Amol Kaushik commited on
Commit
c7e24f5
·
1 Parent(s): cb9b728
Files changed (1) hide show
  1. A9/A9_Report.ipynb +2 -4
A9/A9_Report.ipynb CHANGED
@@ -77,8 +77,6 @@
77
  "source": [
78
  "## 2. Deep Learning Steps Overview\n",
79
  "\n",
80
- "The 7-step Deep Learning workflow:\n",
81
- "\n",
82
  "| Step | Description | Implementation |\n",
83
  "|------|-------------|----------------|\n",
84
  "| 1 | Load Data | `load_all_sequences()` in models.py |\n",
@@ -97,12 +95,12 @@
97
  "source": [
98
  "## 3. Model Architectures\n",
99
  "\n",
100
- "### Dense (MLP)\n",
101
  "- Fully-connected layers\n",
102
  "- Treats each frame independently\n",
103
  "- Configuration: (256, 128, 64) hidden units, ReLU, dropout 0.3\n",
104
  "\n",
105
- "### Conv1D (CNN)\n",
106
  "- 1D convolutional layers for temporal patterns\n",
107
  "- Window size: 30 frames\n",
108
  "- Best variant (v3): filters=(128, 256), kernel=3, pool=3, dense=(256, 128, 64)\n",
 
77
  "source": [
78
  "## 2. Deep Learning Steps Overview\n",
79
  "\n",
 
 
80
  "| Step | Description | Implementation |\n",
81
  "|------|-------------|----------------|\n",
82
  "| 1 | Load Data | `load_all_sequences()` in models.py |\n",
 
95
  "source": [
96
  "## 3. Model Architectures\n",
97
  "\n",
98
+ "### Dense\n",
99
  "- Fully-connected layers\n",
100
  "- Treats each frame independently\n",
101
  "- Configuration: (256, 128, 64) hidden units, ReLU, dropout 0.3\n",
102
  "\n",
103
+ "### Conv1D\n",
104
  "- 1D convolutional layers for temporal patterns\n",
105
  "- Window size: 30 frames\n",
106
  "- Best variant (v3): filters=(128, 256), kernel=3, pool=3, dense=(256, 128, 64)\n",