Changed ML Model

final.py

@@ -12,7 +12,7 @@ dotenv.load_dotenv()
 nlp = spacy.load("en_core_web_sm")
 
 # Load the trained ML model
-model_path = "./results/checkpoint-5030"  # Replace with the actual path to your model
+model_path = "./results/checkpoint-753"  # Replace with the actual path to your model
 tokenizer = AutoTokenizer.from_pretrained('microsoft/deberta-v3-small')
 model = AutoModelForSequenceClassification.from_pretrained(model_path)
 model.eval()

results/checkpoint-5030/optimizer.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:cddba7c9ed0694f75f418657613b8400183c22b1e86f0d5fac90de0153d72e5f
-size 1135260474

results/checkpoint-5030/trainer_state.json

@@ -1,143 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 5.0,
-  "eval_steps": 500,
-  "global_step": 5030,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.4970178926441352,
-      "grad_norm": 11.328213691711426,
-      "learning_rate": 5e-05,
-      "loss": 0.3471,
-      "step": 500
-    },
-    {
-      "epoch": 0.9940357852882704,
-      "grad_norm": 0.29149460792541504,
-      "learning_rate": 4.448123620309051e-05,
-      "loss": 0.1462,
-      "step": 1000
-    },
-    {
-      "epoch": 1.0,
-      "eval_loss": 0.14880910515785217,
-      "eval_runtime": 32.5193,
-      "eval_samples_per_second": 61.871,
-      "eval_steps_per_second": 7.749,
-      "step": 1006
-    },
-    {
-      "epoch": 1.4910536779324055,
-      "grad_norm": 0.04432953894138336,
-      "learning_rate": 3.896247240618102e-05,
-      "loss": 0.0738,
-      "step": 1500
-    },
-    {
-      "epoch": 1.9880715705765408,
-      "grad_norm": 0.004722778219729662,
-      "learning_rate": 3.3443708609271526e-05,
-      "loss": 0.0599,
-      "step": 2000
-    },
-    {
-      "epoch": 2.0,
-      "eval_loss": 0.17704755067825317,
-      "eval_runtime": 32.4526,
-      "eval_samples_per_second": 61.998,
-      "eval_steps_per_second": 7.765,
-      "step": 2012
-    },
-    {
-      "epoch": 2.485089463220676,
-      "grad_norm": 0.0014285552315413952,
-      "learning_rate": 2.792494481236203e-05,
-      "loss": 0.0176,
-      "step": 2500
-    },
-    {
-      "epoch": 2.982107355864811,
-      "grad_norm": 0.0008603875176049769,
-      "learning_rate": 2.240618101545254e-05,
-      "loss": 0.026,
-      "step": 3000
-    },
-    {
-      "epoch": 3.0,
-      "eval_loss": 0.16322186589241028,
-      "eval_runtime": 32.2403,
-      "eval_samples_per_second": 62.406,
-      "eval_steps_per_second": 7.816,
-      "step": 3018
-    },
-    {
-      "epoch": 3.4791252485089466,
-      "grad_norm": 0.000587798363994807,
-      "learning_rate": 1.688741721854305e-05,
-      "loss": 0.0042,
-      "step": 3500
-    },
-    {
-      "epoch": 3.9761431411530817,
-      "grad_norm": 0.00033068188349716365,
-      "learning_rate": 1.1368653421633555e-05,
-      "loss": 0.0012,
-      "step": 4000
-    },
-    {
-      "epoch": 4.0,
-      "eval_loss": 0.20389850437641144,
-      "eval_runtime": 33.2829,
-      "eval_samples_per_second": 60.452,
-      "eval_steps_per_second": 7.571,
-      "step": 4024
-    },
-    {
-      "epoch": 4.473161033797217,
-      "grad_norm": 0.0048806252889335155,
-      "learning_rate": 5.8498896247240626e-06,
-      "loss": 0.0013,
-      "step": 4500
-    },
-    {
-      "epoch": 4.970178926441352,
-      "grad_norm": 0.00042022508569061756,
-      "learning_rate": 3.3112582781456954e-07,
-      "loss": 0.0006,
-      "step": 5000
-    },
-    {
-      "epoch": 5.0,
-      "eval_loss": 0.19458653032779694,
-      "eval_runtime": 33.1006,
-      "eval_samples_per_second": 60.784,
-      "eval_steps_per_second": 7.613,
-      "step": 5030
-    }
-  ],
-  "logging_steps": 500,
-  "max_steps": 5030,
-  "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
-  "save_steps": 500,
-  "stateful_callbacks": {
-    "TrainerControl": {
-      "args": {
-        "should_epoch_stop": false,
-        "should_evaluate": false,
-        "should_log": false,
-        "should_save": true,
-        "should_training_stop": true
-      },
-      "attributes": {}
-    }
-  },
-  "total_flos": 1332007138928640.0,
-  "train_batch_size": 8,
-  "trial_name": null,
-  "trial_params": null
-}

