Added all files from training

README.md

@@ -0,0 +1,92 @@
+---
+tags:
+- generated_from_trainer
+datasets:
+- squad
+model-index:
+- name: Graphcore/bert-base-uncased-squad
+  results: []
+---
+
+# Graphcore/bert-base-uncased-squad11
+
+This model is a fine-tuned version of [Graphcore/bert-base-uncased](https://huggingface.co/Graphcore/bert-base-uncased) on the squad dataset.
+
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+[squad dataset](https://huggingface.co/datasets/squad)
+
+## Training procedure
+
+Model was trained on 16 Graphcore Mk2 IPUs using the [optimum-graphcore](https://github.com/huggingface/optimum-graphcore) library.
+
+Command line:
+
+```
+python examples/question-answering/run_qa.py \
+  --model_name_or_path Graphcore/bert-base-uncased \
+  --ipu_config_name Graphcore/bert-base-ipu \
+  --dataset_name squad \
+  --do_train \
+  --do_eval \
+  --num_train_epochs 3 \
+  --per_device_train_batch_size 2 \
+  --per_device_eval_batch_size 2 \
+  --gradient_accumulation_steps 16 \
+  --pod_type pod16 \
+  --learning_rate 9e-5 \
+  --max_seq_length 384 \
+  --doc_stride 128 \
+  --seed 42\
+  --lr_scheduler_type linear \
+  --loss_scaling 64 \
+  --weight_decay 0.01 \
+  --warmup_ratio 0.2 \
+  --logging_steps 1 \
+  --save_steps 50 \
+  --dataloader_num_workers 64 \
+  --ipu_config_overrides "embedding_serialization_factor=2" \
+  --output_dir squad_v2_bert_base \
+  --overwrite_output_dir
+```
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 6e-05
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: IPU
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 256
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3.0
+- training precision: Mixed Precision
+
+### Training results
+
+```
+{ 
+    "epoch": 3.0,
+    "eval_exact_match": 81.79754020813624,
+    "eval_f1": 88.84840994541061,
+    "eval_samples": 10784
+}
+```
+
+
+### Framework versions
+
+- Transformers 4.18.0.dev0
+- Pytorch 1.10.0+cpu
+- Datasets 1.18.4
+- Tokenizers 0.11.6

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 81.79754020813624,
+    "eval_f1": 88.84840994541061,
+    "eval_samples": 10784,
+    "train_loss": 0.9942761997093901,
+    "train_runtime": 437.6871,
+    "train_samples": 88524,
+    "train_samples_per_second": 606.762,
+    "train_steps_per_second": 2.365
+}

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "Graphcore/bert-base-uncased",
+  "architectures": [
+    "PoptorchPipelinedBertForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 0.001,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float16",
+  "transformers_version": "4.18.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 81.79754020813624,
+    "eval_f1": 88.84840994541061,
+    "eval_samples": 10784
+}

ipu_config.json

@@ -0,0 +1,27 @@
+{
+  "device_iterations": 2,
+  "embedding_serialization_factor": 2,
+  "enable_half_first_order_momentum": true,
+  "enable_half_partials": true,
+  "executable_cache_dir": "./exe_cache",
+  "gradient_accumulation_steps": 16,
+  "inference_device_iterations": 4,
+  "inference_replication_factor": 4,
+  "ipus_per_replica": 4,
+  "layers_per_ipu": [
+    0,
+    4,
+    4,
+    4
+  ],
+  "matmul_proportion": 0.25,
+  "optimizer_state_offchip": false,
+  "optimum_version": "0.1.2",
+  "output_mode": "final",
+  "profile_dir": "",
+  "recompute_checkpoint_every_layer": true,
+  "replicated_tensor_sharding": true,
+  "replication_factor": 4,
+  "seed": 42,
+  "use_popdist": false
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19d2e8b2ee1b19d2ab367416fa4237fbb2bc5805b0b0967a75d9b551a17ee480
+size 217854001

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "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, "special_tokens_map_file": null, "name_or_path": "Graphcore/bert-base-uncased", "tokenizer_class": "BertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.9942761997093901,
+    "train_runtime": 437.6871,
+    "train_samples": 88524,
+    "train_samples_per_second": 606.762,
+    "train_steps_per_second": 2.365
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81847e188efe71a1feccd9f111a09de422f8961c351bac3eca3b470d6c67cdb7
+size 2735