Skip to content
E2E performance

E2E performance Rolling, PyTorch #396

Sign in to view logs

E2E performance Rolling, PyTorch

E2E performance Rolling, PyTorch #396

Run time

Learn about OS pricing on GitHub Actions
Job Run time
Setup
1s
Print inputs
1s
Run test matrix (huggingface, inference, amp_fp16) / Test huggingface amp_fp16 inference performance
13m 54s
Run test matrix (huggingface, inference, amp_bf16) / Test huggingface amp_bf16 inference performance
3m 42s
Run test matrix (huggingface, inference, bfloat16) / Test huggingface bfloat16 inference performance
3m 44s
Run test matrix (huggingface, inference, float32) / Test huggingface float32 inference performance
3m 6s
Run test matrix (huggingface, inference, float16) / Test huggingface float16 inference performance
2m 55s
Run test matrix (huggingface, inference-no-freezing, amp_fp16) / Test huggingface amp_fp16 inference-no-freezing performance
2m 2s
Run test matrix (huggingface, inference-no-freezing, bfloat16) / Test huggingface bfloat16 inference-no-freezing performance
2m 0s
Run test matrix (huggingface, inference-no-freezing, float16) / Test huggingface float16 inference-no-freezing performance
3m 1s
Run test matrix (huggingface, inference-no-freezing, amp_bf16) / Test huggingface amp_bf16 inference-no-freezing performance
2m 3s
Run test matrix (huggingface, inference-no-freezing, float32) / Test huggingface float32 inference-no-freezing performance
2m 6s
Run test matrix (huggingface, training, amp_fp16) / Test huggingface amp_fp16 training performance
3m 24s
Run test matrix (huggingface, training, amp_bf16) / Test huggingface amp_bf16 training performance
2m 4s
Run test matrix (huggingface, training, bfloat16) / Test huggingface bfloat16 training performance
1m 50s
Run test matrix (huggingface, training, float16) / Test huggingface float16 training performance
1m 52s
Run test matrix (huggingface, training, float32) / Test huggingface float32 training performance
1m 48s
Run test matrix (timm_models, inference, amp_fp16) / Test timm_models amp_fp16 inference performance
1m 48s
Run test matrix (timm_models, inference, amp_bf16) / Test timm_models amp_bf16 inference performance
1m 50s
Run test matrix (timm_models, inference, float32) / Test timm_models float32 inference performance
1m 50s
Run test matrix (timm_models, inference, float16) / Test timm_models float16 inference performance
1m 48s
Run test matrix (timm_models, inference, bfloat16) / Test timm_models bfloat16 inference performance
1m 52s
Run test matrix (timm_models, inference-no-freezing, amp_fp16) / Test timm_models amp_fp16 inference-no-freezing performance
1m 50s
Run test matrix (timm_models, inference-no-freezing, bfloat16) / Test timm_models bfloat16 inference-no-freezing performance
1m 51s
Run test matrix (timm_models, inference-no-freezing, float16) / Test timm_models float16 inference-no-freezing performance
1m 52s
Run test matrix (timm_models, inference-no-freezing, float32) / Test timm_models float32 inference-no-freezing performance
1m 50s
Run test matrix (timm_models, inference-no-freezing, amp_bf16) / Test timm_models amp_bf16 inference-no-freezing performance
1m 51s
Run test matrix (timm_models, training, amp_fp16) / Test timm_models amp_fp16 training performance
1m 51s
Run test matrix (timm_models, training, amp_bf16) / Test timm_models amp_bf16 training performance
1m 50s
Run test matrix (timm_models, training, bfloat16) / Test timm_models bfloat16 training performance
1m 49s
Run test matrix (timm_models, training, float32) / Test timm_models float32 training performance
1m 48s
Run test matrix (timm_models, training, float16) / Test timm_models float16 training performance
1m 52s
Run test matrix (torchbench, inference, amp_fp16) / Test torchbench amp_fp16 inference performance
1m 50s
Run test matrix (torchbench, inference, amp_bf16) / Test torchbench amp_bf16 inference performance
1m 52s
Run test matrix (torchbench, inference, bfloat16) / Test torchbench bfloat16 inference performance
1m 50s
Run test matrix (torchbench, inference, float32) / Test torchbench float32 inference performance
1m 48s
Run test matrix (torchbench, inference, float16) / Test torchbench float16 inference performance
1m 48s
Run test matrix (torchbench, inference-no-freezing, bfloat16) / Test torchbench bfloat16 inference-no-freezing performance
1m 49s
Run test matrix (torchbench, inference-no-freezing, amp_fp16) / Test torchbench amp_fp16 inference-no-freezing performance
1m 50s
Run test matrix (torchbench, inference-no-freezing, float16) / Test torchbench float16 inference-no-freezing performance
1m 51s
Run test matrix (torchbench, inference-no-freezing, float32) / Test torchbench float32 inference-no-freezing performance
1m 50s
Run test matrix (torchbench, inference-no-freezing, amp_bf16) / Test torchbench amp_bf16 inference-no-freezing performance
1m 50s
Run test matrix (torchbench, training, amp_fp16) / Test torchbench amp_fp16 training performance
1m 49s
Run test matrix (torchbench, training, amp_bf16) / Test torchbench amp_bf16 training performance
1m 47s
Run test matrix (torchbench, training, bfloat16) / Test torchbench bfloat16 training performance
1m 50s
Run test matrix (torchbench, training, float32) / Test torchbench float32 training performance
1m 50s
Run test matrix (torchbench, training, float16) / Test torchbench float16 training performance
1m 48s
1h 44m 27s