Hugging Face's logo

pietrolesci
/
bert-tiny_mnli_cdc7ea0d50

TensorBoard
Model card Files Metrics Community
bert-tiny_mnli_cdc7ea0d50 / train.log
pietrolesci's picture
pietrolesci
Upload folder using huggingface_hub
629193a verified
raw history blame contribute delete
No virus
7.85 kB
[2024-03-04 17:34:08,795][hydra][INFO] -
estimator:
accelerator: gpu
precision: bf16-true
deterministic: true
tf32_mode: high
convert_to_bettertransformer: false
callbacks:
timer:
_target_: energizer.active_learning.callbacks.Timer
lr_monitor:
_target_: energizer.callbacks.lr_monitor.LearningRateMonitor
model_checkpoint:
_target_: energizer.callbacks.model_checkpoint.ModelCheckpoint
dirpath: .checkpoints
stage: train
frequency: 1:epoch
loggers:
tensorboard:
_target_: energizer.loggers.TensorBoardLogger
root_dir: ./
name: tb_logs
version: null
data:
batch_size: 32
eval_batch_size: 128
shuffle: true
replacement: false
data_seed: 42
drop_last: false
num_workers: 8
pin_memory: true
persistent_workers: false
multiprocessing_context: null
max_length: 512
fit:
max_epochs: 20
optimizer_kwargs:
name: adamw
lr: 3.0e-05
init_kwargs:
fused: true
scheduler_kwargs:
name: constant_schedule_with_warmup
num_warmup_steps: 2000
log_interval: 100
enable_progress_bar: true
limit_train_batches: null
limit_validation_batches: null
model:
name: bert-tiny
revision: null
seed: 42
log_interval: 100
enable_progress_bar: true
limit_batches: null
seed: 42
experiment_group: training
run_name: bert-tiny_2024-03-04T17-34-08
data_path: /home/pl487/coreset-project/data/processed
dataset: mnli
======================================================================
[2024-03-04 17:34:08,796][hydra][INFO] - Seed enabled: 42
[2024-03-04 17:34:09,910][hydra][INFO] - Label distribution:
{<RunningStage.TRAIN: 'train'>: {'0-(entailment)': 130899, '1-(neutral)': 130900, '2-(contradiction)': 130903}}
[2024-03-04 17:34:21,700][hydra][INFO] - Loggers: [<energizer.loggers.tensorboard.TensorBoardLogger object at 0x7f79509062f0>]
[2024-03-04 17:34:21,700][hydra][INFO] - Callbacks: [<energizer.active_learning.callbacks.Timer object at 0x7f792e9ecfd0>, <energizer.callbacks.lr_monitor.LearningRateMonitor object at 0x7f792e9ed030>, <energizer.callbacks.model_checkpoint.ModelCheckpoint object at 0x7f792e9ed540>]
[2024-03-04 17:34:21,702][hydra][INFO] - Model summary:
Total num params: 4.4M
Of which trainable: 4.4M
With a memory footprint of 0.01GB
Total memory allocated 0.03GB
[2024-03-04 17:34:21,702][hydra][INFO] - Dataloading params:
SequenceClassificationDataloaderArgs(batch_size=32, eval_batch_size=128, num_workers=8, pin_memory=True, drop_last=False, persistent_workers=False, shuffle=True, replacement=False, data_seed=42, multiprocessing_context=None, max_length=512)
[2024-03-04 17:34:21,737][hydra][INFO] - Batch:
{<InputKeys.INPUT_IDS: 'input_ids'>: tensor([[ 101, 2092, 7910, 2008, 1005, 1055, 2785, 1997, 5793, 1045,
