Merge branch 'main' of https://huggingface.co/NbAiLab/nb-roberta-base-ext into main

0.3.0

@@ -4,6 +4,15 @@ Requirement already satisfied: jax[tpu] in /home/perk/.local/lib/python3.8/site-
 Requirement already satisfied: scipy>=1.5 in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (1.10.0)
 Requirement already satisfied: opt-einsum in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (3.3.0)
 Requirement already satisfied: numpy>=1.20 in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (1.24.1)
+<<<<<<< HEAD
+Requirement already satisfied: requests in /usr/local/lib/python3.8/dist-packages (from jax[tpu]) (2.27.1)
+Requirement already satisfied: jaxlib==0.4.1 in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (0.4.1)
+Requirement already satisfied: libtpu-nightly==0.1.dev20221212 in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (0.1.dev20221212)
+Requirement already satisfied: certifi>=2017.4.17 in /usr/lib/python3/dist-packages (from requests->jax[tpu]) (2019.11.28)
+Requirement already satisfied: idna<4,>=2.5 in /usr/lib/python3/dist-packages (from requests->jax[tpu]) (2.8)
+Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.8/dist-packages (from requests->jax[tpu]) (2.0.12)
+Requirement already satisfied: urllib3<1.27,>=1.21.1 in /home/perk/.local/lib/python3.8/site-packages (from requests->jax[tpu]) (1.26.13)
+=======
 Requirement already satisfied: libtpu-nightly==0.1.dev20221212 in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (0.1.dev20221212)
 Requirement already satisfied: jaxlib==0.4.1 in /home/perk/.local/lib/python3.8/site-packages (from jax[tpu]) (0.4.1)
 Requirement already satisfied: requests in /usr/local/lib/python3.8/dist-packages (from jax[tpu]) (2.27.1)
@@ -11,3 +20,4 @@ Requirement already satisfied: certifi>=2017.4.17 in /usr/lib/python3/dist-packa
 Requirement already satisfied: urllib3<1.27,>=1.21.1 in /home/perk/.local/lib/python3.8/site-packages (from requests->jax[tpu]) (1.26.13)
 Requirement already satisfied: idna<4,>=2.5 in /usr/lib/python3/dist-packages (from requests->jax[tpu]) (2.8)
 Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.8/dist-packages (from requests->jax[tpu]) (2.0.12)
+>>>>>>> 055fdb68d12179bf9f23af94c6d9a4797a4b36da

config.json

@@ -21,7 +21,7 @@
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.24.0",
+  "transformers_version": "4.25.1",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 250002

events.out.tfevents.1672852728.t1v-n-29919176-w-3.22442.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6516ce9c7746084c51309c2aef27c555a169252a418c8ba2e4a556d2d39d821
+size 40

events.out.tfevents.1672853179.t1v-n-29919176-w-3.140245.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d55f87faf0667d7d84e51d8df81f02ba6640946e52f76323fb3383d0c0e495cf
+size 40

events.out.tfevents.1672853660.t1v-n-29919176-w-3.257607.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df7888d10faf8f207ed12292b63244b752f65670bf6880c4d29a0717ede95467
+size 2651789

events.out.tfevents.1672907612.t1v-n-29919176-w-3.477755.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04fa0820a8fc41435cd7772ace5ca6361aceebfe834d4daac2363ce4da933ed0
+size 40

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6836f8d8a9e97d1f37003a14f0b42eec29f196ec6960a015bfc3f7fc2de0dfa5
+size 1113187999

run_mlm_flax_stream.py

@@ -564,7 +564,7 @@ if __name__ == "__main__":
     train_metrics = []
     eval_metrics = []
     
-    training_iter = iter(torch.utils.data.DataLoader(tokenized_datasets.with_format("torch"), batch_size=1, shuffle=False, num_workers=dataset.n_shards, collate_fn=lambda x: x))
+    training_iter = iter(torch.utils.data.DataLoader(tokenized_datasets.with_format("torch"), batch_size=1, shuffle=False, num_workers=max(33,dataset.n_shards), collate_fn=lambda x: x))
     
     max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)
     eval_samples = advance_iter_and_group_samples(training_iter, data_args.num_eval_samples, max_seq_length)