test

data/text/default-952efb7fdd4e970d/0.0.0/21a506d1b2b34316b1e82d0bd79066905d846e5d7e619823c0dd338d6f1fa6ad/cache-53e75324430950d8.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21163fe350f8123f395134d1e35fa04005898f9dd27b9964d000823314013cdd
+size 325905944

data/text/default-952efb7fdd4e970d/0.0.0/21a506d1b2b34316b1e82d0bd79066905d846e5d7e619823c0dd338d6f1fa6ad/dataset_info.json

@@ -0,0 +1 @@
+{"description": "", "citation": "", "homepage": "", "license": "", "features": {"text": {"dtype": "string", "_type": "Value"}}, "builder_name": "text", "config_name": "default", "version": {"version_str": "0.0.0", "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 123038621, "num_examples": 1000000, "dataset_name": "text"}}, "download_checksums": {"/home/perk/models/SimCSE-test/data/wiki1m_for_simcse.txt": {"num_bytes": 120038621, "checksum": "7b1825863a99aa76479b0456f7c210539dfaeeb69598b41fb4de4f524dd5a706"}}, "download_size": 120038621, "dataset_size": 123038621, "size_in_bytes": 243077242}

data/text/default-952efb7fdd4e970d/0.0.0/21a506d1b2b34316b1e82d0bd79066905d846e5d7e619823c0dd338d6f1fa6ad/text-train.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4cd7d749ccccf59a58dc1f2c4349440ee844c40554214559b7a1f91638f6051
+size 123059952

data/text/default-a93e22393ff62787/0.0.0/21a506d1b2b34316b1e82d0bd79066905d846e5d7e619823c0dd338d6f1fa6ad/cache-0b2eee437c5abe6d.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17254ba04c4c3369f474d8f1cff7af78edd23370a34463384dd511100f3ede4b
+size 320928696

data/text/default-a93e22393ff62787/0.0.0/21a506d1b2b34316b1e82d0bd79066905d846e5d7e619823c0dd338d6f1fa6ad/dataset_info.json

@@ -0,0 +1 @@
+{"description": "", "citation": "", "homepage": "", "license": "", "features": {"text": {"dtype": "string", "_type": "Value"}}, "builder_name": "text", "config_name": "default", "version": {"version_str": "0.0.0", "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 94417493, "num_examples": 1000000, "dataset_name": "text"}}, "download_checksums": {"/home/perk/models/SimCSE-test/data/nor_news_1998_2019_sentences_1M.txt": {"num_bytes": 91417493, "checksum": "a11b0487ea13419b7902c6890723469a3c540ec0143053e82312642509adfaae"}}, "download_size": 91417493, "dataset_size": 94417493, "size_in_bytes": 185834986}

data/text/default-a93e22393ff62787/0.0.0/21a506d1b2b34316b1e82d0bd79066905d846e5d7e619823c0dd338d6f1fa6ad/text-train.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c4d0f3761a50480d487ee5fd4c43ffc579d0158e0ada0823207dbad44e97d75
+size 94438104

run_unsup_GPU.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+
+# In this example, we show how to train SimCSE on unsupervised Wikipedia data.
+# If you want to train it with multiple GPU cards, see "run_sup_example.sh"
+# about how to use PyTorch's distributed data parallel.
+export ZONE=us-central2-b
+export XRT_TPU_CONFIG="localservice;0;localhost:51011"
+export TPU_NUM_DEVICES=4
+export ALLOW_MULTIPLE_LIBTPU_LOAD=1
+
+python3 ../../SimCSE/train.py \
+    --model_name_or_path NbAiLab/nb-bert-base \
+    --train_file data/wiki1m_for_simcse.txt \
+    --output_dir result/unsup-simcse-nb-bert-bert-base-GPU \
+    --num_train_epochs 1 \
+    --per_device_train_batch_size 64 \
+    --learning_rate 3e-5 \
+    --max_seq_length 32 \
+    --evaluation_strategy steps \
+    --metric_for_best_model stsb_spearman \
+    --load_best_model_at_end \
+    --eval_steps 125 \
+    --pooler_type cls \
+    --mlp_only_train \
+    --overwrite_output_dir \
+    --temp 0.05 \
+    --do_train \
+    --do_eval \
+    --fp16 \
+    "$@"
+