forked from tinygrad/tinygrad
tinybox green mlperf submission
This commit is contained in:
Francis Lata
committed by
Chen-Yu Yang
Chen-Yu Yang
parent
8acf215660
commit
edd4fa3b36
@@ -11,6 +11,7 @@ notebooks
|
||||
*.txt
|
||||
build
|
||||
!examples/tinychat/assets/cdn.jsdelivr.net/npm/[email protected]/build/
|
||||
!examples/mlperf/training_submission_*/**/*.txt
|
||||
/dist
|
||||
*.egg-info
|
||||
/env
|
||||
|
||||
@@ -358,11 +358,43 @@ def train_retinanet():
|
||||
|
||||
config, target_metric = {}, 0.34
|
||||
|
||||
config["SEED"] = SEED = getenv("SEED", random.SystemRandom().randint(0, 2**32 - 1))
|
||||
Tensor.manual_seed(SEED)
|
||||
|
||||
NUM_CLASSES = len(MLPERF_CLASSES)
|
||||
BASEDIR = getenv("BASEDIR", BASEDIR)
|
||||
BENCHMARK = getenv("BENCHMARK")
|
||||
# INITMLPERF = getenv("INITMLPERF")
|
||||
INITMLPERF = getenv("INITMLPERF")
|
||||
RUNMLPERF = getenv("RUNMLPERF")
|
||||
|
||||
if getenv("LOGMLPERF"):
|
||||
from mlperf_logging import mllog
|
||||
import mlperf_logging.mllog.constants as mllog_constants
|
||||
|
||||
mllog.config(filename=f"result_retinanet_{SEED}.log")
|
||||
mllog.config(root_dir=Path(__file__).parents[3].as_posix())
|
||||
MLLOGGER = mllog.get_mllogger()
|
||||
MLLOGGER.logger.propagate = False
|
||||
|
||||
if INITMLPERF:
|
||||
assert BENCHMARK, "BENCHMARK must be set for INITMLPERF"
|
||||
MLLOGGER.event(key=mllog_constants.SUBMISSION_ORG, value="tinycorp")
|
||||
MLLOGGER.event(key=mllog_constants.SUBMISSION_PLATFORM, value=getenv("SUBMISSION_PLATFORM", "tinybox"))
|
||||
MLLOGGER.event(key=mllog_constants.SUBMISSION_DIVISION, value=mllog_constants.CLOSED)
|
||||
MLLOGGER.event(key=mllog_constants.SUBMISSION_STATUS, value=mllog_constants.ONPREM)
|
||||
|
||||
MLLOGGER.event(key=mllog_constants.SUBMISSION_BENCHMARK, value=mllog_constants.RETINANET)
|
||||
|
||||
diskcache_clear()
|
||||
MLLOGGER.event(key=mllog_constants.CACHE_CLEAR, value=True)
|
||||
MLLOGGER.start(key=mllog_constants.INIT_START)
|
||||
|
||||
if RUNMLPERF:
|
||||
MLLOGGER.start(key=mllog_constants.RUN_START)
|
||||
MLLOGGER.event(key=mllog_constants.SEED, value=SEED)
|
||||
else:
|
||||
MLLOGGER = None
|
||||
|
||||
config["gpus"] = GPUS = [f"{Device.DEFAULT}:{i}" for i in range(getenv("GPUS", 6))]
|
||||
|
||||
for x in GPUS: Device[x]
|
||||
@@ -415,24 +447,21 @@ def train_retinanet():
|
||||
return out.to(GPUS[0]).realize()
|
||||
|
||||
# ** hyperparameters **
|
||||
config["seed"] = SEED = getenv("SEED", random.SystemRandom().randint(0, 2**32 - 1))
|
||||
config["bs"] = BS = getenv("BS", 16 * len(GPUS) if dtypes.default_float == dtypes.float16 else 12 * len(GPUS))
|
||||
config["eval_bs"] = EVAL_BS = getenv("EVAL_BS", BS)
|
||||
config["epochs"] = EPOCHS = getenv("EPOCHS", 4)
|
||||
config["train_beam"] = TRAIN_BEAM = getenv("TRAIN_BEAM", BEAM.value)
|
||||
config["eval_beam"] = EVAL_BEAM = getenv("EVAL_BEAM", BEAM.value)
