add model weights

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 40.0,
+    "eval_accuracy": 0.4267794423338829,
+    "eval_loss": 3.970010280609131,
+    "eval_runtime": 56.8854,
+    "eval_samples": 1045,
+    "eval_samples_per_second": 18.37,
+    "eval_steps_per_second": 0.299,
+    "test_samples": 1045,
+    "train_loss": 6.490346755981445,
+    "train_runtime": 7961.5563,
+    "train_samples": 3839,
+    "train_samples_per_second": 19.288,
+    "train_steps_per_second": 0.151
+}

config.json

@@ -0,0 +1,73 @@
+{
+  "_name_or_path": "exp_audiocaps_cfmer_2l/beats_ft_frozen_bart_8l_baseline_mincap6_ep10_convds3_embnce_t.5_mixup",
+  "activation_dropout": 0.1,
+  "activation_function": "gelu",
+  "architectures": [
+    "BartConformerBeatsSeq2SeqForCaptioning"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "contrastive_temperature": 0.5,
+  "d_model": 768,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 3072,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "embed_predictor_ffn_dim": 3072,
+  "embed_predictor_out_dim": 768,
+  "encoder_attention_heads": 16,
+  "encoder_downsample_rate": 3,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "encoder_repr_layer_idx": 10,
+  "eos_token_id": 2,
+  "forced_eos_token_id": 2,
+  "freeze_encoder": true,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "lsm_weight": 0.1,
+  "max_cross_position_embeddings": 512,
+  "max_position_embeddings": 128,
+  "min_caption_len": 6,
+  "model_type": "bart",
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "pretrained_beats_path": "/scratch/bbjs/slseanwu/dcase23_aac/beats_baseline/pretrained_weights/BEATs_iter3_plus_AS2M_finetuned_on_AS2M_cpt1.pt",
+  "scale_embedding": false,
+  "spec_aug": {
+    "freq_mask_width_range": [
+      0,
+      64
+    ],
+    "num_freq_mask": 2,
+    "num_time_mask": 5,
+    "time_mask_width_ratio_range": [
+      0,
+      0.12
+    ]
+  },
+  "tokenizer_dir": "facebook/bart-base",
+  "torch_dtype": "float32",
+  "transformers_version": "4.22.2",
+  "use_cache": true,
+  "use_chatgpt_mixup": true,
+  "use_contrastive_embed_loss": true,
+  "use_conv_downsample": true,
+  "use_encoder_embed_mlp": true,
+  "use_weighted_encoder_repr": false,
+  "vocab_size": 50265
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 40.0,
+    "eval_accuracy": 0.42745985711254153,
+    "eval_loss": 3.971204996109009,
+    "eval_runtime": 57.8802,
+    "eval_samples": 1045,
+    "eval_samples_per_second": 18.055,
+    "eval_steps_per_second": 0.294
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30777c73f63e6e556f0501ba7dcc838dd391311ced276f3fd95a5d8ffa4b6f20
+size 869569707

test_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 40.0,
+    "eval_accuracy": 0.4267794423338829,
+    "eval_loss": 3.970010280609131,
+    "eval_runtime": 56.8854,
+    "eval_samples_per_second": 18.37,
+    "eval_steps_per_second": 0.299,
+    "test_samples": 1045
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 40.0,
+    "train_loss": 6.490346755981445,
+    "train_runtime": 7961.5563,
+    "train_samples": 3839,
+    "train_samples_per_second": 19.288,
+    "train_steps_per_second": 0.151
+}