run_mlm_flax_stream.sh

export MODEL_DIR=/home/Z6HJB/roberta-base-danish/roberta-base-danish/

source /home/Z6HJB/test/bin/activate

python3 ./src/run_mlm_flax_stream.py \
    --config_name="${MODEL_DIR}" \
    --output_dir="${MODEL_DIR}" \
    --tokenizer_name="${MODEL_DIR}" \
    --model_type="roberta" \
    --dataset_name="mc4" \
    --dataset_config_name="unshuffled_deduplicated_en" \
    --max_seq_length="128" \
    --per_device_train_batch_size="128" \
    --per_device_eval_batch_size="128" \
    --learning_rate="3e-4" \
    --warmup_steps="1000" \
    --overwrite_output_dir \
    --adam_beta1="0.9" \
    --adam_beta2="0.98" \
    --num_train_steps="300000" \
    --num_eval_samples="5000" \
    --save_steps="1000" \
    --logging_steps="250" \
    --eval_steps="1000" \
    #--push_to_hub \ currently results in this error: ValueError: If not specifying `clone_from`, you need to pass Repository a valid git clone.
    #--model_name_or_path="${MODEL_DIR}" \ used to continue pretrained