|
||||
config["lr"] = lr = getenv("LR", 9.5e-5 * (BS / 96))
|
||||
config["loss_scaler"] = loss_scaler = getenv("LOSS_SCALER", 2**11 if dtypes.default_float == dtypes.float16 else 1.0)
|
||||
config["default_float"] = dtypes.default_float.name
|
||||
config["eval_freq"] = eval_freq = getenv("EVAL_FREQ", 1)
|
||||
config["BS"] = BS = getenv("BS", 16 * len(GPUS) if dtypes.default_float == dtypes.float16 else 12 * len(GPUS))
|
||||
config["EVAL_BS"] = EVAL_BS = getenv("EVAL_BS", BS)
|
||||
config["EPOCHS"] = EPOCHS = getenv("EPOCHS", 4)
|
||||
config["TRAIN_BEAM"] = TRAIN_BEAM = getenv("TRAIN_BEAM", BEAM.value)
|
||||
config["EVAL_BEAM"] = EVAL_BEAM = getenv("EVAL_BEAM", BEAM.value)
|
||||
config["LR"] = lr = getenv("LR", 9.5e-5 * (BS / 96))
|
||||
config["LOSS_SCALER"] = loss_scaler = getenv("LOSS_SCALER", 2**11 if dtypes.default_float == dtypes.float16 else 1.0)
|
||||
config["DEFAULT_FLOAT"] = dtypes.default_float.name
|
||||
config["EVAL_FREQ"] = eval_freq = getenv("EVAL_FREQ", 1)
|
||||
|
||||
# ** initialize wandb **
|
||||
if (WANDB:=getenv("WANDB")):
|
||||
import wandb
|
||||
wandb.init(config=config, project="MLPerf-RetinaNet")
|
||||
|
||||
if SEED: Tensor.manual_seed(SEED)
|
||||
|
||||
# ** model initializers **
|
||||
resnet.BatchNorm = FrozenBatchNorm2dRetinaNet
|
||||
resnet.Linear = Linear
|
||||
@@ -465,8 +494,24 @@ def train_retinanet():
|
||||
optim = Adam(params, lr=lr)
|
||||
|
||||
# ** dataset **
|
||||
config["steps_in_train_epoch"] = steps_in_train_epoch = round_up(get_dataset_count((base_dir_path:=Path(BASEDIR)), False), BS) // BS
|
||||
config["steps_in_val_epoch"] = steps_in_val_epoch = (round_up(get_dataset_count(base_dir_path, True), EVAL_BS) // EVAL_BS)
|
||||
config["STEPS_IN_TRAIN_EPOCH"] = steps_in_train_epoch = round_up(get_dataset_count((base_dir_path:=Path(BASEDIR)), False), BS) // BS
|
||||
config["STEPS_IN_VAL_EPOCH"] = steps_in_val_epoch = (round_up(get_dataset_count(base_dir_path, True), EVAL_BS) // EVAL_BS)
|
||||
|
||||
# log mlperf hparams
|
||||
if MLLOGGER:
|
||||
if RUNMLPERF:
|
||||
MLLOGGER.event(key=mllog_constants.GLOBAL_BATCH_SIZE, value=config["BS"])
|
||||
MLLOGGER.event(key=mllog_constants.TRAIN_SAMPLES, value=config["STEPS_IN_TRAIN_EPOCH"])
|
||||
MLLOGGER.event(key=mllog_constants.EVAL_SAMPLES, value=config["STEPS_IN_VAL_EPOCH"])
|
||||
MLLOGGER.event(key=mllog_constants.EPOCH_COUNT, value=config["EPOCHS"])
|
||||
MLLOGGER.event(key=mllog_constants.FIRST_EPOCH_NUM, value=start_epoch)
|
||||
|
||||
MLLOGGER.event(key=mllog_constants.OPT_NAME, value=mllog_constants.ADAM)
|
||||
MLLOGGER.event(key=mllog_constants.OPT_BASE_LR, value=config["LR"])
|
||||
MLLOGGER.event(key=mllog_constants.OPT_WEIGHT_DECAY, value=0)
|
||||
MLLOGGER.event(key=mllog_constants.OPT_LR_WARMUP_EPOCHS, value=0)
|
||||
MLLOGGER.event(key=mllog_constants.OPT_LR_WARMUP_FACTOR, value=0)
|
||||
