upload flax model

.gitattributes

@@ -6,3 +6,4 @@
 *.tar.gz filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,62 @@
+---
+language:
+  - multilingual
+
+datasets:
+  - squad
+  - arcd
+  - xquad
+---
+
+# Multilingual BERT fine-tuned on SQuADv1.1
+
+[**WandB run link**](https://wandb.ai/salti/mBERT_QA/runs/wkqzhrp2)
+
+**GPU**: Tesla P100-PCIE-16GB
+
+## Training Arguments
+
+```python
+max_seq_length              = 512
+doc_stride                  = 256
+max_answer_length           = 64
+bacth_size                  = 16
+gradient_accumulation_steps = 2
+learning_rate               = 5e-5
+weight_decay                = 3e-7
+num_train_epochs            = 3
+warmup_ratio                = 0.1
+fp16                        = True
+fp16_opt_level              = "O1"
+seed                        = 0
+```
+
+## Results
+
+|   EM   |   F1   |
+| :----: | :----: |
+| 81.731 | 89.009 |
+
+## Zero-shot performance
+
+### on ARCD
+
+|   EM   |   F1   |
+| :----: | :----: |
+| 20.655 | 48.051 |
+
+### on XQuAD
+
+|  Language  |   EM   |   F1   |
+| :--------: | :----: | :----: |
+|   Arabic   | 42.185 | 57.803 |
+|  English   | 73.529 | 85.01  |
+|   German   | 55.882 | 72.555 |
+|   Greek    | 45.21  | 62.207 |
+|  Spanish   | 58.067 | 76.406 |
+|   Hindi    | 40.588 | 55.29  |
+|  Russian   | 55.126 | 71.617 |
+|    Thai    | 26.891 | 39.965 |
+|  Turkish   | 34.874 | 51.138 |
+| Vietnamese | 47.983 | 68.125 |
+|  Chinese   | 47.395 | 58.928 |

config.json

@@ -1,9 +1,11 @@
 {
+  "_name_or_path": "bert-base-multilingual-cased",
   "architectures": [
     "BertForQuestionAnswering"
   ],
   "attention_probs_dropout_prob": 0.1,
   "directionality": "bidi",
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -20,6 +22,9 @@
   "pooler_num_fc_layers": 3,
   "pooler_size_per_head": 128,
   "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.4.0.dev0",
   "type_vocab_size": 2,
+  "use_cache": true,
   "vocab_size": 119547
 }

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": false, "model_max_length": 512, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "name_or_path": "bert-base-multilingual-cased"}