Upload AudioSpectogramTransformerForSequenceClassification

config.json

@@ -3,7 +3,8 @@
     "AudioSpectogramTransformerForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.0,
-  "fstride": 10,
+  "frequency_dimension": 128,
+  "frequency_stride": 10,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
@@ -536,10 +537,7 @@
     "525": "Radio",
     "526": "Field recording"
   },
-  "image_size": 224,
   "initializer_range": 0.02,
-  "input_fdim": 128,
-  "input_tdim": 1024,
   "intermediate_size": 3072,
   "label2id": {
     "A capella": 255,
@@ -1073,11 +1071,11 @@
   "layer_norm_eps": 1e-12,
   "model_type": "audio-spectogram-transformer",
   "num_attention_heads": 12,
-  "num_channels": 3,
   "num_hidden_layers": 12,
   "patch_size": 16,
   "qkv_bias": true,
+  "time_dimension": 1024,
+  "time_stride": 10,
   "torch_dtype": "float32",
-  "transformers_version": "4.23.0.dev0",
-  "tstride": 10
+  "transformers_version": "4.24.0.dev0"
 }