issues
search
krai
/
axs2mlperf
Automated KRAI X workflows for reproducing MLPerf Inference submissions
https://krai.ai
MIT License
1
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Better llm model/tokeniser downloading
#61
Akshat-Tripathi
closed
1 month ago
0
Add recipe for downloading llama2 model and dataset for training
#60
maria-18-git
opened
2 months ago
1
Update numpy version to 2.0.2, fix --dataset-file path.
#59
sahelib25
closed
2 months ago
0
Remove additional mixtral accuracy ranges checks.
#58
sahelib25
opened
3 months ago
0
Llm-dev
#56
sahelib25
closed
3 months ago
1
Convert Llama2 pickle to Llama3
#55
G4V
opened
3 months ago
14
A more detailed summary of performance runs available with 'get performance' command
#54
bmsgit1
closed
3 months ago
4
Cleaned up MOE dataset recipes and made them consistent with OpenOrca's
#53
Akshat-Tripathi
closed
3 months ago
1
Detokenised log now preserves qsl_idx, seq_id and token_count fields
#52
Akshat-Tripathi
closed
3 months ago
0
Updated call to axs2kilt-dev/llama2_hf_weights_recipe
#51
Akshat-Tripathi
closed
4 months ago
1
Fixed bug in setting the number of devices
#50
Akshat-Tripathi
closed
4 months ago
0
Add reference code for `mixtral-8x7b` in `axs`
#49
maria-18-git
opened
4 months ago
29
Run reference code for `mixtral-8x7b`
#48
maria-18-git
closed
2 months ago
17
Replaced the call to .removesuffix() since it's python 3.9+
#47
Akshat-Tripathi
closed
5 months ago
0
Fix sdxl submitter
#46
sahelib25
closed
5 months ago
0
Improved units for "parse_summary"
#45
Akshat-Tripathi
closed
5 months ago
0
Llm detokeniser
#44
Akshat-Tripathi
closed
5 months ago
0
Fix to get checkpoint_path and model_dir path from specific llama2 model variant.
#43
sahelib25
closed
6 months ago
0
Move lscpu parser to axs2cpu-dev.
#42
sahelib25
closed
5 months ago
0
Added server query coalescing
#41
Akshat-Tripathi
closed
7 months ago
0
Update all entries to support of `entry_creator`
#40
maria-18-git
opened
8 months ago
10
Update data_axs.json for bert,resnet50,retinanet to support validate_accuracy.
#39
sahelib25
closed
8 months ago
0
Add optional sdk_ver parameter to the submitter.
#38
sahelib25
closed
5 months ago
0
Power fixes
#37
sahelib25
closed
9 months ago
0
Cleanup code and fix accessing testing entry for power.
#36
sahelib25
closed
9 months ago
0
Merge master into readme
#35
sahelib25
closed
10 months ago
0
Support cpupower
#34
psyhtest
opened
10 months ago
2
New Features
#33
sahelib25
closed
10 months ago
0
Add Stable Diffusion Reference(text to image) to `axs`
#32
maria-18-git
closed
8 months ago
11
Completed issue https://github.com/krai/axs2qaic-dev/issues/24
#31
Akshat-Tripathi
closed
1 year ago
0
Fixed `user.config` creation for `loadgen_multistreamness`, `loadgen_count_override`, `loadgen_duration_s`
#30
maria-18-git
closed
1 year ago
14
Add rule for `collection` creating
#29
maria-18-git
closed
1 year ago
4
Move 'work_collection' from `output_entry` as input parameter
#28
maria-18-git
closed
1 year ago
19
Remove links in power experiments
#27
maria-18-git
closed
1 year ago
11
Change `power` experiment type to `collection`
#26
maria-18-git
opened
1 year ago
0
Add `experiment_begin_timestamp` and `experiment_end_timestamp` for all experiments
#25
maria-18-git
closed
1 year ago
6
Add `axs zip` for zip archive creation
#24
maria-18-git
opened
1 year ago
0
Add creating and running of container input parameters for all benchmarks
#23
maria-18-git
closed
10 months ago
23
Supporting of `ipmitool` sensors for all benchmarks
#22
maria-18-git
opened
1 year ago
0
Move `param_to_conf_pair` to `data_axs.json` in `base_loadgen_program`
#21
maria-18-git
closed
1 year ago
3
Add automated tests. Trigger pull request to stable when master is updated and passed the tests.
#20
elimkwan
closed
1 year ago
0
Support `min_duration` for quick run of experiments
#19
maria-18-git
closed
1 year ago
11
Support no-ranging mode for power experiments
#18
maria-18-git
closed
1 year ago
11
Fix demo and tests
#17
elimkwan
closed
1 year ago
0
Create program for next word prediction( `GPT-J`)
#16
maria-18-git
closed
8 months ago
3
Change `framework=onnx` to `framework=onnxrt` in `axs2mlperf`
#15
maria-18-git
closed
1 year ago
2
Add `lambada` dataset for `gptj`
#14
maria-18-git
closed
8 months ago
2
Add performance test to `test_axs2mlperf.sh` for all benchmarks
#13
maria-18-git
opened
1 year ago
0
Check correct work of `gptj_reference_loadgen` after updates
#12
maria-18-git
closed
1 year ago
13
Add tests for object_detection(retinanet_openimages), image_classification, bert to test_axs2mlperf.sh
#11
maria-18-git
closed
1 year ago
4
Next