results/{checkpoint-5030 → checkpoint-753}/config.json

@@ -1,25 +1,25 @@
 {
-  "_name_or_path": "microsoft/deberta-v3-small",
+  "_name_or_path": "microsoft/deberta-v3-xsmall",
   "architectures": [
     "DebertaV2ForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 384,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 1536,
   "layer_norm_eps": 1e-07,
   "max_position_embeddings": 512,
   "max_relative_positions": -1,
   "model_type": "deberta-v2",
   "norm_rel_ebd": "layer_norm",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 6,
+  "num_attention_heads": 6,
+  "num_hidden_layers": 12,
   "pad_token_id": 0,
   "pooler_dropout": 0,
   "pooler_hidden_act": "gelu",
-  "pooler_hidden_size": 768,
+  "pooler_hidden_size": 384,
   "pos_att_type": [
     "p2c",
     "c2p"

results/{checkpoint-5030 → checkpoint-753}/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f34f9b72aa96cb0927c5cfcdad25c0281212e297d61dd14dcacdb68138c40840
-size 567598552
+oid sha256:5c8bc472032aa1625a83fa5a61358b394aa47e8936084fd5d5fc53d39b4819e7
+size 283347432

results/checkpoint-753/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d54a2486861a93c63c9d3f1ad129317a5ec061c153cc35f88750193eb19c8db
+size 566814714

results/{checkpoint-5030 → checkpoint-753}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d0c9d10259d2c7407ae8f630db471aed45598cb19d4fec8b8a17555906525a5
+oid sha256:bab711e45afdac9084a8d3228aa5d84f0234c10b8536782c428a3e5241e763c0
 size 14244

results/{checkpoint-5030 → checkpoint-753}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f0b07a36064ffcbc9c9cdc658bf6076e72b04ada218a099af03a6b74a3518d1
+oid sha256:2254eb2782bb8f96d8221a7f05be58b9aa6b59a9ac623c10f2d2cc29c6abdd07
 size 1064

results/checkpoint-753/trainer_state.json

@@ -0,0 +1,64 @@
+{
+  "best_metric": 0.13373112678527832,
+  "best_model_checkpoint": "./results\\checkpoint-503",
+  "epoch": 2.99403578528827,
+  "eval_steps": 500,
+  "global_step": 753,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.9980119284294234,
+      "eval_loss": 0.16927649080753326,
+      "eval_runtime": 34.3209,
+      "eval_samples_per_second": 58.623,
+      "eval_steps_per_second": 3.671,
+      "step": 251
+    },
+    {
+      "epoch": 1.9880715705765408,
+      "grad_norm": 3.436805248260498,
+      "learning_rate": 2.53479125248509e-05,
+      "loss": 0.2895,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.13373112678527832,
+      "eval_runtime": 32.7048,
+      "eval_samples_per_second": 61.52,
+      "eval_steps_per_second": 3.853,
+      "step": 503
+    },
+    {
+      "epoch": 2.99403578528827,
+      "eval_loss": 0.1674525886774063,
+      "eval_runtime": 33.2196,
+      "eval_samples_per_second": 60.567,
+      "eval_steps_per_second": 3.793,
+      "step": 753
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 753,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 198349894207488.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

results/{checkpoint-5030 → checkpoint-753}/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e34c99e352dd9e22706f7f1143f42ff1385e64d6b188ee3ed83ab034094c017
+oid sha256:5d020540fc118248e604cd22f9ec20b7acb4023a8953f7fb309148a6a3c3deb8
 size 5240