MLLOGGER.event(key=mllog_constants.GRADIENT_ACCUMULATION_STEPS, value=1)
|
||||
|
||||
if RUNMLPERF:
|
||||
train_dataset = COCO(download_dataset(BASEDIR, "train"))
|
||||
@@ -477,13 +522,16 @@ def train_retinanet():
|
||||
|
||||
for e in range(start_epoch, EPOCHS):
|
||||
# ** training loop **
|
||||
if MLLOGGER and RUNMLPERF:
|
||||
MLLOGGER.start(key=mllog_constants.EPOCH_START, value=e + 1, metadata={"epoch_num": e + 1})
|
||||
|
||||
BEAM.value = TRAIN_BEAM
|
||||
|
||||
if not RUNMLPERF:
|
||||
i, proc = 0, _fake_data_get(BS)
|
||||
else:
|
||||
train_dataloader = batch_load_retinanet(train_dataset, False, base_dir_path, batch_size=BS, seed=SEED)
|
||||
it = iter(tqdm(train_dataloader, total=steps_in_train_epoch, desc=f"epoch {e}", disable=BENCHMARK))
|
||||
it = iter(tqdm(train_dataloader, total=steps_in_train_epoch, desc=f"epoch {e + 1}", disable=BENCHMARK))
|
||||
i, proc = 0, _data_get(it)
|
||||
|
||||
prev_cookies = []
|
||||
@@ -545,8 +593,14 @@ def train_retinanet():
|
||||
if (TRAIN_BEAM or EVAL_BEAM) and e == start_epoch: break
|
||||
return
|
||||
|
||||
if MLLOGGER and RUNMLPERF:
|
||||
MLLOGGER.event(key=mllog_constants.EPOCH_STOP, value=e + 1, metadata={"epoch_num": e + 1})
|
||||
|
||||
# ** eval loop **
|
||||
if (e + 1) % eval_freq == 0:
|
||||
if MLLOGGER and RUNMLPERF:
|
||||
MLLOGGER.start(key=mllog_constants.EVAL_START, value=e + 1, metadata={"epoch_num": e + 1})
|
||||
|
||||
BEAM.value = EVAL_BEAM
|
||||
|
||||
if getenv("RESET_STEP", 1): _train_step.reset()
|
||||
@@ -594,12 +648,15 @@ def train_retinanet():
|
||||
proc, next_proc = next_proc, None
|
||||
i += 1
|
||||
|
||||
if i == BENCHMARK:
|
||||
return
|
||||
|
||||
et = time.time()
|
||||
eval_times.append(et - st)
|
||||
|
||||
if i == BENCHMARK:
|
||||
# assume INITMLPERF has BENCHMARK set
|
||||
if MLLOGGER and INITMLPERF:
|
||||
MLLOGGER.event(key=mllog_constants.INIT_STOP)
|
||||
return
|
||||
|
||||
if getenv("RESET_STEP", 1): _eval_step.reset()
|
||||
total_fw_time = sum(eval_times) / len(eval_times)
|
||||
|
||||
@@ -617,8 +674,16 @@ def train_retinanet():
|
||||
if WANDB:
|
||||
wandb.log({"eval/forward_time": total_fw_time, "eval/metric": val_metric, "epoch": e + 1})
|
||||
|
||||
if MLLOGGER:
|
||||
MLLOGGER.event(key=mllog_constants.EVAL_ACCURACY, value=val_metric, metadata={"epoch_num": e + 1}, clear_line=True)
|
||||
MLLOGGER.end(key=mllog_constants.EVAL_STOP, value=e + 1, metadata={"epoch_num": e + 1})
|
||||
|
||||
if val_metric >= target_metric:
|
||||
print(colored(f"target metric reached: {val_metric:.2f}/{target_metric:.2f}", color="green"))
|
||||
|
||||
if MLLOGGER:
|
||||
MLLOGGER.end(key=mllog_constants.RUN_STOP, metadata={"status": mllog_constants.SUCCESS})
|
||||
|
||||
break
|
||||
|
||||
def train_unet3d():
|
||||
|
||||
+38
@@ -0,0 +1,38 @@
|
||||
# 1. Problem
|
||||
|
||||
This problem uses RetinaNet for SSD.