2812, 2027, 1005, 2128, 2130, 4755, 2009, 2000, 2000, 2073,
2085, 7910, 2008, 2027, 4748, 16874, 5562, 2006, 2694, 2017,
2113, 2065, 2115, 2065, 2017, 7910, 2017, 2113, 2031, 2589,
2023, 2030, 2065, 2017, 2342, 2023, 7910, 7910, 2057, 1005,
2222, 9790, 2005, 2017, 1998, 2017, 2123, 1005, 1056, 2031,
2000, 3477, 2149, 4983, 2017, 2021, 2059, 2054, 2027, 2123,
1005, 1056, 2425, 2017, 2003, 2008, 2065, 2017, 2065, 2027,
2663, 2017, 2507, 2068, 2012, 2560, 1037, 2353, 1997, 1996,
1997, 1996, 2518, 2008, 2027, 2663, 2061, 1045, 2123, 1005,
1056, 2113, 2009, 2003, 7910, 2009, 1005, 1055, 2893, 2000,
2022, 2062, 2449, 2085, 2738, 2084, 7910, 2941, 7910, 7149,
2007, 1996, 4126, 2084, 2007, 7910, 8529, 1996, 7910, 7750,
2027, 1996, 1996, 9559, 2024, 2074, 1999, 2009, 2005, 1996,
2769, 1045, 1005, 1049, 1045, 1005, 1049, 6427, 1045, 2113,
1045, 1045, 5993, 2007, 2017, 1045, 2228, 2017, 1005, 2128,
2613, 2017, 1005, 2128, 2200, 2157, 2008, 1996, 8801, 2323,
1045, 2228, 2027, 2323, 2031, 2019, 5020, 3815, 1997, 2017,
2113, 2672, 2027, 2064, 2031, 1037, 2261, 2021, 1045, 2228,
2087, 1997, 2068, 2323, 2022, 2025, 7910, 9559, 1999, 1996,
2755, 1998, 2008, 1005, 1055, 2126, 2126, 2027, 1005, 2310,
5407, 2046, 4331, 2009, 1005, 1055, 2138, 1997, 1996, 1996,
2375, 2477, 2017, 2113, 1996, 15932, 1998, 2673, 2021, 7910,
2021, 2053, 1045, 2113, 2057, 2009, 8529, 2057, 1005, 2128,
1999, 5374, 1998, 7910, 2057, 2031, 1996, 2168, 2518, 2058,
24375, 4667, 1998, 1998, 7910, 2027, 2024, 5599, 2068, 2041,
1045, 2812, 2074, 1996, 3180, 7173, 6251, 2518, 2027, 2292,
2068, 2041, 2138, 1997, 2027, 2123, 1005, 1056, 2031, 2151,
2173, 2000, 2562, 2000, 2404, 2068, 2061, 8307, 2008, 2071,
2017, 2113, 7910, 4983, 7910, 2042, 1037, 2350, 10048, 2021,
7910, 7910, 2130, 1996, 14255, 14141, 2135, 2210, 4933, 1045,
2812, 1996, 1996, 4364, 2008, 12386, 2651, 1999, 2637, 2003,
1996, 6778, 1998, 9444, 2008, 2089, 2022, 3251, 2009, 1005,
1055, 1037, 13742, 2030, 5850, 2030, 3649, 2842, 8529, 2027,
2024, 1996, 3924, 2008, 2024, 2183, 2000, 3477, 1998, 2027,
2024, 1996, 2028, 2008, 2024, 2183, 2000, 9015, 1998, 1996,
2060, 2711, 2017, 2113, 2065, 2027, 3477, 2065, 2027, 2065,
2027, 7910, 6545, 2068, 7910, 1998, 7910, 2059, 1996, 2017,
2113, 2051, 1996, 2375, 3138, 2058, 7910, 2431, 1996, 2051,
7910, 2027, 2593, 2292, 2068, 2175, 2030, 2027, 2131, 2125,
2007, 1037, 1037, 6251, 2138, 2027, 1005, 2310, 2018, 1037,
5160, 2008, 2017, 2113, 5683, 2008, 2008, 2027, 4694, 1005,
1056, 2035, 2362, 2043, 2027, 2106, 2009, 102, 1045, 2228,
2008, 2045, 2323, 2022, 2019, 5020, 6630, 1997, 15406, 1999,
2256, 8801, 1012, 102]]), <InputKeys.ATT_MASK: 'attention_mask'>: tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]), <InputKeys.LABELS: 'labels'>: tensor([0]), <InputKeys.ON_CPU: 'on_cpu'>: {<SpecialKeys.ID: 'uid'>: [221950]}}