Initial commit

ArabicTransformer8-8-8-TF.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0b5cfe51a364112875263b725ca0388d2b192673742156b9c4921daef87e798
+size 4701659599

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": [
+    8,
+    8,
+    8
+  ],
+  "d_head": 64,
+  "d_inner": 4096,
+  "d_model": 1024,
+  "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": 16,
+  "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 sha256:94419ae8433570a87acced5047cf5f0235c0f139ea6f6e9ca56d398311245926
+size 1515260202

net_config.pytorch.json

@@ -0,0 +1,16 @@
+{
+    "block_size": "8_8_8",
+    "d_embed": 1024,
+    "d_head": 64,
+    "d_inner": 4096,
+    "d_model": 1024,
+    "dropact": 0.0,
+    "dropatt": 0.1,
+    "dropout": 0.1,
+    "n_head": 16,
+    "pool_q_only": true,
+    "pooling_size": 2,
+    "pooling_type": "mean",
+    "separate_cls": true,
+    "vocab_size": 50000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a4399bc65b05a4d9aa007e7f23ea0af7928e0284e8aa8225a85117d0adf41d
+size 1624488789