|
||||
|
||||
## Requirements
|
||||
|
||||
Install tinygrad and mlperf-logging from master.
|
||||
```
|
||||
git clone https://github.com/tinygrad/tinygrad.git
|
||||
python3 -m pip install -e ".[mlperf]"
|
||||
```
|
||||
|
||||
Also install the following dependencies:
|
||||
```
|
||||
pip install tqdm numpy pycocotools boto3 pandas torch torchvision
|
||||
```
|
||||
|
||||
### tinybox_green
|
||||
Install the p2p driver per [README](https://github.com/tinygrad/open-gpu-kernel-modules/blob/550.54.15-p2p/README.md)
|
||||
This is the default on production tinybox green.
|
||||
|
||||
# 2. Directions
|
||||
|
||||
## Steps to download data
|
||||
|
||||
Run the following:
|
||||
```
|
||||
BASEDIR=/raid/datasets/openimages python3 extra/datasets/openimages.py
|
||||
```
|
||||
|
||||
## Running
|
||||
|
||||
### tinybox_green
|
||||
|
||||
#### Steps to run benchmark
|
||||
```
|
||||
examples/mlperf/training_submission_v5.0/tinycorp/benchmarks/retinanet/implementations/tinybox_green/run_and_time.sh
|
||||
```
|
||||
+23
@@ -0,0 +1,23 @@
|
||||
#!/bin/bash
|
||||
|
||||
export PYTHONPATH="." NV=1
|
||||
export MODEL="retinanet"
|
||||
export SUBMISSION_PLATFORM="tinybox_green"
|
||||
export DEFAULT_FLOAT="HALF" GPUS=6 BS=96 EVAL_BS=96
|
||||
|
||||
export TRAIN_BEAM=2 BEAM_UOPS_MAX=1500 BEAM_UPCAST_MAX=64 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=5 BEAM_PADTO=0
|
||||
export IGNORE_JIT_FIRST_BEAM=1
|
||||
export BASEDIR="/raid/datasets/openimages"
|
||||
|
||||
# pip install -e ".[mlperf]"
|
||||
export LOGMLPERF=1
|
||||
|
||||
export SEED=$RANDOM
|
||||
DATETIME=$(date "+%m%d%H%M")
|
||||
LOGFILE="retinanet_green_${DATETIME}_${SEED}.log"
|
||||
|
||||
# init
|
||||
BENCHMARK=10 INITMLPERF=1 python3 examples/mlperf/model_train.py | tee $LOGFILE
|
||||
|
||||
# run
|
||||
PARALLEL=0 RUNMLPERF=1 python3 examples/mlperf/model_train.py | tee -a $LOGFILE
|
||||
+46
@@ -0,0 +1,46 @@
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596628137, "event_type": "POINT_IN_TIME", "key": "submission_org", "value": "tinycorp", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 382}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596628178, "event_type": "POINT_IN_TIME", "key": "submission_platform", "value": "tinybox_green", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 383}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596628178, "event_type": "POINT_IN_TIME", "key": "submission_division", "value": "closed", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 384}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596628178, "event_type": "POINT_IN_TIME", "key": "submission_status", "value": "onprem", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 385}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596628178, "event_type": "POINT_IN_TIME", "key": "submission_benchmark", "value": "retinanet", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 387}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596629954, "event_type": "POINT_IN_TIME", "key": "cache_clear", "value": true, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 390}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745596629955, "event_type": "INTERVAL_START", "key": "init_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 391}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598065772, "event_type": "POINT_IN_TIME", "key": "init_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 658}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598081470, "event_type": "INTERVAL_START", "key": "run_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 394}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598081512, "event_type": "POINT_IN_TIME", "key": "seed", "value": 23282, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 395}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088273, "event_type": "POINT_IN_TIME", "key": "global_batch_size", "value": 96, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 504}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "train_samples", "value": 12191, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 505}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "eval_samples", "value": 259, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 506}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "epoch_count", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 507}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "first_epoch_num", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 508}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "opt_name", "value": "adam", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 510}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "opt_base_learning_rate", "value": 9.5e-05, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 511}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "opt_weight_decay", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 512}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088274, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_epochs", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 513}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088275, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_factor", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 514}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598088275, "event_type": "POINT_IN_TIME", "key": "gradient_accumulation_steps", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 515}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745598144406, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 527, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745605078062, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 598, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745605078063, "event_type": "INTERVAL_START", "key": "eval_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 603, "epoch_num": 1}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745610378469, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.2608930553164607, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 679, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745610378469, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 680, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745610378469, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 527, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745616941326, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 598, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745616941327, "event_type": "INTERVAL_START", "key": "eval_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 603, "epoch_num": 2}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745622185857, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.31207695716564665, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 679, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745622185858, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 680, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745622185858, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 527, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745628713800, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 598, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745628713800, "event_type": "INTERVAL_START", "key": "eval_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 603, "epoch_num": 3}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745633828548, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.32695300496649193, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 679, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745633828548, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 680, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745633828549, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 527, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745640403678, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 598, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745640403679, "event_type": "INTERVAL_START", "key": "eval_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 603, "epoch_num": 4}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745645485614, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.34190927146960864, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 679, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745645485615, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 680, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745645485615, "event_type": "INTERVAL_END", "key": "run_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 686, "status": "success"}}
