 446c878878 8bd6dd14d3
commit 0e6cfbd0a1
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 9 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
TensorFlow/Classification/ConvNets/resnet50v1.5/README.md
View file

@ -194,7 +194,7 @@ To train your model using mixed precision or TF32 with Tensor Cores or FP32, per
1. Clone the repository.
```
git clone https://github.com/NVIDIA/DeepLearningExamples
cd DeepLearningExamples/TensorFlow/Classification/RN50v1.5
cd DeepLearningExamples/TensorFlow/Classification/ConvNets
```
2. Download and preprocess the dataset.
@ -452,10 +452,9 @@ To benchmark the training performance on a specific batch size, run:
Each of these scripts runs 200 warm-up iterations and measures the first epoch.
To control warmup and benchmark length, use the `--warmup_steps`, `--num_iter` and `--iter_unit` flags. Features like XLA or DALI can be controlled
with `--use_xla` and `--use_dali` flags.
with `--use_xla` and `--use_dali` flags. If no `--data_dir=<path to imagenet>` flag is specified then the benchmarks will use a synthetic dataset.
Suggested batch sizes for training are 256 for mixed precision training and 128 for single precision training per single V100 16 GB.
#### Inference performance benchmark
To benchmark the inference performance on a specific batch size, run:
@ -470,6 +469,7 @@ To benchmark the inference performance on a specific batch size, run:
By default, each of these scripts runs 20 warm-up iterations and measures the next 80 iterations.
To control warm-up and benchmark length, use the `--warmup_steps`, `--num_iter` and `--iter_unit` flags.
If no `--data_dir=<path to imagenet>` flag is specified then the benchmarks will use a synthetic dataset.
The benchmark can be automated with the `inference_benchmark.sh` script provided in `resnet50v1.5`, by simply running:
`bash ./resnet50v1.5/inference_benchmark.sh <data dir> <data idx dir>`

5
TensorFlow/Classification/ConvNets/resnext101-32x4d/README.md
View file

@ -203,7 +203,7 @@ To train your model using mixed precision or TF32 with Tensor Cores or FP32, per
1. Clone the repository.
```
git clone https://github.com/NVIDIA/DeepLearningExamples
cd DeepLearningExamples/TensorFlow/Classification/RN50v1.5
cd DeepLearningExamples/TensorFlow/Classification/ConvNets
```
2. Download and preprocess the dataset.
@ -420,7 +420,7 @@ To benchmark the training performance on a specific batch size, run:
Each of these scripts runs 200 warm-up iterations and measures the first epoch.
To control warmup and benchmark length, use the `--warmup_steps`, `--num_iter` and `--iter_unit` flags. Features like XLA or DALI can be controlled
with `--use_xla` and `--use_dali` flags.
with `--use_xla` and `--use_dali` flags. If no `--data_dir=<path to imagenet>` flag is specified then the benchmarks will use a synthetic dataset.
Suggested batch sizes for training are 128 for mixed precision training and 64 for single precision training per single V100 16 GB.
@ -438,6 +438,7 @@ To benchmark the inference performance on a specific batch size, run:
By default, each of these scripts runs 20 warm-up iterations and measures the next 80 iterations.
To control warm-up and benchmark length, use the `--warmup_steps`, `--num_iter` and `--iter_unit` flags.
If no `--data_dir=<path to imagenet>` flag is specified then the benchmarks will use a synthetic dataset.
The benchmark can be automated with the `inference_benchmark.sh` script provided in `resnext101-32x4d`, by simply running:
`bash ./resnext101-32x4d/inference_benchmark.sh <data dir> <data idx dir>`

5
TensorFlow/Classification/ConvNets/se-resnext101-32x4d/README.md
View file

@ -198,7 +198,7 @@ To train your model using mixed precision or TF32 with Tensor Cores or FP32, per
1. Clone the repository.
```
git clone https://github.com/NVIDIA/DeepLearningExamples
cd DeepLearningExamples/TensorFlow/Classification/RN50v1.5
cd DeepLearningExamples/TensorFlow/Classification/ConvNets
```
2. Download and preprocess the dataset.
@ -415,7 +415,7 @@ To benchmark the training performance on a specific batch size, run:
Each of these scripts runs 200 warm-up iterations and measures the first epoch.
To control warmup and benchmark length, use the `--warmup_steps`, `--num_iter` and `--iter_unit` flags. Features like XLA or DALI can be controlled
with `--use_xla` and `--use_dali` flags.
with `--use_xla` and `--use_dali` flags. If no `--data_dir=<path to imagenet>` flag is specified then the benchmarks will use a synthetic dataset.
Suggested batch sizes for training are 96 for mixed precision training and 64 for single precision training per single V100 16 GB.
@ -433,6 +433,7 @@ To benchmark the inference performance on a specific batch size, run:
By default, each of these scripts runs 20 warm-up iterations and measures the next 80 iterations.
To control warm-up and benchmark length, use the `--warmup_steps`, `--num_iter` and `--iter_unit` flags.
If no `--data_dir=<path to imagenet>` flag is specified then the benchmarks will use a synthetic dataset.
The benchmark can be automated with the `inference_benchmark.sh` script provided in `se-resnext101-32x4d`, by simply running:
`bash ./se-resnext101-32x4d/inference_benchmark.sh <data dir> <data idx dir>`