Initial commit

ArabicTransformer4-4-4-TF.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45ae7733ce6ba98d0f936553090ada5d2bd6602557b7d0ddfb12c9b5b1154438
+size 1683288308

config.json

@@ -0,0 +1,37 @@
+{
+  "activation_dropout": 0.0,
+  "architectures": [
+    "FunnelModel"
+  ],
+  "attention_dropout": 0.1,
+  "attention_type": "relative_shift",
+  "block_repeats": [
+    1,
+    1,
+    1
+  ],
+  "block_sizes": [
+    4,
+    4,
+    4
+  ],
+  "d_head": 64,
+  "d_inner": 3072,
+  "d_model": 768,
+  "hidden_act": "gelu_new",
+  "hidden_dropout": 0.1,
+  "initializer_range": 0.1,
+  "initializer_std": null,
+  "layer_norm_eps": 1e-09,
+  "max_position_embeddings": 512,
+  "model_type": "funnel",
+  "n_head": 12,
+  "num_decoder_layers": 2,
+  "pool_q_only": true,
+  "pooling_type": "mean",
+  "rel_attn_type": "factorized",
+  "separate_cls": true,
+  "truncate_seq": true,
+  "type_vocab_size": 3,
+  "vocab_size": 50000
+}

model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd055e2e4e5a9cbf057d4c56910c1bc244513c1998e49d32b7cb559a37ff88f3
+size 522362234

net_config.pytorch.json

@@ -0,0 +1,16 @@
+{
+    "block_size": "4_4_4",
+    "d_embed": 768,
+    "d_head": 64,
+    "d_inner": 3072,
+    "d_model": 768,
+    "dropact": 0.0,
+    "dropatt": 0.1,
+    "dropout": 0.1,
+    "n_head": 12,
+    "pool_q_only": true,
+    "pooling_size": 2,
+    "pooling_type": "mean",
+    "separate_cls": true,
+    "vocab_size": 50000
+}

pytorch_model-base.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:166293638511f21ecd8d7fe5f1d32b7a2fc52c933e2890b817520ebd2b409cbc
+size 522373790

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f6651df0b2be2af8426494bebb773bd951d9abaf1bb01382f868edf881cd740
+size 583834942