|
||||
+46
@@ -0,0 +1,46 @@
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708052929, "event_type": "POINT_IN_TIME", "key": "submission_org", "value": "tinycorp", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 380}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708052970, "event_type": "POINT_IN_TIME", "key": "submission_platform", "value": "tinybox_green", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 381}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708052970, "event_type": "POINT_IN_TIME", "key": "submission_division", "value": "closed", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 382}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708052970, "event_type": "POINT_IN_TIME", "key": "submission_status", "value": "onprem", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 383}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708052970, "event_type": "POINT_IN_TIME", "key": "submission_benchmark", "value": "retinanet", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 385}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708055312, "event_type": "POINT_IN_TIME", "key": "cache_clear", "value": true, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 388}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708055312, "event_type": "INTERVAL_START", "key": "init_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 389}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709484510, "event_type": "POINT_IN_TIME", "key": "init_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 656}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709499880, "event_type": "INTERVAL_START", "key": "run_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 392}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709499922, "event_type": "POINT_IN_TIME", "key": "seed", "value": 3218, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 393}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506804, "event_type": "POINT_IN_TIME", "key": "global_batch_size", "value": 96, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 502}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506805, "event_type": "POINT_IN_TIME", "key": "train_samples", "value": 12191, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 503}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506805, "event_type": "POINT_IN_TIME", "key": "eval_samples", "value": 259, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 504}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506805, "event_type": "POINT_IN_TIME", "key": "epoch_count", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 505}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "first_epoch_num", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 506}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "opt_name", "value": "adam", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 508}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "opt_base_learning_rate", "value": 9.5e-05, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 509}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "opt_weight_decay", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 510}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_epochs", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 511}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_factor", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 512}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709506806, "event_type": "POINT_IN_TIME", "key": "gradient_accumulation_steps", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 513}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745709564057, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745716423332, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745716423333, "event_type": "INTERVAL_START", "key": "eval_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 1}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745721892086, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.2644758301871188, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745721892087, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745721892087, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745728717917, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745728717918, "event_type": "INTERVAL_START", "key": "eval_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 2}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745734129092, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3183940553292647, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745734129092, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745734129092, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745740758848, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745740758849, "event_type": "INTERVAL_START", "key": "eval_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 3}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745746017219, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.330829179299047, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745746017219, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745746017219, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745752685505, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745752685506, "event_type": "INTERVAL_START", "key": "eval_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 4}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757915230, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3430538198992862, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757915231, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757915231, "event_type": "INTERVAL_END", "key": "run_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 684, "status": "success"}}
|
||||
+46
@@ -0,0 +1,46 @@
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757942370, "event_type": "POINT_IN_TIME", "key": "submission_org", "value": "tinycorp", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 380}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757942411, "event_type": "POINT_IN_TIME", "key": "submission_platform", "value": "tinybox_green", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 381}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757942411, "event_type": "POINT_IN_TIME", "key": "submission_division", "value": "closed", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 382}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757942411, "event_type": "POINT_IN_TIME", "key": "submission_status", "value": "onprem", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 383}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757942411, "event_type": "POINT_IN_TIME", "key": "submission_benchmark", "value": "retinanet", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 385}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757943058, "event_type": "POINT_IN_TIME", "key": "cache_clear", "value": true, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 388}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745757943059, "event_type": "INTERVAL_START", "key": "init_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 389}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759379793, "event_type": "POINT_IN_TIME", "key": "init_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 656}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759394363, "event_type": "INTERVAL_START", "key": "run_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 392}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759394404, "event_type": "POINT_IN_TIME", "key": "seed", "value": 7068, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 393}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401265, "event_type": "POINT_IN_TIME", "key": "global_batch_size", "value": 96, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 502}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401265, "event_type": "POINT_IN_TIME", "key": "train_samples", "value": 