File size: 1,403 Bytes
fdc4101
 
 
 
 
 
 
 
 
d9993eb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d33873d
 
fdc4101
 
 
e867aeb
446c112
e867aeb
d9993eb
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
MODEL="t5-base-dutch"

MODEL_DIR="${HOME}/${MODEL}"

mkdir -p "${MODEL_DIR}/runs"

# T5 paper lr 0.01 with batch size 128
# We have a batch size of 8 devices * 32 = 256, so lr = 0.01/2

while true; do

  # Set the seed to random before each run, so date shuffling per epoch is different each run.
  # This kills reproducibility, but is required as long as during training ValueError can be raised.
  SEED=$RANDOM

  ./run_t5_mlm_flax_custom_dataset.py \
      --output_dir="${MODEL_DIR}" \
      --model_type="t5" \
      --config_name="flax-community/${MODEL}" \
      --tokenizer_name="${MODEL_DIR}" \
      --seed="${SEED}" \
      --preprocessing_num_workers="96" \
      --do_train --do_eval \
      --adafactor \
      --max_seq_length="512" \
      --per_device_train_batch_size="32" \
      --per_device_eval_batch_size="32" \
      --learning_rate="5e-3" \
      --dtype="bfloat16" \
      --overwrite_output_dir \
      --num_train_epochs="3" \
      --logging_steps="50" \
      --save_steps="501" \
      --eval_steps="10000000" \
      --resume_from_checkpoint="${MODEL_DIR}" \
      --warmup_steps="3413"

#       \
#      --push_to_hub

  echo "RESTARTING"
  sleep 20
done
#
#     \


#git add pytorch_model.bin
#git commit -m "Update pytorch model after training"
#git push origin main

#    --gradient_accumulation_steps="2" \

#    --resume_from_checkpoint="${MODEL_DIR}/ckpt-18000" \