12191, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 503}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "eval_samples", "value": 259, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 504}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "epoch_count", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 505}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "first_epoch_num", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 506}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "opt_name", "value": "adam", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 508}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "opt_base_learning_rate", "value": 9.5e-05, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 509}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "opt_weight_decay", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 510}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401266, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_epochs", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 511}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401267, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_factor", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 512}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759401267, "event_type": "POINT_IN_TIME", "key": "gradient_accumulation_steps", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 513}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759458864, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745766229351, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745766229352, "event_type": "INTERVAL_START", "key": "eval_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 1}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745771664180, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.2618442233208197, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745771664180, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745771664181, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745778271730, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745778271731, "event_type": "INTERVAL_START", "key": "eval_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 2}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745783649281, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3050222595524408, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745783649281, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745783649282, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745790269694, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745790269695, "event_type": "INTERVAL_START", "key": "eval_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 3}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745795622203, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.33438554461867026, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745795622204, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745795622204, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745802201592, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745802201593, "event_type": "INTERVAL_START", "key": "eval_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 4}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745807409644, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3413173788267323, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745807409644, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745807409644, "event_type": "INTERVAL_END", "key": "run_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 684, "status": "success"}}
|
||||
+46
@@ -0,0 +1,46 @@
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708740629, "event_type": "POINT_IN_TIME", "key": "submission_org", "value": "tinycorp", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 380}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708740670, "event_type": "POINT_IN_TIME", "key": "submission_platform", "value": "tinybox_green", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 381}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708740670, "event_type": "POINT_IN_TIME", "key": "submission_division", "value": "closed", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 382}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708740670, "event_type": "POINT_IN_TIME", "key": "submission_status", "value": "onprem", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 383}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708740670, "event_type": "POINT_IN_TIME", "key": "submission_benchmark", "value": "retinanet", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 385}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708753515, "event_type": "POINT_IN_TIME", "key": "cache_clear", "value": true, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 388}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745708753515, "event_type": "INTERVAL_START", "key": "init_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 389}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710196875, "event_type": "POINT_IN_TIME", "key": "init_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 656}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710211866, "event_type": "INTERVAL_START", "key": "run_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 392}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710211906, "event_type": "POINT_IN_TIME", "key": "seed", "value": 1934, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 393}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219928, "event_type": "POINT_IN_TIME", "key": "global_batch_size", "value": 96, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 502}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219929, "event_type": "POINT_IN_TIME", "key": "train_samples", "value": 12191, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 503}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219929, "event_type": "POINT_IN_TIME", "key": "eval_samples", "value": 259, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 504}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219929, "event_type": "POINT_IN_TIME", "key": "epoch_count", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 505}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219929, "event_type": "POINT_IN_TIME", "key": "first_epoch_num", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 506}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219930, "event_type": "POINT_IN_TIME", "key": "opt_name", "value": "adam", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 508}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219930, "event_type": "POINT_IN_TIME", "key": "opt_base_learning_rate", "value": 9.5e-05, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 509}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219930, "event_type": "POINT_IN_TIME", "key": "opt_weight_decay", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 510}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219930, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_epochs", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 511}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219930, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_factor", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 512}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710219930, "event_type": "POINT_IN_TIME", "key": "gradient_accumulation_steps", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 513}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745710276595, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745717038732, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745717038733, "event_type": "INTERVAL_START", "key": "eval_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 1}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745722476155, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.24994336549495808, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745722476156, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745722476156, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745729177485, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745729177486, "event_type": "INTERVAL_START", "key": "eval_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 2}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745734589630, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.30947442932060776, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745734589630, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745734589630, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745741107714, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745741107714, "event_type": "INTERVAL_START", "key": "eval_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 3}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745746523920, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3304143886715271, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745746523920, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745746523920, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745753134001, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745753134002, "event_type": "INTERVAL_START", "key": "eval_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 4}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758428287, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3429861420134466, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758428288, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758428288, "event_type": "INTERVAL_END", "key": "run_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 684, "status": "success"}}
|
||||
+45
@@ -0,0 +1,45 @@
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758455763, "event_type": "POINT_IN_TIME", "key": "submission_org", "value": "tinycorp", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 380}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758455804, "event_type": "POINT_IN_TIME", "key": "submission_platform", "value": "tinybox_green", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 381}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758455804, "event_type": "POINT_IN_TIME", "key": "submission_division", "value": "closed", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 382}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758455804, "event_type": "POINT_IN_TIME", "key": "submission_status", "value": "onprem", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 383}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758455805, "event_type": "POINT_IN_TIME", "key": "submission_benchmark", "value": "retinanet", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 385}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758457940, "event_type": "POINT_IN_TIME", "key": "cache_clear", "value": true, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 388}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745758457941, "event_type": "INTERVAL_START", "key": "init_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 389}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759900517, "event_type": "POINT_IN_TIME", "key": "init_stop", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 656}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759915495, "event_type": "INTERVAL_START", "key": "run_start", "value": null, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 392}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759915536, "event_type": "POINT_IN_TIME", "key": "seed", "value": 25159, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 393}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922365, "event_type": "POINT_IN_TIME", "key": "global_batch_size", "value": 96, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 502}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922366, "event_type": "POINT_IN_TIME", "key": "train_samples", "value": 12191, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 503}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922366, "event_type": "POINT_IN_TIME", "key": "eval_samples", "value": 259, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 504}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922366, "event_type": "POINT_IN_TIME", "key": "epoch_count", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 505}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922366, "event_type": "POINT_IN_TIME", "key": "first_epoch_num", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 506}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922366, "event_type": "POINT_IN_TIME", "key": "opt_name", "value": "adam", "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 508}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922366, "event_type": "POINT_IN_TIME", "key": "opt_base_learning_rate", "value": 9.5e-05, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 509}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922367, "event_type": "POINT_IN_TIME", "key": "opt_weight_decay", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 510}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922367, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_epochs", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 511}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922367, "event_type": "POINT_IN_TIME", "key": "opt_learning_rate_warmup_factor", "value": 0, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 512}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759922367, "event_type": "POINT_IN_TIME", "key": "gradient_accumulation_steps", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 513}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745759981024, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745766937876, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745766937877, "event_type": "INTERVAL_START", "key": "eval_start", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 1}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745772433927, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.25660616888772175, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745772433927, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 1, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 1}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745772433927, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745779249804, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745779249805, "event_type": "INTERVAL_START", "key": "eval_start", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 2}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745784709047, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3114751446994825, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745784709048, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 2, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 2}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745784709048, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745791366481, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745791366482, "event_type": "INTERVAL_START", "key": "eval_start", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 3}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745796796512, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.33395135022162803, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745796796512, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 3, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 3}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745796796512, "event_type": "INTERVAL_START", "key": "epoch_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 525, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745803562272, "event_type": "POINT_IN_TIME", "key": "epoch_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 596, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745803562273, "event_type": "INTERVAL_START", "key": "eval_start", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 601, "epoch_num": 4}}
|
||||
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745808971898, "event_type": "POINT_IN_TIME", "key": "eval_accuracy", "value": 0.3397162205764848, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 677, "epoch_num": 4}}
|
||||
:::MLLOG {"namespace": "", "time_ms": 1745808971899, "event_type": "INTERVAL_END", "key": "eval_stop", "value": 4, "metadata": {"file": "tinygrad/examples/mlperf/model_train.py", "lineno": 678, "epoch_num": 4}}
|
||||
Reference in New Issue
Block a user